Search results for: large scale optimization

Authors: Tsegay Giday Woldu, Haibin Zhang, Xin Zhang, Yemane Hailu Fissuh

Abstract:

It is well known that nonlinear conjugate gradient method is one of the widely used first order methods to solve large scale unconstrained smooth optimization problems. Because of the low memory requirement, attractive theoretical features, practical computational efficiency and nice convergence properties, nonlinear conjugate gradient methods have a special role for solving large scale unconstrained optimization problems. Large scale optimization problems are with important applications in practical and scientific world. However, nonlinear conjugate gradient methods have restricted information about the curvature of the objective function and they are likely less efficient and robust compared to some second order algorithms. To overcome these drawbacks, the new modified nonlinear conjugate gradient method is presented. The noticeable features of our work are that the new search direction possesses the sufficient descent property independent of any line search and it belongs to a trust region. Under mild assumptions and standard Wolfe line search technique, the global convergence property of the proposed algorithm is established. Furthermore, to test the practical computational performance of our new algorithm, numerical experiments are provided and implemented on the set of some large dimensional unconstrained problems. The numerical results show that the proposed algorithm is an efficient and robust compared with other similar algorithms.

Keywords: conjugate gradient method, global convergence, large scale optimization, sufficient descent property

Procedia PDF Downloads 163

Authors: Mohammed Belloufi, Rachid Benzine, Badreddine Sellami

Abstract:

Conjugate gradient methods is useful for solving large scale optimization problems in scientific and engineering computation, characterized by the simplicity of their iteration and their low memory requirements. It is well known that the search direction plays a main role in the line search method. In this paper, we propose a search direction with the Wolfe line search technique for solving unconstrained optimization problems. Under the above line searches and some assumptions, the global convergence properties of the given methods are discussed. Numerical results and comparisons with other CG methods are given.

Keywords: unconstrained optimization, conjugate gradient method, strong Wolfe line search, global convergence

Procedia PDF Downloads 384

Authors: Henrik Prinzhorn, Peter Nyhuis, Johannes Wagner, Peter Burggräf, Torben Schmitz, Christina Reuter

Abstract:

Assembling large-scale products, such as airplanes, locomotives, or wind turbines, involves frequent process interruptions induced by e.g. delayed material deliveries or missing availability of resources. This leads to a negative impact on the logistical performance of a producer of xxl-products. In industrial practice, in case of interruptions, the identification, evaluation and eventually the selection of an alternative order of assembly activities (‘assembly alternative’) leads to an enormous challenge, especially if an optimized logistical decision should be reached. Therefore, in this paper, an innovative, optimization model for the identification of assembly alternatives that addresses the given problem is presented. It describes make-to-order, large-scale product assembly processes as a resource constrained project scheduling (RCPS) problem which follows given restrictions in practice. For the evaluation of the assembly alternative, a cost-based definition of the logistical objectives (delivery reliability, inventory, make-span and workload) is presented.

Keywords: assembly scheduling, large-scale products, make-to-order, optimization, rescheduling

Procedia PDF Downloads 421

Authors: F. Bellali, M. Kharroubi, Y. Rady, N. Bourhim

Abstract:

In Morocco, fish processing industry is an important source income for a large amount of by-products including skins, bones, heads, guts, and scales. Those underutilized resources particularly scales contain a large amount of proteins and calcium. Sardina plichardus scales from resulting from the transformation operation have the potential to be used as raw material for the collagen production. Taking into account this strong expectation of the regional fish industry, scales sardine upgrading is well justified. In addition, political and societal demands for sustainability and environment-friendly industrial production systems, coupled with the depletion of fish resources, drive this trend forward. Therefore, fish scale used as a potential source to isolate collagen has a wide large of applications in food, cosmetic, and biomedical industry. The main aim of this study is to isolate and characterize the acid solubilize collagen from sardine fish scale, Sardina pilchardus. Experimental design methodology was adopted in collagen processing for extracting optimization. The first stage of this work is to investigate the optimization conditions of the sardine scale deproteinization on using response surface methodology (RSM). The second part focus on the demineralization with HCl solution or EDTA. And the last one is to establish the optimum condition for the isolation of collagen from fish scale by solvent extraction. The advancement from lab scale to pilot scale is a critical stage in the technological development. In this study, the optimal condition for the deproteinization which was validated at laboratory scale was employed in the pilot scale procedure. The deproteinization of fish scale was then demonstrated on a pilot scale (2Kg scales, 20l NaOH), resulting in protein content (0,2mg/ml) and hydroxyproline content (2,11mg/l). These results indicated that the pilot-scale showed similar performances to those of lab-scale one.

