Search results for: biogas upgrading energetic cost
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 6453

Search results for: biogas upgrading energetic cost

4923 Development of an Indigenous Motorized Planter for the Sustainable Production of Grain Crops in Nigeria

Authors: Babatunde Oluwamayokun Soyoye

Abstract:

This technology, whose development revolves round culture, tradition, and prevailing needs of the people, is seen as a solution in promoting development in poor rural communities in many parts of Nigeria. The research was based on one of the food security agenda of the Federal Government of Nigeria by developing a motorized multi-grain crop planter suitable for planting operations in tropical soils. The ergonomic design is tailored towards the ease of planting operations for would-be users, improve crop yields and profitability by minimizing the cost of production. Some properties of the grain crops were determined and were used to develop and assemble the locally-made motorized planter. These properties were used in establishing the design criteria of various components of the planter. The geometric mean diameter of the maize, cowpea, groundnut, and soybean were 8.26 mm, 8.72 mm, 9.51 mm and 6.52 mm respectively, with respective groove depths of 8 mm, 7 mm, 9 mm and 6 mm. The results obtained from the evaluation of the planter confirmed that the planter has a uniform discharge and application rates. The field capacity of the planter was determined to be 0.187 ha/h. Also, the average performance efficiency of the planter was 95.5%, with the average discharge and application rates of 7.86 kg/h and 42.1 kg/ha, respectively. The motorized multi-grain planter can be used in increasing food production, reduce time, cost of production, and can become a major tool to fast-track the food security agenda of the government of Nigeria.

Keywords: design and fabrication, food security, grain crop, motorized planter

Procedia PDF Downloads 129
4922 Non-Linear Finite Element Investigation on the Behavior of CFRP Strengthened Steel Square HSS Columns under Eccentric Loading

Authors: Tasnuba Binte Jamal, Khan Mahmud Amanat

Abstract:

Carbon Fiber-Reinforced Polymer (CFRP) composite materials have proven to have valuable properties and suitability to be used in the construction of new buildings and in upgrading the existing ones due to its effectiveness, ease of implementation and many more. In the present study, a numerical finite element investigation has been conducted using ANSYS 18.1 to study the behavior of square HSS AISC sections under eccentric compressive loading strengthened with CFRP materials. A three-dimensional finite element model for square HSS section using shell element was developed. Application of CFRP strengthening was incorporated in the finite element model by adding an additional layer of shell elements. Both material and geometric nonlinearities were incorporated in the model. The developed finite element model was applied to simulate experimental studies done by past researchers and it was found that good agreement exists between the current analysis and past experimental results, which established the acceptability and validity of the developed finite element model to carry out further investigation. Study was then focused on some selected non-compact AISC square HSS columns and the effects of number of CFRP layers, amount of eccentricities and cross-sectional geometry on the strength gain of those columns were observed. Load was applied at a distance equal to the column dimension and twice that of column dimension. It was observed that CFRP strengthening is comparatively effective for smaller eccentricities. For medium sized sections, strengthening tends to be effective at smaller eccentricities as well. For relatively large AISC square HSS columns, with increasing number of CFRP layers (from 1 to 3 layers) the gain in strength is approximately 1 to 38% to that of unstrengthened section for smaller eccentricities and slenderness ratio ranging from 27 to 54. For medium sized square HSS sections, effectiveness of CFRP strengthening increases approximately by about 12 to 162%. The findings of the present study provide a better understanding of the behavior of HSS sections strengthened with CFRP subjected to eccentric compressive load.

Keywords: CFRP strengthening, eccentricity, finite element model, square hollow section

Procedia PDF Downloads 139
4921 Finite Element Analysis of Connecting Rod

Authors: Mohammed Mohsin Ali H., Mohamed Haneef

Abstract:

The connecting rod transmits the piston load to the crank causing the latter to turn, thus converting the reciprocating motion of the piston into a rotary motion of the crankshaft. Connecting rods are subjected to forces generated by mass and fuel combustion. This study investigates and compares the fatigue behavior of forged steel, powder forged and ASTM a 514 steel cold quenched connecting rods. The objective is to suggest for a new material with reduced weight and cost with the increased fatigue life. This has entailed performing a detailed load analysis. Therefore, this study has dealt with two subjects: first, dynamic load and stress analysis of the connecting rod, and second, optimization for material, weight and cost. In the first part of the study, the loads acting on the connecting rod as a function of time were obtained. Based on the observations of the dynamic FEA, static FEA, and the load analysis results, the load for the optimization study was selected. It is the conclusion of this study that the connecting rod can be designed and optimized under a load range comprising tensile load and compressive load. Tensile load corresponds to 360o crank angle at the maximum engine speed. The compressive load is corresponding to the peak gas pressure. Furthermore, the existing connecting rod can be replaced with a new connecting rod made of ASTM a 514 steel cold quenched that is 12% lighter and 28% cheaper.

Keywords: connecting rod, ASTM a514 cold quenched material, static analysis, fatigue analysis, stress life approach

Procedia PDF Downloads 298
4920 Feasibility of Two Positive-Energy Schools in a Hot-Humid Tropical Climate: A Methodological Approach

Authors: Shashwat, Sandra G. L. Persiani, Yew Wah Wong, Pramod S. Kamath, Avinash H. Anantharam, Hui Ling Aw, Yann Grynberg

Abstract:

Achieving zero-energy targets in existing buildings is known to be a difficult task, hence targets are addressed at new buildings almost exclusively. Although these ultra-efficient case-studies remain essential to develop future technologies and drive the concepts of Zero-energy, the immediate need to cut the consumption of the existing building stock remains unaddressed. This work aims to present a reliable and straightforward methodology for assessing the potential of energy-efficient upgrading in existing buildings. Public Singaporean school buildings, characterized by low energy use intensity and large roof areas, were identified as potential objects for conversion to highly-efficient buildings with a positive energy balance. A first study phase included the development of a detailed energy model for two case studies (a primary and a secondary school), based on the architectural drawings provided, site-visits and calibrated using measured end-use power consumption of different spaces. The energy model was used to demonstrate compliances or predict energy consumption of proposed changes in the two buildings. As complete energy monitoring is difficult and substantially time-consuming, short-term energy data was collected in the schools by taking spot measurements of power, voltage, and current for all the blocks of school. The figures revealed that the bulk of the consumption is attributed in decreasing order of magnitude to air-conditioning, plug loads, and lighting. In a second study-phase, a number of energy-efficient technologies and strategies were evaluated through energy-modeling to identify the alternatives giving the highest energy saving potential, achieving a reduction in energy use intensity down to 19.71 kWh/m²/y and 28.46 kWh/m²/y for the primary and the secondary schools respectively. This exercise of field evaluation and computer simulation of energy saving potential aims at a preliminary assessment of the positive-energy feasibility enabling future implementation of the technologies on the buildings studied, in anticipation of a broader and more widespread adoption in Singaporean schools.

