Characterization of sintered discs containing distinct sawdust content in the bottom-layer obtained from mercury intrusion porosimetry.Sawdust content (wt.%)Open porosity (%)Bulk density (g/cm3)Median pore diameter (μm)2.54.82.301.255.42.2415.81011.12.1640.8Full-size tableTable optionsView in workspaceDownload as CSVFig. 6. Pore size distribution curves (calculated from mercury intrusion data) for sintered discs produced with distinct sawdust content.Figure optionsDownload full-size imageDownload as PowerPoint slideTo comply with the standard, porcelain stoneware tiles must have bending strength values higher than 35 MPa (ISO 10545/4). In this work, the mechanical resistance of the sintered bi-layered ceramic discs is controlled by the degree of porosity in the bottom-layer of the discs. Fig. 7 presents the bending strength and Young\’s modulus of the fired bodies. Results show that both parameters diminish with increasing sawdust content. Nevertheless, the incorporation of up to 10 wt% of sawdust creates materials that still comply with these specifications. As for other standard properties required for porcelain stoneware tiles they Ro3306 are ensured by the dense top-layer.Fig. 7. Mechanical properties of bi-layered ceramic tiles as a function of sawdust content: (a) bending strength and (b) Young\’s modulus.Figure optionsDownload full-size imageDownload as PowerPoint slideThe specific strength of samples was also evaluated, as proposed by Ashby (2005). Experimental values ranged between 3.2 and 4.0 MPa0.5 cm3/g, being similar to those reported in the literature for lightweight porcelain stoneware tiles (Bernardo et?al., 2010 and Novais et?al., 2014).The creation of porosity in the bottom layer sintered discs is expected to decrease the thermal conductivity of the bodies. In fact, the thermal conductivity strongly decreases with porosity, as shown in Fig. 8. For comparison purposes, the thermal conductivity of a standard ceramic sample (prepared without porogen addition) was included in the figure. A threefold decrease in the thermal conductivity (from 0.71 to 0.23 W/m K) was observed when only 5 wt% sawdust was added to the bottom layer of the bi-layered discs. This observation is consistent with the above-mentioned porogen percolation threshold. Indeed, SEM micrographs in Fig. 5b and f clearly show the formation of networks between adjacent pores, hence reducing the solid paths throughout the ceramic body. The thermal conductivity attenuation with porosity level observed in Fig. 8 was steeper than that observed when polypropylene and polymethyl methacrylate were used as porogen agents (Novais et al., 2014). The thermal insulation achieved with sawdust incorporation endows porcelain stoneware ceramic tiles with new features that may extend the range of applications of this common product.Fig. 8. Thermal conductivity of the porous layer sintered discs, prepared with sawdust, as function of open porosity level. The horizontal line corresponds to the thermal conductivity of a standard composition (without porogen).Figure optionsDownload full-size imageDownload as PowerPoint slide4. ConclusionsThis study evaluated the possibility of using wood wastes (sawdust) as a pore forming agent for producing porcelain stoneware ceramic tiles with novel features.Lightweight bi-layered bodies showing suitable mechanical resistance and low thermal conductivity were fabricated, attesting to the potential of using sawdust as a pore forming agent in such fast-fired ceramic products.Optical microscopy and mercury intrusion porosimetry characterization demonstrated that the porosity level is controlled by sawdust content, and therefore can be tuned considering the application envisaged.Sawdust presents fast and complete combustion, without leaving residues or ashes, and does not induce defects in the ceramics bodies. Additionally, the heat released from its decomposition brings value to the ceramic tile manufacturing process, allowing energy savings.The incorporation of sawdust in the bottom layer of the bi-layered ceramics promotes weight reduction (up to 7.5%) and simultaneous thermal conductivity attenuation (up to 76%). The low porogen percolation threshold (5 wt%) achieved endorsed a threefold decrease in the ceramic tile\’s thermal conductivity in comparison to commercial stoneware tiles. At the same time, the product complies with mechanical strength requirements when sawdust incorporation level is below 10 wt%.Results demonstrate that innovative products with excellent features can be produced by incorporation of sawdust into porcelain stoneware ceramic tiles. The novel ceramic tiles ensure environmental, technical and economic advantages: waste valorisation by sawdust reuse (environmental advantage); density reduction of the product which decreases the tiles transportation and distribution costs (economic advantage); restrain energy loss (technical advantage). These new and exciting features may widen the range of applications of porcelain stoneware tiles while simultaneously contributing towards sustainable construction.AcknowledgementsThe authors acknowledge the financial support from Portuguese Innovation Agency (Adi) through project ThermoCer, to CICECO (PEst C/CTM/LA0011/2013) and RNME – Pole University of Aveiro (FCT Project REDE/1509/RME/2005) for instrument use, scientific and technical assistance. The authors acknowledge CINCA for providing the spray-dried powder, and the assistance of Dr. R.C. Pullar with editing English language in this paper.

## GW841819X Markov matrix of regional energy efficiency between and

Venture capital; Entity industry; Green innovation1. IntroductionAs the important foundation of national economy, entity industry refers to real industry satisfying material and cultural needs of human, including agriculture, manufacturing and most service industries. Entity industry can be divided into green industry and non-green industry from the perspective whether it is conducive to resource conservation and environmental protection. Green industry is conducive to resource conservation and environmental protection. Narrow-sense green industry refers to service industry of GW841819X conservation and environmental management services, while general green industry is the industry consuming less resource and producing less environmental pollution. The so-called non-green industry refers to industries with large consumption of resources and heavy environmental pollution.Green innovation refers to technological innovation that ecological concept is introduced into various stages of technological innovation for entity industry, thus benefiting resource conservation and environmental protection (Zhang, 2013). Practice in developed countries has proved its important supporting role in energy conservation. For example, the use of aeration technology played a huge role in the pollution control project of UK Thames in early 1960s. In 1970s, Japan introduced the world’s most stringent standards of sulfur dioxide emission, greatly reducing sulfur dioxide emissions through desulfurization technology (Bu, 2006).The support of financial industry is indispensable to promote green innovation activities. It is reasonable and necessary for government to provide financial supports due to significant positive externalities of green innovation. However, financial resources form government is very limited compared to the fund demand of green innovation. After all, aspects of response to climate change, pollution control, eco-economy development, and sustainable development are common aspiration of mankind throughout the world. It is the inevitable trend of economic and social development to transform economic development mode and lifestyle with construction of ecological civilization. Thus, expansion of green innovation funding sources has become an inevitable choice for entity industries. According to the prediction of US Energy Foundation and China National Development and Reform Commission, annual financing gap of Chinese energy saving industry, new energy industry and environmental management industry is about 200 billion RMB; it will reach at least two trillion RMB by 2020 (subject group, 2009). Therefore, industries of energy conservation, new energy development and environmental management cannot be promoted for green innovation without active use of financial instruments, thus making it difficult to promote green innovation.Academic research has proved the supporting role of venture capital in technology innovation. Kortum and Lerner (2000) found that venture capital greatly promoted technology innovation in economy in the United States – the promoting effect of 1