Keywords: deproteinization, pilot scale, scale, sardine pilchardus

Procedia PDF Downloads 415

Authors: Wen-Bao Qiao, Jean-Charles Créput

Abstract:

To accelerate the solution for large scale traveling salesman problems (TSP), a parallel 2-opt local search algorithm with simple implementation based on Graphics Processing Unit (GPU) is presented and tested in this paper. The parallel scheme is based on technique of data decomposition by dynamically assigning multiple K processors on the integral tour to treat K edges’ 2-opt local optimization simultaneously on independent sub-tours, where K can be user-defined or have a function relationship with input size N. We implement this algorithm with doubly linked list on GPU. The implementation only requires O(N) memory. We compare this parallel 2-opt local optimization against sequential exhaustive 2-opt search along integral tour on TSP instances from TSPLIB with more than 10000 cities.

Keywords: parallel 2-opt, double links, large scale TSP, GPU

Procedia PDF Downloads 592

Authors: Andrey A. Chernousov, Ben Y. B. Chan

Abstract:

The thermal behavior of a large-scale, phase change material (PCM) enhanced building envelope system was studied in regard to the need for pre-fabricated construction in subtropical regions. The proposed large-scale envelope consists of a reinforced aluminum skin, insulation core, phase change material and reinforced gypsum board. The PCM impact on an energy efficiency of an enveloped room was resolved by validation of the Energy Plus numerical scheme and optimization of a smart material location in the core. The PCM location was optimized by a minimization method of a cooling energy demand. It has been shown that there is good agreement between the test and simulation results. The optimal location of the PCM layer in Hong Kong summer conditions has been then recomputed for core thicknesses of 40, 60 and 80 mm. A non-dimensional value of the optimal PCM location was obtained to be same for all the studied cases and the considered external and internal conditions.

Keywords: thermal performance, phase change material, energy efficiency, PCM optimization

Procedia PDF Downloads 372

Authors: F. Bellali, M. Kharroubi, M. Loutfi, N.Bourhim

Abstract:

In Morocco, fish processing industry is an important source income for a large amount of byproducts including skins, bones, heads, guts and scales. Those underutilized resources particularly scales contain a large amount of proteins and calcium. Scales from Sardina plichardus resulting from the transformation operation have the potential to be used as raw material for the collagen production. Taking into account this strong expectation of the regional fish industry, scales sardine upgrading is well justified. In addition, political and societal demands for sustainability and environment-friendly industrial production systems, coupled with the depletion of fish resources, drive this trend forward. Therefore, fish scale used as a potential source to isolate collagen has a wide large of applications in food, cosmetic and bio medical industry. The main aim of this study is to isolate and characterize the acid solubilize collagen from sardine fish scale, Sardina pilchardus. Experimental design methodology was adopted in collagen processing for extracting optimization. The first stage of this work is to investigate the optimization conditions of the sardine scale deproteinization on using response surface methodology (RSM). The second part focus on the demineralization with HCl solution or EDTA. Moreover, the last one is to establish the optimum condition for the isolation of collagen from fish scale by solvent extraction. The basic principle of RSM is to determinate model equations that describe interrelations between the independent variables and the dependent variables.