Keywords: energy simulation, school building, tropical climate, zero energy buildings, positive energy

Procedia PDF Downloads 143
4919 Overview of Multi-Chip Alternatives for 2.5 and 3D Integrated Circuit Packagings

Authors: Ching-Feng Chen, Ching-Chih Tsai

Abstract:

With the size of the transistor gradually approaching the physical limit, it challenges the persistence of Moore’s Law due to the development of the high numerical aperture (high-NA) lithography equipment and other issues such as short channel effects. In the context of the ever-increasing technical requirements of portable devices and high-performance computing, relying on the law continuation to enhance the chip density will no longer support the prospects of the electronics industry. Weighing the chip’s power consumption-performance-area-cost-cycle time to market (PPACC) is an updated benchmark to drive the evolution of the advanced wafer nanometer (nm). The advent of two and half- and three-dimensional (2.5 and 3D)- Very-Large-Scale Integration (VLSI) packaging based on Through Silicon Via (TSV) technology has updated the traditional die assembly methods and provided the solution. This overview investigates the up-to-date and cutting-edge packaging technologies for 2.5D and 3D integrated circuits (ICs) based on the updated transistor structure and technology nodes. The author concludes that multi-chip solutions for 2.5D and 3D IC packagings are feasible to prolong Moore’s Law.

Keywords: moore’s law, high numerical aperture, power consumption-performance-area-cost-cycle time to market, 2.5 and 3D- very-large-scale integration, packaging, through silicon via

Procedia PDF Downloads 112
4918 Improvement of Energy Consumption toward Sustainable Ceramic Industry in Indonesia

Authors: Sawarni Hasibuan, Rudi Effendi Listyanto

Abstract:

The industrial sector is the largest consumer of energy consumption in Indonesia. The ceramics industry includes one of seven industries categorized as an energy-intensive industry. Energy costs on the ceramic floor production process reached 40 percent of the total production cost. The kiln is one of the machines in the ceramic industry that consumes the most gas energy reach 51 percent of gas consumption in ceramic production. The purpose of this research is to make improvement of energy consumption in kiln machine part with the innovation of burner tube to support the sustainability of Indonesian ceramics industry. The tube burner is technically designed to be able to raise the temperature and stabilize the air pressure in the burner so as to facilitate the combustion process in the kiln machine which implies the efficiency of gas consumption required. The innovation of the burner tube also has an impact on the decrease of the combustion chamber pressure in the kiln and managed to keep the pressure of the combustion chamber according to the operational standard of the kiln; consequently, the smoke fan motor power can be lowered and the kiln electric energy consumption is also more efficient. The innovation of burner tube succeeded in saving consume of gas and electricity respectively by 0.0654 GJ and 1,693 x 10-3 GJ for every ton of ceramics produced. Improvement of this energy consumption not only implies the cost savings of production but also supports the sustainability of the Indonesian ceramics industry.

Keywords: sustainable ceramic industry, burner tube, kiln, energy efficiency

Procedia PDF Downloads 318
4917 Assessing the Influence of Using Traditional Methods of Construction on Cost and Quality of Building Construction

Authors: Musoke Ivan, Birungi Racheal

Abstract:

The construction trend is characterized by increased use of modern methods yet traditional methods are cheaper in terms of costs, in addition to the benefits it offers to the construction sector, like providing more jobs that could have been worked with the intensive machines. The purpose of this research was to assess the influence of using Traditional methods of construction (TMC) on the costs and quality of building structures and determine the different ways. Traditional methods of construction (TMC) can be applicable and integrated into the construction trend, and propose ways how this can be a success. The study adopted a quantitative method approach targeting various construction professionals like Architects, Quantity surveyors, Engineers, and Construction Managers. Questionnaires and analyses of literature were used to obtain research data and findings. Simple random sampling was used to select 40 construction professionals to which questionnaires were administered. The data was then analyzed using Microsoft Excel. The findings of the research indicate that Traditional methods of construction (TMCs) in Uganda are cheaper in terms of costs, but the quality is still low. This is attributed to a lack of skilled labour and efficient supervision while undertaking tasks leading to low quality. The study identifies strategies that would improve Traditional methods of construction (TMC), which include the employment of skilled manpower and effective supervision. It also identifies the need by stakeholders like the government, clients, and professionals to appreciate Traditional methods of construction (TMCs) and allow for a levelled ground for Traditional Methods of Construction and Modern methods of construction (MMCs).

Keywords: traditional methods of construction, integration, cost, quality

Procedia PDF Downloads 53
4916 Separation of Composites for Recycling: Measurement of Electrostatic Charge of Carbon and Glass Fiber Particles

Authors: J. Thirunavukkarasu, M. Poulet, T. Turner, S. Pickering

Abstract:

Composite waste from manufacturing can consist of different fiber materials, including blends of different fiber. Commercially, the recycling of composite waste is currently limited to carbon fiber waste and recycling glass fiber waste is currently not economically viable due to the low cost of virgin glass fiber and the reduced mechanical properties of the recovered fibers. For this reason, the recycling of hybrid fiber materials, where carbon fiber is combined with a proportion of glass fiber, cannot be processed economically. Therefore, a separation method is required to remove the glass fiber materials during the recycling process. An electrostatic separation method is chosen for this work because of the significant difference between carbon and glass fiber electrical properties. In this study, an experimental rig has been developed to measure the electrostatic charge achievable as the materials are passed through a tube. A range of particle lengths (80-100 µm, 6 mm and 12 mm), surface state conditions (0%SA, 2%SA and 6%SA), and several tube wall materials have been studied. A polytetrafluoroethylene (PTFE) tube and recycled without sizing agent was identified as the most suitable parameters for the electrical separation method. It was also found that shorter fiber lengths helped to encourage particle flow and attain higher charge values. These findings can be used to develop a separation process to enable the cost-effective recycling of hybrid fiber composite waste.