## Venture capital Entity industry Green innovation IntroductionAs the

Spatial Markov matrix of regional GW841819X efficiency between 1999 and 2010 in China.Spatial lagti/ti+1n1: <75%2: <100%3: <125%4: >125%11200.950.050.000.002120.050.860.090.00340.000.001.000.00400.000.000.000.0021650.900.080.000.022560.050.850.090.013290.000.150.780.074180.000.000.180.8231121.000.000.000.002150.000.690.310.003390.000.190.700.114180.000.050.140.814100.000.000.000.00200.000.000.000.003140.000.000.800.204580.000.000.030.97Full-size tableTable optionsView in workspaceDownload as CSVTable 5 illustrates three factors.First, the spatial relationship between regions plays an important role in the convergence club of energy efficiency in China. With different neighbors, the transition probabilities of regional energy are different. In other words, if the background of a region does not change, the four conditional matrices in the same period in Table 5 should be similar to each other. In fact, the background of a region does not change.Second, different regional backgrounds play different roles in the transfer of energy efficiency type. The probability of an upward shift will increase and the probability of a downward shift will decrease if a region is within the regional neighborhood with a high level of energy efficiency. Conversely, the probability of an upward shift will decrease and the probability of a downward shift will increase if a region is within the regional neighborhood with a low level of energy efficiency. Between 1999 and 2010, when a region with low energy efficiency is adjacent to regions with low energy efficiency, the probability of an upward shift is 5%; meanwhile, if a region is adjacent to regions with medium-low, medium-high, or high-level energy efficiency, the upward shift probability is increased to 8%. The probability of an upward shift is 19%, and the probability of a downward shift is 10% if a region with medium-low energy efficiency is adjacent to regions with low or medium-low energy efficiency; the probability of an upward shift is 31% and the probability of a downward shift is 0% if a region is adjacent to regions with medium-high or high energy efficiency. When a region with medium-high energy efficiency is adjacent to regions with low, medium-low, or medium-high energy efficiency, the probability of an upward shift is 11% and the probability of a downward shift is 34%; meanwhile, if a region is adjacent to regions with high or medium-high energy efficiency, the probability of an upward shift is 20% and the probability of a downward shift is 0%. When a region with high energy efficiency is adjacent to regions with lower energy efficiency, the probability of a downward shift is 32%; meanwhile, if a region is adjacent to regions with higher energy efficiency, the probability of a downward shift is 3%.Third, the matrix of the spatial Markov transition probability provides a spatial interpretation for the “club convergence” phenomenon. A region will be negatively influenced by its geographical neighbors with a low level of energy efficiency. Between 1999 and 2010, if the geographical neighbors of a region have a low level of energy efficiency, the probability of this region to maintain a low level of energy efficiency after several years is 95%. This probability is higher than the probability that ignores the regional neighbors in Table 4, which is 0.92 in the same period. Between 1999 and 2010, the probability of a region to maintain a high level of energy efficiency is 97% if its geographical neighbors are at a high level as well; this probability is higher than the probability in Table 4 in the same period, which is 0.90.4. ConclusionWe adopt DEA in this paper to calculate regional energy efficiency from the perspective of total-factor energy efficiency, and the club convergence of the regional energy efficiency in China is subsequently tested using the Markov chain and spatial Markov chain methods. We draw the following conclusions:(1)The “club convergence” phenomenon exists in the regional energy efficiency in China between 1999 and 2010, and the levels of club convergence are low, medium-low, medium-high, and high. Moreover, the stability of both low- and high-level club convergence is high.(2)The energy efficiency class transitions in China are highly constrained by their regional backgrounds. The regional transitions are positively influenced by regions with a high level of energy efficiency and are negatively influenced by regions with a low level of energy efficiency. These empirical analyses provide a spatial explanation to the existence of the “club convergence” phenomenon of regional energy efficiency in China.(3)In accordance with the dynamic evolution of regional energy efficiency in China, special attention should be paid to spatial effect, and regional cooperation should be strengthened. Policy that favors the “enrich the neighbor” approach should be used in regions with a high level of energy efficiency. Simultaneously considering the geography, population, industry, resources, etc., attains a win–win situation on energy efficiency. Preferential policies should be implemented in the low-level and low-growth regions of energy efficiency to enhance the opening-up level, thus accelerating the adjustment and optimization of the industrial structure, and the promotion of energy efficiency of these areas.AcknowledgementsThis paper is the stage achievement of the National Natural Science Foundation of China (71303029) and the National Social Science Fund Project (10BGL066). The author is grateful for the support of the National Natural Science Foundation of China and the National Social Science Foundation of China.

## leukotriene receptor antagonists During daylight hours the red orange

During daylight hours the red, orange and yellow-coded foraging environments may represent the vertical stratification of the different prey communities targeted by SES ranging from deep mesopelagic, intermediate depth mesopelagic to epipelagic resources. The decision of the SES to target different layers is likely to depend on their local occurrence and profitability. While yellow and red-coded environments were almost exclusively identified during daylight hours (see Fig. 7), the orange-coded environment is detected during both day and night but at different depths (see Fig. 5 and Fig. 7). At night, we are currently unable to determine if the orange-coded environment represents or not a mixture between on one hand the day red-coded environment representative of a deep mesopelagic prey assemblage migrating closer to the surface at night and on the other hand the intermediate depth mesopelagic one which extend its distribution to epipelagic waters and possibly mix with the yellow-coded epipelagic resources observed during the day.