Keywords: Sardina pilchardus, scales, valorization, collagen extraction, response surface methodology

Procedia PDF Downloads 382

Authors: Takahiro Hino, Michiharu Maeda

Abstract:

Particle swarm optimization (PSO) is an optimization approach that achieves the social model of bird flocking and fish schooling. PSO works in continuous space and can solve continuous optimization problem with high quality. Set-based particle swarm optimization (SPSO) functions in discrete space by using a set. SPSO can solve combinatorial optimization problem with high quality and is successful to apply to the large-scale problem. In this paper, we present an algorithm of SPSO with status memory to decide the position based on the previous position for solving traveling salesman problem (TSP). In order to show the effectiveness of our approach. We examine SPSOSM for TSP compared to the existing algorithms.

Keywords: combinatorial optimization problems, particle swarm optimization, set-based particle swarm optimization, traveling salesman problem

Procedia PDF Downloads 508

Authors: Kweon Oh-Sang, Kim Heung-Youl

Abstract:

Estimating the engineering properties of fires is important to be prepared for the complex and various fire risks of large-scale structures such as super-tall buildings, large stadiums, and multi-purpose structures. In this study, a mock-up of a compartment which was 2.4(L) x 3.6 (W) x 2.4 (H) meter in dimensions was fabricated at the 10MW LSC (Large Scale Calorimeter) and combustible office supplies were placed in the compartment for a real-scale fire test. Maximum heat release rate was 4.1 MW and total energy release obtained through the application of t2 fire growth rate was 6705.9 MJ.

Keywords: fire growth, fire experiment, t2 curve, large scale calorimeter

Procedia PDF Downloads 303

Authors: Xiaoming Jiang, Jinqiao Shi, Qingfeng Tan, Wentao Zhang, Xuebin Wang, Muqian Chen

Abstract:

Nowadays, big companies such as Google, Microsoft, which have adequate proxy servers, have perfectly implemented their web crawlers for a certain website in parallel. But due to lack of expensive proxy servers, it is still a puzzle for researchers to crawl large amounts of information from a single website in parallel. In this case, it is a good choice for researchers to use free public proxy servers which are crawled from the Internet. In order to improve efficiency of web crawler, the following two issues should be considered primarily: (1) Tasks may fail owing to the instability of free proxy servers; (2) A proxy server will be blocked if it visits a single website frequently. In this paper, we propose Proxisch, an optimization approach of large-scale unstable proxy servers scheduling, which allow anyone with extremely low cost to run a web crawler efficiently. Proxisch is designed to work efficiently by making maximum use of reliable proxy servers. To solve second problem, it establishes a frequency control mechanism which can ensure the visiting frequency of any chosen proxy server below the website’s limit. The results show that our approach performs better than the other scheduling algorithms.

Keywords: proxy server, priority queue, optimization algorithm, distributed web crawling

Procedia PDF Downloads 183

Authors: Fatima Bellali, Mariem Kharroubi

Abstract:

Fish scale wastes are one of the main sources of production of value-added products such as collagen. The main aim of this study is to investigate the optimization conditions of the sardine scale deproteinization using response surface methodology (RSM) on a pilot scale. In order to look for the optimal conditions, a Box–Behnken-based design of experiment (DOE) method was carried out. The model predicted values of product coal ash content were in good agreement with the experiment values (R2 = 0.9813). Finally, model-based optimization was carried out to identify the operating parameters (reaction time=4h and the solid-liquid ratio= 1/10) and to obtain the lowest collagen content.

Keywords: pilot scale, Plackett and Burman design, fish waste, deproteinization

Procedia PDF Downloads 115

Authors: Nur Iffika Ruslan, Ahmad Rosly Abbas, Munirah Stapah@Salleh, Nurfaziera Rahim

Abstract:

Increased concerns and awareness of environmental hazards by fossil fuels burning for energy have become the major factor driving the transition toward green energy. It is expected that an additional of 2,000 MW of renewable energy is to be recorded from the renewable sources by 2025 following the implementation of Large Scale Solar projects in Peninsular Malaysia, including Large Scale Floating Solar projects. Floating Solar has better advantages over its landed counterparts such as the requirement for land acquisition is relatively insignificant. As part of the site selection process established by TNB Research Sdn. Bhd., a set of mandatory and rejection criteria has been developed in order to identify only sites that are feasible for the future development of Large Scale Floating Solar power plant. There are a total of 85 lakes and reservoirs identified within Peninsular Malaysia. Only lakes and reservoirs with a minimum surface area of 120 acres will be considered as potential sites for the development of Large Scale Floating Solar power plant. The result indicates a total of 10 potential Large Scale Floating Solar sites identified which are located in Selangor, Johor, Perak, Pulau Pinang, Perlis and Pahang. This paper will elaborate on the various mandatory and rejection criteria, as well as on the various site selection process required to identify potential (suitable) Large Scale Floating Solar sites in Peninsular Malaysia.

Keywords: Large Scale Floating Solar, Peninsular Malaysia, Potential Sites, Renewable Energy

Procedia PDF Downloads 150

Authors: Lokendra Kumar Devangan, Ajay Mishra

Abstract:

This paper describes the implementation of a large-scale SAS/OR model with significant pre-processing, scenario analysis, and post-processing work done using SAS. A large cement manufacturer with ten geographically distributed manufacturing plants for two variants of cement, around 400 warehouses serving as transshipment points, and several thousand distributor locations generating demand needed to optimize this multi-echelon, multi-modal transport supply chain separately for planning and allocation purposes. For monthly planning as well as daily allocation, the demand is deterministic. Rail and road networks connect any two points in this supply chain, creating tens of thousands of such connections. Constraints include the plant’s production capacity, transportation capacity, and rail wagon batch size constraints. Each demand point has a minimum and maximum for shipments received. Price varies at demand locations due to local factors. A large mixed integer programming model built using proc OPTMODEL decides production at plants, demand fulfilled at each location, and the shipment route to demand locations to maximize the profit contribution. Using base SAS, we did significant pre-processing of data and created inputs for the optimization. Using outputs generated by OPTMODEL and other processing completed using base SAS, we generated several reports that went into their enterprise system and created tables for easy consumption of the optimization results by operations.

Keywords: production planning, mixed integer optimization, network model, network optimization

Procedia PDF Downloads 27

Authors: Yitao Lei, Xingxiang Zhai, Burra Venkata Durga Kumar

Abstract:

As the quantity and complexity of computing in large-scale software systems increase, distributed system computing becomes increasingly important. The distributed system realizes high-performance computing by collaboration between different computing resources. If there are no efficient resource scheduling resources, the abuse of distributed computing may cause resource waste and high costs. However, resource scheduling is usually an NP-hard problem, so we cannot find a general solution. However, some optimization algorithms exist like genetic algorithm, ant colony optimization, etc. The large scale of distributed systems makes this traditional optimization algorithm challenging to work with. Heuristic and machine learning algorithms are usually applied in this situation to ease the computing load. As a result, we do a review of traditional resource scheduling optimization algorithms and try to introduce a deep reinforcement learning method that utilizes the perceptual ability of neural networks and the decision-making ability of reinforcement learning. Using the machine learning method, we try to find important factors that influence the performance of distributed system computing and help the distributed system do an efficient computing resource scheduling. This paper surveys the application of deep reinforcement learning on distributed system computing resource scheduling proposes a deep reinforcement learning method that uses a recurrent neural network to optimize the resource scheduling, and proposes the challenges and improvement directions for DRL-based resource scheduling algorithms.

Keywords: resource scheduling, deep reinforcement learning, distributed system, artificial intelligence

Procedia PDF Downloads 80

Authors: B. Sellami, Y. Laskri, M. Belloufi

Abstract:

Conjugate gradient methods are an important class of methods for unconstrained optimization, especially for large-scale problems. Recently, they have been much studied. In this paper, a new family of conjugate gradient method is proposed for unconstrained optimization. This method includes the already existing two practical nonlinear conjugate gradient methods, which produces a descent search direction at every iteration and converges globally provided that the line search satisfies the Wolfe conditions. The numerical experiments are done to test the efficiency of the new method, which implies the new method is promising. In addition the methods related to this family are uniformly discussed.