Keywords: electrostatic charging, hybrid fiber composites, recycling, short fiber composites

Procedia PDF Downloads 122
4915 Parametric Study of a Solar-Heating-And-Cooling System with Hybrid Photovoltaic/Thermal Collectors in North China

Authors: Ruobing Liang, Jili Zhang, Chao Zhou

Abstract:

A solar-heating-and-cooling (SHC) system, consisting of a hybrid photovoltaic/ thermal collector array, a hot water storage tank, and an absorption chiller unit is designed and modeled to satisfy thermal loads (space heating, domestic hot water, and space cooling). The system is applied for Dalian, China, a location with cold climate conditions, where cooling demand is moderate, while space heating demand is slightly high. The study investigates the potential of a solar system installed and operated onsite in a detached single-family household to satisfy all necessary thermal loads. The hot water storage tank is also connected to an auxiliary heater (electric boiler) to supplement solar heating, when needed. The main purpose of the study is to model the overall system and contact a parametric study that will determine the optimum economic system performance in terms of design parameters. The system is compared, through a cost analysis, to an electric heat pump (EHP) system. This paper will give the optimum system combination of solar collector area and volumetric capacity of the hot water storage tank, respectively.

Keywords: absorption chiller, solar PVT collector, solar heating and cooling, solar air-conditioning, parametric study, cost analysis

Procedia PDF Downloads 414
4914 50/50 Oil-Water Ratio Invert Emulsion Drilling Mud Using Vegetable Oil as Continuous Phase

Authors: P. C. Ihenacho, M. Burby, G. G. Nasr, G. C. Enyi

Abstract:

Formulation of a low oil-water ratio drilling mud with vegetable oil continuous phase without adversely affecting the mud rheology and stability has been a major challenge. A low oil-water ratio is beneficial in producing low fluid loss which is essential for wellbore stability. This study examined the possibility of 50/50 oil-water ratio invert emulsion drilling mud using a vegetable oil continuous phase. Jatropha oil was used as continuous phase. 12 ml of egg yolk which was separated from the albumen was added as the primary emulsifier additive. The rheological, stability and filtration properties were examined. The plastic viscosity and yield point were found to be 36cp and 17 Ib/100 ft2 respectively. The electrical stability at 48.9ºC was 353v and the 30 minutes fluid loss was 6ml. The results compared favourably with a similar formulation using 70/30 oil - water ratio giving plastic viscosity of 31cp, yield point of 17 Ib/100 ft2, electrical stability value of 480v and 12ml for the 30 minutes fluid loss. This study indicates that with a good mud composition using guided empiricism, 50/50 oil-water ratio invert emulsion drilling mud is feasible with a vegetable oil continuous phase. The choice of egg yolk as emulsifier additive is for compatibility with the vegetable oil and environmental concern. The high water content with no fluid loss additive will also minimise the cost of mud formulation.

Keywords: environmental compatibility, low cost of mud formulation, low fluid loss, wellbore stability

Procedia PDF Downloads 387
4913 Assessment Environmental and Economic of Yerba Mate as a Feed Additive on Feedlot Lamb

Authors: Danny Alexander R. Moreno, Gustavo L. Sartorello, Yuli Andrea P. Bermudez, Richard R. Lobo, Ives Claudio S. Bueno, Augusto H. Gameiro

Abstract:

Meat production is a significant sector for Brazil's economy; however, the agricultural segment has suffered censure regarding the negative impacts on the environment, which consequently results in climate change. Therefore, it is essential the implementation of nutritional strategies that can improve the environmental performance of livestock. This research aimed to estimate the environmental impact and profitability of the use of yerba mate extract (Ilex paraguariensis) as an additive in the feeding of feedlot lamb. Thirty-six castrated male lambs (average weight of 23.90 ± 3.67 kg and average age of 75 days) were randomly assigned to four experimental diets with different levels of inclusion of yerba mate extract (0, 1, 2, and 4 %) based on dry matter. The animals were confined for fifty-three days and fed with 60:40 corn silage to concentrate ratio. As an indicator of environmental impact, the carbon footprint (CF) was measured as kg of CO₂ equivalent (CO₂-eq) per kg of body weight produced (BWP). The greenhouse gas (GHG) emissions such as methane (CH₄) generated from enteric fermentation, were calculated using the sulfur hexafluoride gas tracer (SF₆) technique; while the CH₄, nitrous oxide (N₂O - emissions generated by feces and urine), and carbon dioxide (CO₂ - emissions generated by concentrate and silage processing) were estimated using the Intergovernmental Panel on Climate Change (IPCC) methodology. To estimate profitability, the gross margin was used, which is the total revenue minus the total cost; the latter is composed of the purchase of animals and food. The boundaries of this study considered only the lamb fattening system. The enteric CH₄ emission from the lamb was the largest source of on-farm GHG emissions (47%-50%), followed by CH₄ and N₂O emissions from manure (10%-20%) and CO₂ emission from the concentrate, silage, and fossil energy (17%-5%). The treatment that generated the least environmental impact was the group with 4% of yerba mate extract (YME), which showed a 3% reduction in total GHG emissions in relation to the control (1462.5 and 1505.5 kg CO₂-eq, respectively). However, the scenario with 1% YME showed an increase in emissions of 7% compared to the control group. In relation to CF, the treatment with 4% YME had the lowest value (4.1 kg CO₂-eq/kg LW) compared with the other groups. Nevertheless, although the 4% YME inclusion scenario showed the lowest CF, the gross margin decreased by 36% compared to the control group (0% YME), due to the cost of YME as a food additive. The results showed that the extract has the potential for use in reducing GHG. However, the cost of implementing this input as a mitigation strategy increased the production cost. Therefore, it is important to develop political strategies that help reduce the acquisition costs of input that contribute to the search for the environmental and economic benefit of the livestock sector.