Oceanographic parameters other than the ones considered in our analyses are also likely to affect prey distribution and foraging habitat, such as leukotriene receptor antagonists density, which is linked to primary production (Moore and Abbott, 2000), salinity, a key physical parameter of oceans (Caldwell, 1974), or dissolved oxygen, a major biogeochemical component for marine ectotherms (Karna, 2003). Including them in our habitat models could refine our analysis improve the distinction between different habitat environments.

Investigating species distribution in association with their surrounding environmental conditions is essential for improving our understanding of the marine ecosystem, particularly within the context of climate change (Chen-Tung, 2008, Ishizaka, 2010 and Doney et al., 2012). These changes are expected to have substantial biological consequences on marine ecosystems by impacting both the horizontal and vertical distribution of food resources (Cantin et al., 2011) and consequently the foraging efficiency of their natural predators. Currently, these biological consequences remain poorly understood. The results in this study provide a better understanding of the delimitation and the characterization of foraging environments of a top predator, but also emphasize how eclectic SESs are in their foraging environments. This suggests that this species, according to its horizontal and vertical range, is likely to adapt to future climatic perturbations. Indeed, this study provides insight into the horizontal and vertical variability of mesopelagic resource distribution targeted by a deep diving predator. However, we would like to stress that it would be highly beneficial to combine these results to data obtained from acoustic survey and trawl net sampling (MyctO-3D -MAP project; Fielding et al., 2012) to improve our assessment of prey distribution, and compare them with other CLIOTOP approaches such as the Seapodym model (Lehodey et al., 2010).

## The Frame for Food education in

The Frame for Food education in early childhood education developed in Finland. SBFE = Sensory-based food education, ECEC = Local curricula for childhood education and care.

Figure options

In 2013, 50 municipalities and cities from Finland and approximately 270 daycare centres started to develop food education for children based on the knowledge of sensory education and Sapere-based application (Koistinen & Ruhanen, 2009). Since 2009, approximately 7000 teachers in kindergartens, nursery nurses and catering staff members were trained in food education activities in daily pedagogy and were educated in the basics of nutrition for children, child-oriented participatory food education and catering. In some municipalities and cities, food education is already included in their Curriculum Guidelines on ECEC.

Daycare and children’s catering staff in selected municipalities were challenged to jointly develop child-oriented food activities. In practice, this Tubastatin A manufacturer meant development work in their organizations at the management and unit levels. Moreover, in those units, the entire staff was educated on the basics of nutrition for children. In addition, the staff was trained in sensory tasks that were developed for children under the school-age. A pedagogic menu was developed at the Central Finland Health Care District as a new tool for activating and helping the education staff to make sensory tasks for children during and outside mealtimes and as a tool for the kitchen staff to offer ingredients cost-effectively for food education activities. The menu contains meals and a selection of food components anther can be used as is or as components for sensory or cooking and baking tasks with children. Ready and easy food themes were also built in the menu. Specifically, more vegetables, fruits, berries, fish, and whole grain cereals were included in the menu.

The Finnish sensory-based application for food education includes sensory training, cooking and other activities with real food. In addition, a pedagogic menu supports the pedagogic work with children. However, a collaboration between all of the adults is essential. Pedagogic menu foods used in sensory task on a daily level are delivered to the groups of children according to the menu. The Sapere-based food education is considered to be a good practice in food education in early childhood education according the questionnaire targeted at the trained kindergarten staff and appears to be a promising approach for improving nutrition of children in early education. In addition, as a result of common training and new tools, the staff determined that there was an increase in cooperation among the daycare and catering staff, which inspired them to jointly develop food education for children.

## There are a few studies that have investigated the

There are a few studies that have investigated the difficulties of integration. Tress et al., 2005 and Tress et al., 2007 have identified the following barriers: a) difficulties in communication and cooperation, 2) lack of necessary resources, 3) lack of interest in cooperation, 4) lack of knowledge of other disciplines, 5) time demands, 6) lack of common terminology, and 7) lack of knowledge about integrative methods. Other authors confirm these barriers in one way or another (Ayre and Nettle, 2015, B?hm, 2006, Klein, 2012 and Newell, 2012).

Klein (2012) stresses that purchase c-Myc Peptide should be conducted from the very beginning and possibly at every sub-project level. Meanwhile, Stokols et al. (2008), Stokols, Harvey, Gress, Fuqua, and Phillips (2005) endorse the relevance of communication for integration and mutual understanding, as by means of discussions and exchange, the “building blocks” of integrative collaboration arise. Newell (2012) argues that for effective “integrative practice the development of a shared understanding between individuals with a wide range of conceptual systems” is essential. Thus, time and space for communication are crucial: No integration without discussion (Tress et al., 2005).

It is the aim of this article to reflect upon the integration process by taking the above-mentioned barriers and challenges into account. In order to do so, we draw on experiences gained through the ELaN1 project, which dealt with integrated water and land management issues. Within this project, scenarios were used as boundary objects for interdisciplinary exchange.

After a brief introduction of the ELaN project (Section 2) and the scenario method (Section 3), the challenges of integration based on scenarios are presented (Section 4). Section 5 discusses the suitability of the method for integration and improving the quality of the interdisciplinary process. We then conclude by relating the experiences or the project to the literature in this field (Section 6).

## Incorporating orbit evolution into the design process In the

5. Incorporating orbit evolution into the design process

In the previous sections, a review of the evolution of the libration point orbits is offered. Methods of consolidating this information into charts that can be employed within the mission design process would prove useful in selecting orbits that best meet mission requirements. For example, knowledge about the evolution of periodic orbit stability with Jacobi constant value is useful to determine the existence of periodic orbits as well as their associated stable/unstable and center manifold structures. The stability information from Fig. 3, Fig. 4, Fig. 5 and Fig. 8 is collected to produce the chart in Fig. 16.

