Résultats de recherche pour « MARL »

High concentrations of phosphate ion precipitate with carbonates by simple coprecipitation as the pH of marl lake water is increased. High pH microenvironments associated with photosynthesis probably induce phosphate‐carbonate precipitation, which in turn functions as one of several indirect density‐dependent population controls.

Near-real-time water-quality monitoring in uncertain environments such as rivers, lakes, and water reservoirs of different variables is critical to protect the aquatic life and to prevent further propagation of the potential pollution in the water. In order to measure the physical values in a region of interest, adaptive sampling is helpful as an energy- and time-efficient technique since an exhaustive search of an area is not feasible with a single vehicle. We propose an adaptive sampling algorithm using multiple autonomous vehicles, which are well-trained, as agents, in a Multi-Agent Reinforcement Learning (MARL) framework to make efficient sequence of decisions on the adaptive sampling procedure. The proposed solution is evaluated using experimental data, which is fed into a simulation framework. Experiments were conducted in the Raritan River, Somerset and in Carnegie Lake, Princeton, NJ during July 2019.

Abstract The paper outlines the physical parameters and textural differences of clay marl, marl, and calcareous marl, focusing on the physical changes that are caused by water. The characterization of marls is often very difficult since their properties are variable, covering a broad range from low-strength friable rock to highly cemented durable lithologies. Consequently, the prediction of their physical parameters requires further consideration. The paper addresses this issue by physically and texturally characterizing Eocene marly lithologies and comparing their properties to various limestones. Density, water absorption, ultrasonic pulse velocity, uniaxial compressive strength (UCS), Brazilian tensile strength, Young’s modulus, and their interdependencies are explained. The link between physical properties and various micro-fabrics is also outlined, with the help of new equations. Our tests show that UCS of calcareous marl can be as high as 100 MPa, but it decreases when the sample is water saturated. Relationships were calculated between the physical parameters of air-dry and water-saturated samples. Correlations between Young’s moduli and uniaxial compressive strength are also explained. The results of our analyses were compared to published physical parameters of marls and limestone, and relationships were outlined between the parameters of these carbonates.

The physicochemical and sunscreen properties of marl as a function of particle size were investigated. The research findings established that the marl under investigation consisted of more than 95% calcium carbonate (CaCO3). The particles of marl inspected under a scanning electron microscope were calcite, which is the stable polymorph of CaCO3, with a rhombohedral structure. The particle size classification by the sieving method showed that grinding using a ball mill could downsize the marl particles by 2 to 3 times, reaching below 15 µm on average. Marl particles showed a tendency to reflect ultraviolet A (UVA) rays rather than UVB rays and a possibility to steadily absorb both UVAII and UVAI. Finer particles obtained after a longer grinding process demonstrated higher efficacy regarding UV reflection and absorption properties. The 3 wt.% marl displayed a sun protection factor (SPF) value of 1 to 2. However, marl demonstrated a good ability to protect against radiation over a broad spectrum range with a critical wavelength above 370 nm. The addition of marl in the formulation containing avobenzone and octinoxate had a positive synergistic effect because the marl was able to increase the UV absorbance efficacy (based on the area under the curve (AUC) value) and SPF value of the cream. Furthermore, it was also discovered that the added marl powder could slow the decrease in UV protection efficacy of the products in terms of the AUC calculated from the absorbance profile after exposure to simulated UV rays with an amplitude range of 10 J/cm2 to 40 J/cm2 for 30 min, which was similar to the results obtained from octocrylene and bemotrizinol.

ABSTRACT Relatively pure lacustrine carbonates referred to as marl are being deposited in Littlefield Lake, central Michigan, a hard‐water lake with little terrigenous clastic influx. Thick accumulations of marl form both progradational marl benches along lake margins, and islands or lakemounts in the lake centre. Marl benches develop flat platforms up to 20 m wide in very shallow water and steeply inclined slopes, up to 30°, extending into deep water. The flat landward platform is frequently covered by algal pisoliths while the upper portion of the lakeward‐sloping bottom is overgrown by Chara which in the summertime becomes thickly encrusted with low‐magnesian calcite. Marl islands are flat‐topped features that formed over relict highs on Pleistocene drift which underlies the lake basin. These are fringed by marl benches identical to those found along lake margins. Marl benches are composed of four units: two thin facies deposited on the shallow‐water bench platform and two thicker faces deposited on the bench slope developed in moderate water depths. These in turn overlie a fifth facies deposited in deep water. A coarsening‐upward sequence is developed in these sediments as a result of both mechanical sorting, and primary production of carbonate sand and gravel in shallow water. In addition to facies sequences and size grading, trends upsection of increasing carbonate content and decreasing insoluble content may serve to identify temperate‐region lacustrine carbonate deposits in the rock record.