Keywords: meat production, natural additives, profitability, sheep

Procedia PDF Downloads 132
4912 Techno-Economic Assessment of Distributed Heat Pumps Integration within a Swedish Neighborhood: A Cosimulation Approach

Authors: Monica Arnaudo, Monika Topel, Bjorn Laumert

Abstract:

Within the Swedish context, the current trend of relatively low electricity prices promotes the electrification of the energy infrastructure. The residential heating sector takes part in this transition by proposing a switch from a centralized district heating system towards a distributed heat pumps-based setting. When it comes to urban environments, two issues arise. The first, seen from an electricity-sector perspective, is related to the fact that existing networks are limited with regards to their installed capacities. Additional electric loads, such as heat pumps, can cause severe overloads on crucial network elements. The second, seen from a heating-sector perspective, has to do with the fact that the indoor comfort conditions can become difficult to handle when the operation of the heat pumps is limited by a risk of overloading on the distribution grid. Furthermore, the uncertainty of the electricity market prices in the future introduces an additional variable. This study aims at assessing the extent to which distributed heat pumps can penetrate an existing heat energy network while respecting the technical limitations of the electricity grid and the thermal comfort levels in the buildings. In order to account for the multi-disciplinary nature of this research question, a cosimulation modeling approach was adopted. In this way, each energy technology is modeled in its customized simulation environment. As part of the cosimulation methodology: a steady-state power flow analysis in pandapower was used for modeling the electrical distribution grid, a thermal balance model of a reference building was implemented in EnergyPlus to account for space heating and a fluid-cycle model of a heat pump was implemented in JModelica to account for the actual heating technology. With the models set in place, different scenarios based on forecasted electricity market prices were developed both for present and future conditions of Hammarby Sjöstad, a neighborhood located in the south-east of Stockholm (Sweden). For each scenario, the technical and the comfort conditions were assessed. Additionally, the average cost of heat generation was estimated in terms of levelized cost of heat. This indicator enables a techno-economic comparison study among the different scenarios. In order to evaluate the levelized cost of heat, a yearly performance simulation of the energy infrastructure was implemented. The scenarios related to the current electricity prices show that distributed heat pumps can replace the district heating system by covering up to 30% of the heating demand. By lowering of 2°C, the minimum accepted indoor temperature of the apartments, this level of penetration can increase up to 40%. Within the future scenarios, if the electricity prices will increase, as most likely expected within the next decade, the penetration of distributed heat pumps can be limited to 15%. In terms of levelized cost of heat, a residential heat pump technology becomes competitive only within a scenario of decreasing electricity prices. In this case, a district heating system is characterized by an average cost of heat generation 7% higher compared to a distributed heat pumps option.

Keywords: cosimulation, distributed heat pumps, district heating, electrical distribution grid, integrated energy systems

Procedia PDF Downloads 145
4911 Comparative Analysis of Fused Deposition Modeling and Binding-Jet 3D Printing Technologies

Authors: Mohd Javaid, Shahbaz Khan, Abid Haleem

Abstract:

Purpose: Large numbers of 3D printing technologies are now available for sophisticated applications in different fields. Additive manufacturing has established its dominance in design, development, and customisation of the product. In the era of developing technologies, there is a need to identify the appropriate technology for different application. In order to fulfil this need, two widely used printing technologies such as Fused Deposition Modeling (FDM), and Binding-Jet 3D Printing are compared for effective utilisation in the current scenario for different applications. Methodology: Systematic literature review conducted for both technologies with applications and associated factors enabling for the same. Appropriate MCDM tool is used to compare critical factors for both the technologies. Findings: Both technologies have their potential and capabilities to provide better direction to the industry. Additionally, this paper is helpful to develop a decision support system for the proper selection of technologies according to their continuum of applications and associated research and development capability. The vital issue is raw materials, and research-based material development is key to the sustainability of the developed technologies. FDM is a low-cost technology which provides high strength product as compared to binding jet technology. Researcher and companies can take benefits of this study to achieve the required applications in lesser resources. Limitations: Study has undertaken the comparison with the opinion of experts, which may not always be free from bias, and some own limitations of each technology. Originality: Comparison between these technologies will help to identify best-suited technology as per the customer requirements. It also provides development in this different field as per their extensive capability where these technologies can be successfully adopted. Conclusion: FDM and binding jet technology play an active role in industrial development. These help to assist the customisation and production of personalised parts cost-effectively. So, there is a need to understand how these technologies can provide these developments rapidly. These technologies help in easy changes or in making revised versions of the product, which is not easily possible in the conventional manufacturing system. High machine cost, the requirement of skilled human resources, low surface finish, and mechanical strength of product and material changing option is the main limitation of this technology. However, these limitations vary from technology to technology. In the future, these technologies are to be commercially viable for efficient usage in direct manufacturing of varied parts.

Keywords: 3D printing, comparison, fused deposition modeling, FDM, binding jet technology

Procedia PDF Downloads 100
4910 Students’ Perception of Effort and Emotional Costs in Chemistry Courses

Authors: Guizella Rocabado, Cassidy Wilkes

Abstract:

It is well known that chemistry is one of the most feared courses in college. Although many students enjoy learning about science, most of them perceive that chemistry is “too difficult”. These perceptions of chemistry result in many students not considering Science, Technology, Engineering, and Mathematics (STEM) majors because they require chemistry courses. Ultimately, these perceptions are also thought to be related to high attrition rates of students who begin STEM majors but do not persist. Students perceived costs of a chemistry class can be many, such as task effort, loss of valued alternatives, emotional, and others. These costs might be overcome by students’ interests and goals, yet the level of perceived costs might have a lasting impact on the students’ overall perception of chemistry and their desire to pursue chemistry and other STEM careers in the future. In this mixed methods study, we investigated task effort and emotional cost, as well as a mastery or performance goal orientation, and the impact these constructs may have on achievement in general chemistry classrooms. Utilizing cluster analysis as well as student interviews, we investigated students’ profiles of perceived cost and goal orientation as it relates to their final grades. Our results show that students who are well prepared for general chemistry, such as those who have taken chemistry in high school, display less negative perceived costs and thus believe they can master the material more fully. Other interesting results have also emerged from this research, which has the potential to have an impact on future instruction of these courses.

Keywords: chemistry education, motivation, affect, perceived costs, goal orientations

Procedia PDF Downloads 82
4909 Numerical Simulation of Different Configurations for a Combined Gasification/Carbonization Reactors

Authors: Mahmoud Amer, Ibrahim El-Sharkawy, Shinichi Ookawara, Ahmed Elwardany

Abstract:

Gasification and carbonization are two of the most common ways for biomass utilization. Both processes are using part of the waste to be accomplished, either by incomplete combustion or for heating for both gasification and carbonization, respectively. The focus of this paper is to minimize the part of the waste that is used for heating biomass for gasification and carbonization. This will occur by combining both gasifiers and carbonization reactors in a single unit to utilize the heat in the product biogas to heating up the wastes in the carbonization reactors. Three different designs are proposed for the combined gasification/carbonization (CGC) reactor. These include a parallel combination of two gasifiers and carbonized syngas, carbonizer and combustion chamber, and one gasifier, carbonizer, and combustion chamber. They are tested numerically using ANSYS Fluent Computational Fluid Dynamics to ensure homogeneity of temperature distribution inside the carbonization part of the CGC reactor. 2D simulations are performed for the three cases after performing both mesh-size and time-step independent solutions. The carbonization part is common among the three different cases, and the difference among them is how this carbonization reactor is heated. The simulation results showed that the first design could provide only partial homogeneous temperature distribution, not across the whole reactor. This means that the produced carbonized biomass will be reduced as it will only fill a specified height of the reactor. To keep the carbonized product production high, a series combination is proposed. This series configuration resulted in a uniform temperature distribution across the whole reactor as it has only one source for heat with no temperature distribution on any surface of the carbonization section. The simulations provided a satisfactory result that either the first parallel combination of gasifier and carbonization reactor could be used with a reduced carbonized amount or a series configuration to keep the production rate high.