Evolution of stability for L1 and L2 orbit families; (nC?2)/2=0 (red), 1 (green), 2 (blue) (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

Here the orbit family names appear in the boxed labels, and the bifurcations from Table 1 are included as well. Members from the various periodic orbit families are represented by their associated value of Jacobi constant, and are colored according to the dimension of the unstable subspace nU=nS such that nU=2→ red, nU=1→ green, and nU=0→ blue. In the chart, the abscissa is inverted so that moving along the horizontal axis corresponds to increasing 3X FLAG tag manufacturer (decreasing Jacobi constant). Note that the vertical families extend beyond the limits of the chart, thus, not all orbits in the family appear in the plot. Also, the L3 families of halo and axial orbits exist for a range of Jacobi constant values that does not appear within the axis limits on this chart. For the L1 and L2 halo families, more than one orbit may exist for a particular value of C . In such cases, multiple lines appear on the chart to represent the family, where one line above another indicates a higher value of the amplitude ratio Az/Ay, and Ay and Az represent the maximal y- and z-excursions along an orbit. For example, two L2 halo orbits exist between the values C=3.015?3.059; for this range of C , the orbits with a higher value of Az/Ay appear as a second line above the first. So that the orbit families for each of the collinear points are clearly represented in one chart, the vertical axis in Fig. 16 serves only to differentiate between the orbit families. To include additional information, such as orbit amplitudes, it is useful to consider a reduced set of orbits for clarity. For example, the plot in Fig. 17 represents orbit amplitude information for the Earth–Moon L2 Lyapunov, vertical, halo, and axial orbits for a range of Jacobi constant values. The horizontal blue and orange dashed lines, and vertical black dashed lines are included for an upcoming example.

Amplitude and stability information for L2 Lyapunov, vertical, halo, and axial …

Amplitude and stability information for L2 Lyapunov, vertical, halo, and axial orbits (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

Additional information, such as representative transfer costs, could also be included in the analysis. Folta, Bosanac, Guzzetti, and Howell implement an interactive approach to compare transfer and stationkeeping costs, among other parameters, across various orbit types [16] and [17]. Folta et al. compute the cost to transfer to Lyapunov, halo, quasi-halo, vertical, and Lissajous orbits for a variety of Jacobi constant values, assuming a direct transfer from a 200 km low-Earth orbit (LEO) to a libration point orbit, achieved by performing one tangential maneuver (Δv1) at LEO, and one maneuver (Δv2) at the x -axis crossing of the libration point orbit where View the MathML source[27]. Varying the energy level of the orbit impacts the cost of direct transfer significantly, where cost is defined as the magnitude of Δv2 for locally optimal transfers. Generally, the cost for direct transfers decreases with Jacobi constant for the halo and Lyapunov orbits. For two orbits at the same value of C, the cost is less for that maneuver possessing the smaller z-amplitude at insertion on the libration point orbit. Many other transfer types exist and could be included in the analysis. For example, transfers to halo orbits employing stable manifolds have been computed by Parker and Born [28]; Folta et al. [27] additionally demonstrated a reduction in cost achieved by including a maneuver near the Moon, where the cost to insert into an L2 halo orbit decreases with Az.

5.1. Informing the orbit selection process

Including charts that supply information about the global solution space in the vicinity of the collinear points within the mission design process allows mission designers to exploit the currently available information, and improves the efficiency of the design process. The charts in Fig. 16 and Fig. 17 are useful to

## Stability information for Lyapunov and vertical orbits in the

Stability information for Lyapunov and vertical orbits in the Earth–Moon system; Jacobi constant denoted by color (a) L1 Lyapunov family (b) L1 vertical family (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

Stability information for Lyapunov and vertical orbits in the Earth–Moon system; …

Stability information for Lyapunov and vertical orbits in the Earth–Moon system; Jacobi constant denoted by color (a) L2 Lyapunov family (b) L2 vertical family (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

Stability information for Lyapunov and vertical orbits in the Earth–Moon system; …

Stability information for Lyapunov and vertical orbits in the Earth–Moon system; Jacobi constant denoted by color (a) L3 Lyapunov family (b) L3 vertical family (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

Examining the buy BMS777607 of the orbital stability across the Lyapunov and vertical families, several bifurcations exist and are summarized in Table 1. The first bifurcation, Ly-1, in each of the Lyapunov families yields the families of halo orbits. The northern halo families are plotted for the Earth–Moon system in Fig. 6. Southern families are computed by reflecting the northern families across the x?y plane. The second bifurcation, Ly-2, signals the emergence of the families of axial orbits from the plane. Portions of the axial families appear in Fig. 7. Only those orbits for which z>0 at the maximal value of y are plotted. The second half of the families are computed by reflecting these members across the x?y plane. Again, the individual orbits within the families in Fig. 6 and Fig. 7 are colored according to the associated value of Jacobi constant, however, the color mapping is not consistent across the different families. The L1 and L2 Lyapunov families possess a third bifurcation, Ly-3, which corresponds to a period-doubling bifurcation. The third bifurcation, Ly-4, in the L3 Lyapunov family connects to families of planar orbits that originate from the equilateral points, L4 and L5[8]. From examination of the stability plots for the vertical families, several additional bifurcations are apparent. The first bifurcation in each family is labeled V-1, and corresponds to a bifurcation to the respective axial families. The bifurcation, V-2, corresponds to a period-doubling bifurcation from a family of planar orbits [21]. The L3 vertical family connects to the L4 and L5 families of vertical orbits via the bifurcation V-3 [8].

Bifurcations in periodic orbit families.