The term “marl” is used to designate soft, carbonate-rich, fine-grained soils, which pose concerns related to both settlement and stability. Despite the prevalence of marls in Indiana and the concerns associated with their behavior, very limited work has been done to study the engineering properties of these soils. This was the motivation for this research project, which involved two primary activities: a) the creation of a map and database of existing information on marl deposits in Indiana; and b) an in-depth characterization of the properties of a marl deposit in Daviess County, which was considered representative of similar deposits encountered in Indiana. The marl database was generated using ArcGIS 10.0.from information available at the INDOT, and involved mining data from over five thousand boreholes. The second part of the project involved field tests (seismic cone penetration tests, standard penetration tests, field vane shear tests), and laboratory experiments (index tests, incremental and constant rate of strain consolidation tests, and K0-consolidated undrained triaxial tests) conducted on high quality Shelby tubes samples. Additionally, the mineralogy and the microstructure of the soil were studied in detail. The laboratory tests reveal that the deposit was not homogeneous as was initially anticipated, but was, instead, formed by two types of soils that repeat in horizontal thin layers. These two soils, referred to as ‘soil M’ and ‘soil C’, are both characterized by very high calcium carbonate contents but show distinct index and engineering properties, that may be ascribed to differences in mineralogy and composition. This stratification is not detected by the field tests. The consolidation tests show that the deposit has an OCR less than 2 and compressibility parameters markedly dependent on stress level, as typical of sensitive soils. K0-consolidated undrained compression triaxial tests show that both soils exhibit normalized behavior, and that the relationship between strength and stress history is well described by the SHANSEP equation (although the SHANSEP parameters differ for the two soils). Comparison of the field data and laboratory results provides the means to validate published correlations for interpretation of the geotechnical properties of marls from field results. For the site examined, correlations to estimate shear wave velocity, stress history, and undrained strength from CPT results are identified. Implementation recommendations are provided for soil identification, sampling and specimen preparation, interpretation of filed data, and preliminary design.

We present a comparative study of multi-agent reinforcement learning (MARL) algorithms for cooperative warehouse robotics. We evaluate QMIX and IPPO on the Robotic Warehouse (RWARE) environment and a custom Unity 3D simulation. Our experiments reveal that QMIX's value decomposition significantly outperforms independent learning approaches (achieving 3.25 mean return vs. 0.38 for advanced IPPO), but requires extensive hyperparameter tuning -- particularly extended epsilon annealing (5M+ steps) for sparse reward discovery. We demonstrate successful deployment in Unity ML-Agents, achieving consistent package delivery after 1M training steps. While MARL shows promise for small-scale deployments (2-4 robots), significant scaling challenges remain. Code and analyses: https://pallman14.github.io/MARL-QMIX-Warehouse-Robots/

KENTUCKY marls or calcareous clays are usually soft, calcareous-magnesian shales or soft, clay-containing magnesian limestone which breaks down on weathering into finely divided particles. In recent years marl has been used extensively because of its cheapness and easy accessibility, as a source of lime for soils. The neutralizing values of these marls vary from 20 to 98 percent (estimated as calcium carbonate). Because of the general distribution of marls in certain localities in Kentucky the establishment of their value as sources of calcium for laying hens is of economic interest. Three typical marls and a high grade limestone were selected, the analyses of which were: These marls and limestone were ground to pass through a 100-mesh sieve and added to a basal ration which consisted of shipstuff 60, ground yellow corn 20, meat scrap 20 and cod liver oil 1 pound and fed to Rhode Island Red pullets. The analysis . . .