Keywords: numerical simulation, carbonization, gasification, biomass, reactor

Procedia PDF Downloads 97
4908 Predictive Maintenance of Industrial Shredders: Efficient Operation through Real-Time Monitoring Using Statistical Machine Learning

Authors: Federico Pittino, Thomas Arnold

Abstract:

The shredding of waste materials is a key step in the recycling process towards the circular economy. Industrial shredders for waste processing operate in very harsh operating conditions, leading to the need for frequent maintenance of critical components. Maintenance optimization is particularly important also to increase the machine’s efficiency, thereby reducing the operational costs. In this work, a monitoring system has been developed and deployed on an industrial shredder located at a waste recycling plant in Austria. The machine has been monitored for one year, and methods for predictive maintenance have been developed for two key components: the cutting knives and the drive belt. The large amount of collected data is leveraged by statistical machine learning techniques, thereby not requiring very detailed knowledge of the machine or its live operating conditions. The results show that, despite the wide range of operating conditions, a reliable estimate of the optimal time for maintenance can be derived. Moreover, the trade-off between the cost of maintenance and the increase in power consumption due to the wear state of the monitored components of the machine is investigated. This work proves the benefits of real-time monitoring system for the efficient operation of industrial shredders.

Keywords: predictive maintenance, circular economy, industrial shredder, cost optimization, statistical machine learning

Procedia PDF Downloads 118
4907 Out of Pocket Costs for Patients with Tuberculosis in Colombia: Evidence from Three Metropolitan Areas

Authors: Jose Hernandez, Lina Martínez, Gustavo Gonzalez, Carlos Lázaro, Diana Castrillon, Jonathan Cardona, Laura Mejía, Yina Sanchez, Luisa Ochoa, Evert Jimenez

Abstract:

Objectives: Economic analyses of tuberculosis control interventions are usually focused on the payer’s perspective. To assess the overall economic impact of the disease, out-of-pocket and indirect costs are also required. This research is aimed to estimate overall economic impact under DOTS-strategy (Directly Observed Therapy Short Course). Methods: A cross-sectional survey of 91 adult tuberculosis patients in treatment for at least two months was conducted from the society perspective. A standardized questionnaire was used in three different cities of Colombia: Medellin (poverty is 17.7%), Monteria (poverty is 36.9%) and Quibdó (poverty is 51.2%). Costs were converted to 2013 USD and categorized into two periods: diagnostics phase and treatment. Results: The median cost during diagnostics was 13$ (±SD 9.5). The median monthly patient out-of-pocket costs during treatment were 32$ (±SD 6.8), equivalent to 17% of patient’s median monthly income, estimated in 186$ (±SD 23). Costs recorded in Medellin were 47$ in Monteria was 18$ and in Quibdó was 13$. Conclusion: Patient costs under DOTS strategy are high even when services are provided free of charge. The creation or strengthening of community-based treatment supervisors could greatly impact costs of tuberculosis and lower drop-outs.

Keywords: tuberculosis, costs and cost analysis, health promotion, Colombia

Procedia PDF Downloads 361
4906 Analysis of Energy Efficiency Behavior with the Use of Train Dynamics Simulator and Statistical Tools: Case Study of Vitoria Minas Railway, Brazil

Authors: Eric Wilson Santos Cabral, Marta Monteiro Da Costa Cruz, Fabio Luis Maciel Machado, Henrique Andrade, Rodrigo Pirola Pestana, Vivian Andrea Parreira

Abstract:

The large variation in the price of diesel in Brazil directly affects the variable cost of companies operating in the transportation sector. In rail transport, the great challenge is to overcome the annual budget, cargo and ore transported with cost reduction in relation to previous years, becoming more efficient every year. Some effective measures are necessary to achieve the reduction of the liter ratio consumed by KTKB (Gross Ton per Kilometer multiplied by thousand). This acronym represents the indicator of energy efficiency of some railroads in the world. This study is divided into two parts: the first, to identify using statistical tools, part of the controlled variables in the railways, which have a correlation with the energy efficiency indicator, seeking to aid decision-making. The second, with the use of the train dynamics simulator, within scenarios defined in the operational reality of a railroad, seeks to optimize the train formations and the train stop model for the change of train drivers. With the completion of the study, companies in the rail sector are expected to be able to reduce some of their transportation costs.

Keywords: railway transport, railway simulation, energy efficiency, fuel consumption

Procedia PDF Downloads 331
4905 Reducing the Computational Cost of a Two-way Coupling CFD-FEA Model via a Multi-scale Approach for Fire Determination

Authors: Daniel Martin Fellows, Sean P. Walton, Jennifer Thompson, Oubay Hassan, Kevin Tinkham, Ella Quigley

Abstract:

Structural integrity for cladding products is a key performance parameter, especially concerning fire performance. Cladding products such as PIR-based sandwich panels are tested rigorously, in line with industrial standards. Physical fire tests are necessary to ensure the customer's safety but can give little information about critical behaviours that can help develop new materials. Numerical modelling is a tool that can help investigate a fire's behaviour further by replicating the fire test. However, fire is an interdisciplinary problem as it is a chemical reaction that behaves fluidly and impacts structural integrity. An analysis using Computational Fluid Dynamics (CFD) and Finite Element Analysis (FEA) is needed to capture all aspects of a fire performance test. One method is a two-way coupling analysis that imports the updated changes in thermal data, due to the fire's behaviour, to the FEA solver in a series of iterations. In light of our recent work with Tata Steel U.K using a two-way coupling methodology to determine the fire performance, it has been shown that a program called FDS-2-Abaqus can make predictions of a BS 476 -22 furnace test with a degree of accuracy. The test demonstrated the fire performance of Tata Steel U.K Trisomet product, a Polyisocyanurate (PIR) based sandwich panel used for cladding. Previous works demonstrated the limitations of the current version of the program, the main limitation being the computational cost of modelling three Trisomet panels, totalling an area of 9 . The computational cost increases substantially, with the intention to scale up to an LPS 1181-1 test, which includes a total panel surface area of 200 .The FDS-2-Abaqus program is developed further within this paper to overcome this obstacle and better accommodate Tata Steel U.K PIR sandwich panels. The new developments aim to reduce the computational cost and error margin compared to experimental data. One avenue explored is a multi-scale approach in the form of Reduced Order Modeling (ROM). The approach allows the user to include refined details of the sandwich panels, such as the overlapping joints, without a computationally costly mesh size.Comparative studies will be made between the new implementations and the previous study completed using the original FDS-2-ABAQUS program. Validation of the study will come from physical experiments in line with governing body standards such as BS 476 -22 and LPS 1181-1. The physical experimental data includes the panels' gas and surface temperatures and mechanical deformation. Conclusions are drawn, noting the new implementations' impact factors and discussing the reasonability for scaling up further to a whole warehouse.