Label Bifurcation

Ly-1 Halo family

Ly-2 Axial family

Ly-3 Period-doubling

V-1 Axial family

V-2 ‘Reverse’ period-doubling

V-3 L4 and L5 vertical families

Table options

Sample members from the families of halo orbits in the Earth–Moon system; Jacobi …

Sample members from the families of halo orbits in the Earth–Moon system; Jacobi constant denoted by color (a) L1 family (b) L2 family (c) L3 family (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

Sample members from the families of axial orbits in the Earth–Moon system; …

Sample members from the families of axial orbits in the Earth–Moon system; Jacobi constant denoted by color (a) L1 family (b) L2 family (c) L3 family (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

Plots representing the stability corresponding to the halo orbits appear in Fig. 8. For the halo families, the number of complex eigenvalue pairs is represented as a function of orbit amplitude ratio Az/Ay. At the points H-1, H-2 and H-4, the L1 and L2 families experience period-doubling bifurcations. A stability change occurs in the family at H-3 and H-5, but does not lead to any new orbit families [5]. The L2 halo family undergoes a period-doubling bifurcation, H-6, that yields the family of L2 butterfly orbits [8]. Only those orbits with perilune above the surface of the Moon are included in the plots, thus, a bifurcation from the L1 family of halo orbits to the L4 and L5 families of axials orbits does not appear in the L1 halo stability chart [8]. The axial orbits are unstable (nc=2) for each of the families.

Stability information for halo orbits in the Earth–Moon system (a) L1 family (b) …

Stability information for halo orbits in the Earth–Moon system (a) L1 family (b) L2 family (c) L3 family (For interpretation of the references to color in this figure caption, the reader is referred to the web version of this paper.)

Figure options

3.2. Quasi-periodic orbits

## Once their covariance matrices are

Once their covariance matrices are translated to the B-plane reference system, the combined covariance matrix can be obtained. By means of the deterministic formulation, the collision risk associated with each point of the related mean B-plane is computed. Those points with an associated collision risk larger than the ACPL define an area (manoeuvring area) composed of those points that would involve an avoidance manoeuvre. The number of avoidance manoeuvres, related to each one of the debris object groups, is then dependent on the size of that area and the associated flux, caused by the catalogued objects (Fj? in the following). The total manoeuvre rate MA is the sum of the number of avoidance manoeuvre rates over all the catalogued groups.

equation(5)

View the MathML source

Turn MathJax on

The risk associated with the catalogued objects may also be divided into the same orbit groups, thus, Eq. (2) (accounting only for catalogued objects) can be written in the following way:

equation(6)

View the MathML source

Turn MathJax on

The integral operations in this former equation can be split in two parts: the first part associated with the manoeuvring area, and the second covering the rest of the space.

equation(7)

View the MathML source

Turn MathJax on

2.3.2. Risk reduction and residual risk

The first term in Eq. (6) is associated with the reduction of risk related to the avoidance strategy. The second term represents the existing risk that is not intended to be reduced. The ratio between these two quantities and the total risk associated with the catalogued objects are known as the fractional risk reduction and fractional residual risk, respectively. They provide information on the appropriateness of the avoidance strategy.

As explained before, ACRC does not provide the value of the total risk, but the risk due to catalogued objects. Thus, the residual risk is the risk that could be avoided, but is not. If the total remaining risk is of interest, ACRw has to be used: subtracting the risk reduction from ACRw, the total remaining risk is obtained. A fractional remaining risk can also be obtained by normalising the remaining risk with the total ACRw.

TLE and CSM are the sources of data for the operational collision avoidance activities, but an estimated ASP 1517 cost is needed for the analysis of collision avoidance requirements in mission design. The assessment of the number of near-miss events and avoidance manoeuvres during the lifetime of a satellite requires the knowledge of the object population that can cause conjunction events. The population provided by the MASTER-2009 environment model[19] for objects larger than 1 cm in size is used within the ARES software package described in this paper.

MASTER provides a semi-deterministic population of natural and man-made objects for a reference epoch. Natural objects are discarded for collision avoidance manoeuvres computations. MASTER provides data on the whole orbiting population, but avoidance manoeuvres are dependent on that part of the population that can be tracked. Thus, the incompleteness of the catalogue must be taken into account by introducing a correcting function to consider only population data within the catalogue size limit.

As already mentioned, a main issue related to risk reduction is the uncertainty in the orbital data. The orbit determination uncertainty of the population objects, together with the orbit determination of operating spacecraft play an important role in the risk associated with each near-miss event. The orbit determination uncertainty of the spacecraft and population objects is handled in terms of the associated error covariance matrices. In the case of the debris objects, the covariance matrices are dependent on the type of orbit. ARES provides suitable covariance data for different orbital regimes and orbital information data sources (TLE or CSM) [11]. The following section provides some information on the uncertainty associated with those orbital data sets, and how this information is obtained for ARES.

2.5. Accuracy of the orbital data Sets

ARES provides position uncertainty information for the debris populations. Two kinds of covariances are provided based on TLE accuracy and based on JSpOC CSM accuracy. These covariances were the result of an extensive study described in Domínguez-González, 2013 [8]. In that reference:

CSM typical uncertainties in an UVW (radial/along-track/cross-track) orbital frame were obtained by averaging the CSM-provided covariance matrices as a function of the time to event.

TLE typical uncertainties in an UVW (radial/along-track/cross-track) orbital frame were obtained by comparing a large number of precise (post-processed) reference orbit arcs with the orbit arcs predicted by TLEs.

Fig. 2 shows the TLE-type position uncertainties for an example satellite, as a function of the SGP-4 propagation interval. For each orbit regime, uncertainties from several satellites are averaged. ARES users can use the pre-computed covariances for the debris population and (optionally) for the spacecraft determination. Also it is possible to define arbitrary scaling factors for the covariances, and even use a custom covariance set.

## order moexipril The th of November marked then the first

The 12th of November 2014 marked then the first landing of a spacecraft – the lander Philae – on a comet. The push-off from the Rosetta orbiter and descend towards the selected landing site ?Agilkia? were nominal during the actual landing attempt. Touchdown occurred very close to the targeted landing site coordinates. This event was preceded by a Landing Site order moexipril Process (LSSP) which led to the selection of ?Agilkia? [4]. A landing gear performance and touchdown safety assessment was made in the frame of this site selection process. This paper describes the analysis tools and methods developed for this assessment. Note, that the analyses assumed a working ADS.