Keywords: fire testing, numerical coupling, sandwich panels, thermo fluids

Procedia PDF Downloads 70
4904 Sustainable Use of Agricultural Waste to Enhance Food Security and Conserve the Environment

Authors: M. M. Tawfik, Ezzat M. Abd El Lateef, B. B. Mekki, Amany A. Bahr, Magda H. Mohamed, Gehan S. Bakhoom

Abstract:

The rapid increase in the world’s population coupled by decrease the arable land per capita has resulted into an increased demand for food which has in turn led to the production of large amounts of agricultural wastes, both at the farmer, municipality and city levels. Agricultural wastes can be a valuable resource for improving food security. Unfortunately, agricultural wastes are likely to cause pollution to the environment or even harm to human health. This calls for increased public awareness on the benefits and potential hazards of agricultural wastes, especially in developing countries. Agricultural wastes (residual stalks, straw, leaves, roots, husks, shells etcetera) and animal waste (manures) are widely available, renewable and virtually free, hence they can be an important resource. They can be converted into heat, steam, charcoal, methanol, ethanol, bio diesel as well as raw materials (animal feed, composting, energy and biogas construction etcetera). agricultural wastes are likely to cause pollution to the environment or even harm to human health, if it is not used in a sustainable manner. Organic wastes could be considered an important source of biofertilizer for enhancing food security in the small holder farming communities that would not afford use of expensive inorganic fertilizers. Moreover, these organic wastes contain high levels of nitrogen, phosphorus, potassium, and organic matter important for improving nutrient status of soils in urban agriculture. Organic compost leading to improved crop yields and its nutritional values as compared with inorganic fertilization. This paper briefly reviews how agricultural wastes can be used to enhance food security and conserve the environment.

Keywords: agricultural waste, organic compost, environment, valuable resources

Procedia PDF Downloads 514
4903 Standalone Docking Station with Combined Charging Methods for Agricultural Mobile Robots

Authors: Leonor Varandas, Pedro D. Gaspar, Martim L. Aguiar

Abstract:

One of the biggest concerns in the field of agriculture is around the energy efficiency of robots that will perform agriculture’s activity and their charging methods. In this paper, two different charging methods for agricultural standalone docking stations are shown that will take into account various variants as field size and its irregularities, work’s nature to which the robot will perform, deadlines that have to be respected, among others. Its features also are dependent on the orchard, season, battery type and its technical specifications and cost. First charging base method focuses on wireless charging, presenting more benefits for small field. The second charging base method relies on battery replacement being more suitable for large fields, thus avoiding the robot stop for recharge. Existing many methods to charge a battery, the CC CV was considered the most appropriate for either simplicity or effectiveness. The choice of the battery for agricultural purposes is if most importance. While the most common battery used is Li-ion battery, this study also discusses the use of graphene-based new type of batteries with 45% over capacity to the Li-ion one. A Battery Management Systems (BMS) is applied for battery balancing. All these approaches combined showed to be a promising method to improve a lot of technical agricultural work, not just in terms of plantation and harvesting but also about every technique to prevent harmful events like plagues and weeds or even to reduce crop time and cost.

Keywords: agricultural mobile robot, charging methods, battery replacement method, wireless charging method

Procedia PDF Downloads 144
4902 Paying Less and Getting More: Evidence on the Effect of Corporate Purpose from Two Natural Field Experiments

Authors: Nikolai Brosch, Alwine Mohnen

Abstract:

Academics and business leaders increasingly call for a (re)definition of a corporate purpose beyond profit-maximization to contribute to the welfare of society. This study investigates the effect of communicating such a pro-social corporate purpose on three employee-level outcomes that constitute major cost components for most organizations: workers reservation wage, work quality, and work misbehavior. To provide causal evidence, two natural field experiments were conducted with almost 2,000 workers recruited from different online labor marketplaces. Workers were randomly assigned to treatments manipulating whether or not they received information about the employer’s corporate purpose and subsequently performed a short, real-effort task for payment. The main findings in both experiments show that receiving information about an employer’s pro-social corporate purpose causes workers to accept lower wages (9% lower in the first experiment and 28% lower in the second experiment) for the same job. Workers that personally assess high importance to organizations having a pro-social purpose are most responsive. At the same time, sacrificing wage for a corporate purpose comes at no cost of quality and even decreases the likelihood of engaging in work misbehavior. In a broader context, the results provide some evidence that the (re)definition of corporate purpose in commercial organizations is not ultimately at odds with creating profits.

Keywords: corporate purpose, natural field experiment, reservation wage, work misbehavior, work quality

Procedia PDF Downloads 227
4901 Isolation and Screening of Fungal Strains for β-Galactosidase Production

Authors: Parmjit S. Panesar, Rupinder Kaur, Ram S. Singh

Abstract:

Enzymes are the biocatalysts which catalyze the biochemical processes and thus have a wide variety of applications in the industrial sector. β-Galactosidase (E.C. 3.2.1.23) also known as lactase, is one of the prime enzymes, which has significant potential in the dairy and food processing industries. It has the capability to catalyze both the hydrolytic reaction for the production of lactose hydrolyzed milk and transgalactosylation reaction for the synthesis of prebiotics such as lactulose and galactooligosaccharides. These prebiotics have various nutritional and technological benefits. Although, the enzyme is naturally present in almonds, peaches, apricots and other variety of fruits and animals, the extraction of enzyme from these sources increases the cost of enzyme. Therefore, focus has been shifted towards the production of low cost enzyme from the microorganisms such as bacteria, yeast and fungi. As compared to yeast and bacteria, fungal β-galactosidase is generally preferred as being extracellular and thermostable in nature. Keeping the above in view, the present study was carried out for the isolation of the β-galactosidase producing fungal strain from the food as well as the agricultural wastes. A total of more than 100 fungal cultures were examined for their potential in enzyme production. All the fungal strains were screened using X-gal and IPTG as inducers in the modified Czapek Dox Agar medium. Among the various isolated fungal strains, the strain exhibiting the highest enzyme activity was chosen for further phenotypic and genotypic characterization. The strain was identified as Rhizomucor pusillus on the basis of 5.8s RNA gene sequencing data.

Keywords: beta-galactosidase, enzyme, fungal, isolation

Procedia PDF Downloads 248
4900 Fluoride Removal from Groundwater in the East Nile Area (Sudan) Using Locally Available Charcoal

Authors: Motwkel M. Alhaj, Bashir M. Elhassan

Abstract:

The East Nile area is located in Khartoum state. The main source of drinking water in the East Nile Area (Sudan) is groundwater. However, fluoride concentration in the water is more than the maximum allowable dose, which is 1.5 mg/l. This study aims to demonstrate and innovative, affordable, and efficient filter to remove fluoride from drinking water. Many researchers have found that aluminum oxide-coated adsorbent is the most affordable technology for fluoride removal. However, adsorption is pH-dependent, and the water pH in the East Nile area is relatively high (around 8), which is hindering the adsorption process. Locally available charcoal was crushed, sieved, and coated with aluminum oxide. Then, different coating configurations were tested in order to produce an adsorbent with a high pH point of zero charge pH PZC in order to overcome the effect of high pH of water. Moreover, different methods were used to characterize the adsorbent, including: Scanning Electron Microscope (SEM), Energy Dispersive X-Ray Spectroscopy (EDX), Brunauer - Emmett - Teller (BET) method, and pH point of zero charge pH PZC. The produced adsorbent has pH PZC of 8.5, which is essential in enhancing the fluoride adsorption process. A pilot household fluoride filter was also designed and installed in a house that has water with 4.34 mg/l F- and pH of 8.4. The filter was operated at a flow rate 250 cm³/min. The total cost of treating one cubic meter was about 0.63$, while the cost for the same water before adsorbent coating modification was 2.33$⁄cm³.

Keywords: water treatment, fluoride, adsorption, charcoal, Sudan

Procedia PDF Downloads 110
4899 The Importance of Downstream Supply Chain in Supply Chain Risk Management: Multi-Objective Optimization

Authors: Zohreh Khojasteh-Ghamari, Takashi Irohara

Abstract:

One of the efficient ways in supply chain risk management is avoiding the interruption in Supply Chain (SC) before it occurs. Although the majority of the organizations focus on their first-tier suppliers to avoid risk in the SC, studies show that in only 60 percent of the disruption cases the reason is first tier suppliers. In the 40 percent of the SC disruptions, the reason is downstream SC, which is the second tier and lower. Due to the increasing complexity and interrelation of modern supply chains, the SC elements have become difficult to trace. Moreover, studies show that there is a vital need to better understand the integration of risk and visibility, especially in the context of multiple objectives. In this study, we propose a multi-objective programming model to avoid disruption in SC. The objective of this study is evaluating the effect of downstream SCV on managing supply chain risk. We propose a multi-objective mathematical programming model with the objective functions of minimizing the total cost and maximizing the downstream supply chain visibility (SCV). The decision variable is supplier selection. We assume there are several manufacturers and several candidate suppliers. For each manufacturer, our model proposes the best suppliers with the lowest cost and maximum visibility in downstream supply chain. We examine the applicability of the model by numerical examples. We also define several scenarios for datasets and observe the tendency. The results show that minimum visibility in downstream SC is needed to have a safe SC network.

Keywords: downstream supply chain, optimization, supply chain risk, supply chain visibility

Procedia PDF Downloads 241
4898 Electrostatic Solitary Waves in Degenerate Relativistic Quantum Plasmas

Authors: Sharmin Sultana, Reinhard Schlickeiser

Abstract:

A degenerate relativistic quantum plasma (DRQP) system (containing relativistically degenerate electrons, degenerate/non-degenerate light nuclei, and non-degenerate heavy nuclei) is considered to investigate the propagation characteristics of electrostatic solitary waves (in the ionic scale length) theoretically and numerically. The ion-acoustic solitons are found to be associated with the modified ion-acoustic waves (MIAWs) in which inertia (restoring force) is provided by mass density of the light or heavy nuclei (degenerate pressure of the cold electrons). A mechanical-motion analog (Sagdeev-type) pseudo-potential approach is adopted to study the properties of large amplitude solitary waves. The basic properties of the large amplitude MIAWs and their existence domain in terms of soliton speed (Mach number) are examined. On the other hand, a multi-scale perturbation approach, leading to an evolution equation for the envelope dynamics, is adopted to derive the cubic nonlinear Schrödinger equation (NLSE). The criteria for the occurrence of modulational instability (MI) of the MIAWs are analyzed via the nonlinear dispersion relation of the NLSE. The possibility for the formation of highly energetic localized modes (e.g. peregrine solitons, rogue waves, etc.) is predicted in such DRQP medium. Peregrine solitons or rogue waves with amplitudes of several times of the background are observed to form in DRQP. The basic features of these modulated waves (e.g. envelope solitons, peregrine solitons, and rogue waves), which are found to form in DRQP, and their MI criteria (on the basis of different intrinsic plasma parameters), are investigated. It is emphasized that our results should be useful in understanding the propagation characteristics of localized disturbances and the modulation dynamics of envelope solitons, and their instability criteria in astrophysical DRQP system (e.g. white dwarfs, neutron stars, etc., where matters under extreme conditions are assumed to exist) and also in ultra-high density experimental plasmas.

Keywords: degenerate plasma, envelope solitons, modified ion-acoustic waves, modulational instability, rogue waves

Procedia PDF Downloads 197
4897 How to Improve the Environmental Performance in a HEI in Mexico, an EEA Adaptation

Authors: Stephanie Aguirre Moreno, Jesús Everardo Olguín Tiznado, Claudia Camargo Wilson, Juan Andrés López Barreras

Abstract:

This research work presents a proposal to evaluate the environmental performance of a Higher Education Institution (HEI) in Mexico in order to minimize their environmental impact. Given that public education has limited financial resources, it is necessary to conduct studies that support priorities in decision-making situations and thus obtain the best cost-benefit ratio of continuous improvement programs as part of the environmental management system implemented. The methodology employed, adapted from the Environmental Effect Analysis (EEA), weighs the environmental aspects identified in the environmental diagnosis by two characteristics. Number one, environmental priority through the perception of the stakeholders, compliance of legal requirements, and environmental impact of operations. Number two, the possibility of improvement, which depends of factors such as the exchange rate that will be made, the level of investment and the return time of it. The highest environmental priorities, or hot spots, identified in this evaluation were: electricity consumption, water consumption and recycling, and disposal of municipal solid waste. However, the possibility of improvement for the disposal of municipal solid waste is higher, followed by water consumption and recycling, in spite of having an equal possibility of improvement to the energy consumption, time of return and cost-benefit is much greater.