A high-fidelity multi-body simulation of the lander – described in Section 2 – represents the landing gear dynamics and system response upon touchdown. The lander kinematics and force laws implemented in the model which determine the touchdown dynamics are introduced in this section. Performance metrics and safety figures are defined in Section 3 to measure the success of a particular landing case and the remaining landing gear performance margins. Site and landing scenario specific expected values for a safe landing and its margins are derived from this analysis in conjunction with Monte Carlo trajectory data from the flight dynamics analysis (Section 4).

The actual landing – however – was hampered by failures of both the hold-down thrust and the anchoring harpoons. Consequently, the lander bounced-off the surface after initial touchdown and drifted to its final, unintended landing position [5]. Section 5 of this paper reviews the touchdown simulation and assessment process in view of the actual landing taking into account also the subsystem failures.

2. Philae touchdown dynamics simulation

The purpose of a high-fidelity touchdown (abbreviated: T/D) simulation is to provide an accurate representation of the lander dynamics and surface interaction upon touchdown. Such an engineering simulation has already been used during the design and development phase of the lander in 1996–2002. These early simulation are described in [6]. An experimental landing test campaign [3] was done to review the landing performance and to be incorporated into the operations planning of the upcoming landing. This test campaign made use of a new test facility to provide data beyond those available during the design, development and qualification of Philae. The numerical simulation was newly set up with improved fidelity in view of new landing gear test data and findings from this campaign. The simulation is checked and validated against these experimental data.

The landing system is modeled as a multibody mechanical system. Its implementation uses the commercial multibody simulation software tool SIMPACK [7]. Model elements are bodies, joints or constraints and forces. Bodies represent the geometry and mass properties of the lander, joints or constraints are connecting the bodies and determine the degrees of freedom of the complete assembly. The resulting multibody topology is depicted in Fig. 1.

Philae lander multibody topology showing kinematic relations of different lander components, their relative degrees of freedom and applied forces.

Figure options

The force elements act on the various bodies according to certain force laws (e.g. the thrust profile of the Active Descend System ADS). Control logic is implemented to mimic the onboard logic which detects the touchdown event and initiates events such as the firing of anchoring harpoons and the ADS activation. The following functionalities and forces are represented:

2.1. Damper assembly

The electro-mechanical damping device (Fig. 2) translates the damper stroke into a rotation which drives an electric generator. The electrical energy is then dissipated within a resistor. The resulting damping force Fd depends on the relative actuation velocity vd between the landing gear assembly and the lander body and is described by a complex transfer function (Eq. (1)).

equation(1)

View the MathML source

Turn MathJax on

Schematic view of Philae landing gear damper assembly: the stroke of the damper tube is translated via a cable pulley and spindle/nut combination into a rotation driving an electrical generator. The energy is dissipated by a resistor.

Figure options

The moment of inertia of all of the rotating elements is collectively described by IR, whereas kd stands for the stiffness of (primarily) the cables, σ is the spindle thread pitch and d is a damping coefficient. The quasi-stationary transfer behavior (Eq. (2)) shows that the damping force is linearly proportional to the velocity with b=567 N s/m. A detailed description of this assembly and its dynamics is given in [3].

equation(2)

View the MathML source

Turn MathJax on

2.2. Footpad–soil-contact

This force law models the comet surface as a kind of granular material (dust/ice agglomerates). Its compressive strength provides the resistance against the penetrating feet as the surface yields. The adopted soil force model is the simplest imaginable for granular beds in micro-gravity, thus with no depth-dependence. Dynamical resistance is neglected. The penetration resistance Fs,norm of any lander element is then given by the compressive strength sc of this idealized material times the penetrating cross section Av=0.017 m2 (Eq. (3)). This area is mainly determined by the two soles of each foot, see [8]. The landing feet are retained by the shear forces between the comet material and the ice screws of the lander. The retention force is given by material?s shear strength ss times the side wall area Ah=0.0069 m2 of the penetrating object (Eq. (3)). The assumed strength values – compressive sc=7 kPa, shear ss=1 kPa – are taken from [8]. The model above is consistent with laboratory measurements (slow horizontal drag of a test body in a granular medium under 1 g, [9]) but not with assumed tri-directional compression because the particles cannot evade to the sides when an object is penetrating as in [10]. While these parameter assume a lower bound for a weak surface in the simulated landing scenarios, a second case considers a hard, nearly rigid ice crust with a compressive strength of sc=2 MPa. The ice screws are not extended and the footpad penetration is negligible. Consequently, with Av=0 this case, there is no retention force acting in this hard surface case.

## Design parameters of the integration Sign Definition Value

Design parameters of the integration.

Sign Definition Value

Geometric parameter L Length of integration 3800 mm

Xcoj Intersection in X axis 700 mm

s Height ratio of Intersection 0.9

Ls Straight length of ICC 10 mm

n Exponent of ICC curve 2

Lu Straight length of FCC 100 mm

db Width of Busemann inlet 700 mm

Flow parameter M∞ Design freestream Mach 6

Hfh Design flight height 30 km

Table options

In Fig. 15, the two-stage osculating cones waverider of the integration is completely identical to the first two stages of the three-stage waverider. From bottom view, the Busemann-inlet portion of the integration has greater IWR-1-endo cost surface than the third compression surface of the three-stage waverider. From front view, both of the two models have closely same mass flowrate, for the cross-sectional areas of the integration and the three-stage waverider are respectively 0.221 m2 and 0.219 m2 at the isolator. Because the Busemann inlet compresses airflow more slowly, the integration has a greater third-stage compression surface.

5.2. Performance analysis of the integration

First, the aforementioned model derived by the integration method is analyzed, and its inviscid and viscous performances are determined numerically in the design and off-design conditions. The computational states are 30 km height, Mach number 5, 6, 7 of the freestream, 0°, 2°, 4°, 6° of the attack angle. The CFD code [21] developed by our research team is employed to complete the simulations. In this paper, a LU-SGS scheme for temporal discretisation, a Roe scheme for spatial discretisation and a S–A turbulence model if in need are executed for the inviscid and viscous states. And the CFD results are non-dimensionally presented by Eq. (15).

View the MathML source

Turn MathJax on

Here, the overline ‘～’ indicates dimensional variables, and the subscript ‘∞’ indicates the parameters of freestream.

Fig. 16 shows the Mach contour of the integration at the inlet entrance under the design conditions (View the MathML source, Ma =6, α=0°). In the figure, the blue dotted line is the predicted location of the ICC. The inviscid Mach contour on the left side shows that the location of the shock wave has good agreement with the predicted ICC, which indicates the validity of the integration method based on the mutistage compression waverider and Busemann inlet under the inviscid condition. However, the viscous Mach contour on the right side shows that the location of the shock wave at the inlet entrance has a slightly downward deviation relative to the predicted ICC, and the airflow leaks out a bit at the intersection edge of the compression surface and freestream surface. This is because of the presence of the boundary layer. The density contour at the symmetric plane in Fig. 17 also shows that the two shock waves of the two-stage osculating cones waverider intersect at the inlet lip (which is the predicted ICC). The results show that the proposed design can meet the prospective requirements except a bit of leak at the inlet lip and a greater low-pressure area at the start of isolator under the viscous condition.

Mach contours of the integration at the inlet entrance under design conditions (View the MathML source, Ma =6, α=0°).

Figure options

Density contours of the integration at the symmetric plane under design conditions (View the MathML source, Ma =6, α=0°).

Figure options

Fig. 18 shows the aerodynamic parameters of the integration at different attack angles including lift coefficient, drag coefficient, pitching moment coefficient and lift-to-drag ratio under inviscid and viscous conditions, as Eq. (16).

View the MathML source

Turn MathJax on

Aerodynamic parameters of the integration at different attack angles.

Figure options

Here, L, D, and Mo are lift, drag, and pitching moment, respectively, of the forebody. q∞ is dynamic pressure of freestream, Sref and lref are reference area and length of the forebody respectively. In this study, the reference area is the area of the freestream capture section shown in Fig. 19, which is 0.876968 m, under the design conditions (View the MathML source, Ma =6, α=0°). The height of the model (0.9123 m) is used as the reference length, and the highest point of the model at the inlet entrance section is used as the pitching moment reference center, as shown in Fig. 19. When integrating the aerodynamic parameters on the forebody surface (freestream surface and three-stage compression surface), the base surface of forebody is integrated using the freestream parameters in consideration of the forebody closure and the comparability of aerodynamic parameters. In Fig. 18, the variation trends of the inviscid and viscous aerodynamic parameters stay consistent at different attack angles under the design and off-design conditions. The lift, drag, and the pitching moment increase with increasing attack angle and decrease with increasing Mach number. The lift-to-drag ratio decreases when Mach number increases and its max value occurs at the design attack angle 0° at different Mach numbers. Because of the influence of viscosity, the viscous lift coefficient and drag coefficient both increase relative to the inviscid ones. This is probably because the hypersonic viscous interaction effects increase the surface pressure (which increases the shock wave drag) and the surface friction resistance (which increases the skin friction drag). Both the two kinds of resistance can increase the total drag of the model. So the drag interference is larger relative to lift interference. The max variations of the drag coefficient are ?23.85%, ?22.58%, and ?22.12% respectively at Mach number 5, 6, and 7. However, the hypersonic viscous interaction effects work on both of the freestream surface and compression surface and mostly counteract in the lift direction. So the lift coefficient just increases slightly [22] and [23]. The max variations of the lift coefficient are 1.67%, 1.40%, and 1.52% respectively at Mach number 5, 6, and 7. The pitching moment coefficient is mostly affected by the lift so that it has the same variation trend with lift coefficient. Mainly owing to the increasing drag, the lift-to-drag ratio decreases with the increasing Mach number and attack angle.

## Baseline scenario without constellation red lines vs constellation scenario

Baseline scenario without constellation (red lines) vs. constellation scenario where the constellation performs direct re-entry with 90% success (black lines). The thick lines are means of the MC runs and the dashed lines are the standard deviation around the means with a 95% confidence interval.

Figure options

Constellation scenario where the constellation performs direct re-entry with 90% success (red lines) vs. constellation scenario where the constellation disposes to eccentric orbit with 10 year lifetime with 90% success (black lines). The thick lines are means of the MC runs and the dashed lines are the standard deviation around the means with a 95% confidence interval.

Figure options

Constellation scenario where the constellation disposes to eccentric orbit with 10 year lifetime with 90% success (black lines) vs. constellation scenario where the constellation disposes to eccentric orbit with 25 year lifetime with 90% success (red lines). The thick lines are means of the MC runs and the dashed lines are the standard deviation around the means with a 95% confidence interval.

Figure options

1. Introduction

Rosetta is a cornerstone scientific mission of the European Space Agency [1], [2] and [3], launched on 2nd March 2004 on an Ariane 5G+ rocket. Its main scientific objective was to rendezvous with the nucleus of comet Churyumov-Gerasimenko in 2014, to orbit it buy SB 203580 hydrochloride for about 1.5 years and to deliver onto the nucleus’ surface a landing module named Philae. Seven years of active cruise, in which several planet gravity assist manoeuvres and two asteroid fly-bys were carried out [4], [5], [6], [7], [8], [9] and [10], were followed by the final part of the cruise, where Rosetta (Fig. 1) had to fly at distances from the Sun that had never been reached before by a solar-powered spacecraft (aphelion was reached on 3rd October 2012 at about 5.3AU distance) [11]. Notwithstanding the large solar array (64 m2), in order to survive at such Sun distances, the spacecraft had to be almost fully deactivated from June 2011 to January 2014 [12] and [13] to drastically reduce the power consumption. After reactivation, the final mission phase was flown towards its target, which was reached early August 2014; a full comet mapping and characterisation phase was then carried out, leading to the selection of a landing site for Philae [14].

Rosetta interplanetary trajectory.

Figure options

Section II of this paper describes the operations of the comet orbital phase including the landing event. The evolution of the spacecraft orbit and the planning scheme as a function of the growing comet activity is described in section III. Section IV describes the perihelion phase and the plan for the next mission phases. Lessons learned and conclusions are addressed in sections V and VI respectively.