Keywords: environmental performance, environmental priority, possibility of improvement, continuous improvement programs

Procedia PDF Downloads 489
4896 MAOD Is Estimated by Sum of Contributions

Authors: David W. Hill, Linda W. Glass, Jakob L. Vingren

Abstract:

Maximal accumulated oxygen deficit (MAOD), the gold standard measure of anaerobic capacity, is the difference between the oxygen cost of exhaustive severe intensity exercise and the accumulated oxygen consumption (O2; mL·kg–1). In theory, MAOD can be estimated as the sum of independent estimates of the phosphocreatine and glycolysis contributions, which we refer to as PCr+glycolysis. Purpose: The purpose was to test the hypothesis that PCr+glycolysis provides a valid measure of anaerobic capacity in cycling and running. Methods: The participants were 27 women (mean ± SD, age 22 ±1 y, height 165 ± 7 cm, weight 63.4 ± 9.7 kg) and 25 men (age 22 ± 1 y, height 179 ± 6 cm, weight 80.8 ± 14.8 kg). They performed two exhaustive cycling and running tests, at speeds and work rates that were tolerable for ~5 min. The rate of oxygen consumption (VO2; mL·kg–1·min–1) was measured in warmups, in the tests, and during 7 min of recovery. Fingerprick blood samples obtained after exercise were analysed to determine peak blood lactate concentration (PeakLac). The VO2 response in exercise was fitted to a model, with a fast ‘primary’ phase followed by a delayed ‘slow’ component, from which was calculated the accumulated O2 and the excess O2 attributable to the slow component. The VO2 response in recovery was fitted to a model with a fast phase and slow component, sharing a common time delay. Oxygen demand (in mL·kg–1·min–1) was determined by extrapolation from steady-state VO2 in warmups; the total oxygen cost (in mL·kg–1) was determined by multiplying this demand by time to exhaustion and adding the excess O2; then, MAOD was calculated as total oxygen cost minus accumulated O2. The phosphocreatine contribution (area under the fast phase of the post-exercise VO2) and the glycolytic contribution (converted from PeakLac) were summed to give PCr+glycolysis. There was not an interaction effect involving sex, so values for anaerobic capacity were examined using a two-way ANOVA, with repeated measures across method (PCr+glycolysis vs MAOD) and mode (cycling vs running). Results: There was a significant effect only for exercise mode. There was no difference between MAOD and PCr+glycolysis: values were 59 ± 6 mL·kg–1 and 61 ± 8 mL·kg–1 in cycling and 78 ± 7 mL·kg–1 and 75 ± 8 mL·kg–1 in running. Discussion: PCr+glycolysis is a valid measure of anaerobic capacity in cycling and running, and it is as valid for women as for men.

Keywords: alactic, anaerobic, cycling, ergometer, glycolysis, lactic, lactate, oxygen deficit, phosphocreatine, running, treadmill

Procedia PDF Downloads 131
4895 Mixotrophic Cultivation of Microalgae as a Feasible Strategy for Carotenoid Production

Authors: Jian Li

Abstract:

Carotenoids area group of metabolites in mostly photosynthetic organisms such as plants and microalgae and have wide applications in cosmetics, food, feed, and health industries. Although phototrophic cultivation of microalgae has been developed to produce some carotenoids for decades, most carotenoids are currently synthesized chemically at industrial scales because of affordable production costs. Chemical carotenoids are regarded not as safe for human beings as natural carotenoids and are restricted only for animal feed markets, and the industries call for inexpensive sources of natural products. Microalgae grow much quicker in mixotrophy than in phototrophy, and thus mixotrophic cultivation processes have great potential to reduce the production cost of carotenoids from microalgae. However, much more expensive photobioreactor systems and more strictly controlled sterile processes are needed to avoid contamination by heterotrophic organisms during mixotrophic cultivation processes, which makes mixotrophy, in fact, much more expensive than phototrophic cultivation. Recently technical breakthroughs have been reported to overcome contamination problems in photobioreactor systems traditionally used for phototrophic cultivation, and a much lower process cost of mixotrophic cultivation than that of phototrophic cultivation might be achieved for carotenoid production. These reviews intend to summarize recent technical advancements in mixotrophic cultivation of microalgae, to evaluate the economic viability of carotenoid production from mixotrophically cultivated microalgae, and to prospect mixotrophy as a strategy to produce a variety of carotenoids for industrial applications.

Keywords: microalgae, carotenoid, mixotrophy, biotechnology

Procedia PDF Downloads 150
4894 MLProxy: SLA-Aware Reverse Proxy for Machine Learning Inference Serving on Serverless Computing Platforms

Authors: Nima Mahmoudi, Hamzeh Khazaei

Abstract:

Serving machine learning inference workloads on the cloud is still a challenging task at the production level. The optimal configuration of the inference workload to meet SLA requirements while optimizing the infrastructure costs is highly complicated due to the complex interaction between batch configuration, resource configurations, and variable arrival process. Serverless computing has emerged in recent years to automate most infrastructure management tasks. Workload batching has revealed the potential to improve the response time and cost-effectiveness of machine learning serving workloads. However, it has not yet been supported out of the box by serverless computing platforms. Our experiments have shown that for various machine learning workloads, batching can hugely improve the system’s efficiency by reducing the processing overhead per request. In this work, we present MLProxy, an adaptive reverse proxy to support efficient machine learning serving workloads on serverless computing systems. MLProxy supports adaptive batching to ensure SLA compliance while optimizing serverless costs. We performed rigorous experiments on Knative to demonstrate the effectiveness of MLProxy. We showed that MLProxy could reduce the cost of serverless deployment by up to 92% while reducing SLA violations by up to 99% that can be generalized across state-of-the-art model serving frameworks.

Keywords: serverless computing, machine learning, inference serving, Knative, google cloud run, optimization

Procedia PDF Downloads 174