2. Comet orbital phase

After having characterised the comet to the level of detail necessary for orbiting and landing (global mapping phase), Rosetta began its close observation phase early October 2014 when the orbit radius was changed from ca. 29–19 km and the spacecraft, after an excursion to the night side (120 deg phase angle with the Sun), was moved to the terminator plane (see Fig. 2). On this orbit only the edge of the large solar panels, which are pointing to the Sun, is exposed to the – mostly radial – gas flow originating from the comet; this reduced to the minimum the aerodynamic accelerations, which could not be predicted accurately enough yet by the comet engineering model being developed in parallel.

Global mapping and close observation orbits.

Figure options

A further reduction of the orbit radius to ca. 9 km was then conducted once the operations team had reached an adequate level of confidence in their engineering models of the comet environment. This was possible only after having performed an excursion at lower altitudes when the pericenter of a 9×19 km2 orbit was flown. During this pericenter pass the error on the predicted vs. actual spacecraft position reached a value of ca. 1.7 km at a distance of ca. 10 km thus resulting in an attitude pointing error of ca. 10 deg (see Fig. 3), i.e. translated in pointing error relative to the comet this well beyond the field of view (FoV) of the vast majority of the remote sensing instruments (e.g. navigation and science cameras) and in particular preventing optical navigation.

Errors at 9 km distance.

Figure options

With the data collected during this pericenter pass the model of the gravity field was updated with second order harmonics (see Fig. 4) and sufficiently accurate navigation on a 9 km circular orbit became possible.

Accelerations and SC-comet distance (Jul to Oct 2014).

Figure options

After an initial screening of the whole comet surface, the comet models developed with data collected throughout this phase allowed the selection of the 5 candidate landing sites (see Fig. 5).

Candidate landing sites. Credits: ESA/Rosetta/MPS for OSIRIS Team MPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDA

## Hence in this study we are concerned to numerically

Hence, in this study we are concerned to numerically investigate the scale effect of varying injection diameters, in a cold kerosene-fueled scramjet combustor, on the interaction between incident shock wave and transversal cavity injection using a combined three dimensional Couple Level Set & Volume of Fluids (CLSVOF) approach with an improved K-H & R-T model.

2. CFD model and simulation approach

Fig. 1 schematically illustrates the real prototype of scramjet combustor presented by Liu [16]. In the original design, for the purpose of experimental measurements, the scramjet combustor contains flange 1 and 6, pressure pad of upper glass window 3 and upper glass window 4. To simply the CFD model, however, the foregoing components are ignored because they do not affect the interaction between kerosene and supersonic flow. Though the strut 8 affects the flow patterns of the mixture in the combustor, it is not consider either due to the fact that there is not the component in the present experimental setup of Liu\’s. As a result, the scramjet combustor in this study consists of rear cover 2, upper cover 5, lower cover 7 and cavity 9. Fig. 2 shows the simplified geometry of the three-dimensional scramjet combustor by using a combined feature-based modeling approach and Virtual Assembly Technique (VAT) [17]. The primary specifications used for the calculation are listed in Table 1. Note that the incident shock wave, generated by kerosene, is injected from the orifice at the center of the cavity (see Fig. 2(b)).

Figure options

(a) A three-dimensional model of the scramjet combustor, (b) computational domain of (a) and (c) the cavity configuration.

Figure options

Specifications of the scramjet combustor.

Item Dimension

Table options

Fig. 3 shows the meshed CFD model of the scramjet combustor. There are totally 330,000 hexahedron protein kinase used for the combustor, in which the mesh cells are concentrated around the walls and the region near the cavity due to the strong interaction between incident shock wave and transversal cavity injection.

Details of numerical grid of the scramjet combustor.

Figure options

The interaction between incoming flow and kerosene was simulated by commercial CFD software ANSYS Fluent 14.0 to understand the interface breakup and coalesce with another interface. As for this, Volume of Fluids (VOF) model is generally used for the underlying physical mechanisms [18]. However, additional re-meshing is necessitated when VOF is applied for a large deformation of the gas-liquid interface [18]. The mixing process of incident shock wave and transversal cavity injection in supersonic flows generally results in more complex turbulent structures [19]. Hence, in this study three dimensional Couple Level Set & Volume of Fluids (CLSVOF) model was proposed to predict the scale effect of various injection diameters on the breakup behaviors of kerosene droplets in the cold kerosene-fueled scramjet combustor. This may be explained as due to CLSVOF coupling the LS (Level Set) and VOF. In detail, CLSVOF model contains the advantage of VOF: automatically deal with topological changes with a higher-order of accuracy, and also overcomes the disadvantage of VOF: unable to accurately compute such important properties as the curvature and the normal to the interface [20]. Consequently, CLSVOF is more accurate than both the standalone LS and VOF models [20]. Note that the LS and VOF methods belong to one fluid method. A single set of the conservation equation is therefore used for the whole domain, and there are no separate gas-liquid velocities. In this case the Navier-Stokes has the following form [20]:

equation(1)

View the MathML source

Turn MathJax on

and the continuity equation is

equation(2)

View the MathML source

Turn MathJax on

where View the MathML source is the velocity vector, ρ is the fluid density defined by Eq. (3), t is the time, μ is the fluid viscosity defined by Eq. (4), D is the viscous deformation tensor defined by Eq. (5), g is the gravity vector and the body force due to the surface tension,Fst, is defined using the immersed boundary method to represent the presence of the solid surface in the fluid. We refer the readers to the work of Yu [21] and Mènard [22] for details of the equations (3)~(5) and the implementation of the immersed boundary method.

equation(3)

Turn MathJax on

equation(4)

Turn MathJax on

equation(5)

Turn MathJax on

In addition, both energy equation and the state equation of the gaseous mixture were also taken into account in this study. The detailed formulas are presented as follows:

The energy equation for a droplet is [23]:

equation(6)

View the MathML source

Turn MathJax on

where View the MathML source, View the MathML source-the energy of phase transitions, cvs-specific heat capacity, hL-the latent heat of evaporation, q―heat flux to a single droplet from the surrounding gas flow. It is determined by the following forms [24]:

equation(7)

View the MathML source