Keywords: conjugate gradient method, global convergence, line search, unconstrained optimization

Procedia PDF Downloads 375

Authors: Jali Octariady, Ida Herliningsih, Ade K. Mulyana, Annisa Fitria, Diaz C. K. Yuwana

Abstract:

Awareness of the importance large-scale maps for development of a country is growing in all walks of life, especially for governments in Indonesia. Various parties, especially local governments throughout Indonesia demanded for immediate availability the large-scale maps of 1:5000 for regional development. But in fact, the large-scale maps of 1:5000 is only available less than 5% of the entire territory of Indonesia. Unavailability precise GCP at the entire territory of Indonesia is one of causes of slow availability the large scale maps of 1:5000. This research was conducted to find an alternative solution to this problem. This study was conducted to assess the accuracy of ArcGIS base maps coordinate when it shall be used as GCP for creating a map scale of 1:5000. The study was conducted by comparing the GCP coordinate from Field survey using GPS Geodetic than the coordinate from ArcGIS basemaps in various locations in Indonesia. Some areas are used as a study area are Lombok Island, Kupang City, Surabaya City and Kediri District. The differences value of the coordinates serve as the basis for assessing the accuracy of ArcGIS basemaps coordinates. The results of the study at various study area show the variation of the amount of the coordinates value given. Differences coordinate value in the range of millimeters (mm) to meters (m) in the entire study area. This is shown the inconsistency quality of ArcGIS base maps coordinates. This inconsistency shows that the coordinate value from ArcGIS Basemaps is careless. The Careless coordinate from ArcGIS Basemaps indicates that its cannot be used as GCP for large-scale mapping on the entire territory of Indonesia.

Keywords: accuracy, ArcGIS base maps, GCP, large scale maps

Procedia PDF Downloads 344

Authors: Byeong-Sam Kim, Kyeongwoo Park

Abstract:

In this study, induced structural optimization is performed to compare the trade-off between wing weight and induced drag for wing panel extensions, construction of wing panel and winglets. The aerostructural optimization problem consists of parameters with strength condition, and two maneuver conditions using residual stresses in panel production. The results of kinematic motion analysis presented a homogenization based theory for 3D beams and 3D shells for wing panel. This theory uses a kinematic description of the beam based on normalized displacement moments. The displacement of the wing is a significant design consideration as large deflections lead to large stresses and increased fatigue of components cause residual stresses. The stresses in the wing panel are small compared to the yield stress of aluminum alloy. This study describes the implementation of a large wing panel, aerostructural analysis and structural parameters optimization framework that couples a three-dimensional panel method.

Keywords: wing panel, aerostructural optimization, FEM, structural analysis

Procedia PDF Downloads 552

Authors: Arash Parehkar, S. Jamshid Mousavi, Shoubo Bayazidi, Vahid Karami, Laleh Shahidi, Arash Azaranfar, Ali Moridi, M. Shabakhti, Tayebeh Ariyan, Mitra Tofigh, Kaveh Masoumi, Alireza Motahari

Abstract:

Water resource systems modelling have constantly been a challenge through history for human being. As the innovative methodological development is evolving alongside computer sciences on one hand, researches are likely to confront more complex and larger water resources systems due to new challenges regarding increased water demands, climate change and human interventions, socio-economic concerns, and environment protection and sustainability. In this research, an automatic calibration scheme has been applied on the Gilan’s large-scale water resource model using mathematical programming. The water resource model’s calibration is developed in order to attune unknown water return flows from demand sites in the complex Sefidroud irrigation network and other related areas. The calibration procedure is validated by comparing several gauged river outflows from the system in the past with model results. The calibration results are pleasantly reasonable presenting a rational insight of the system. Subsequently, the unknown optimized parameters were used in a basin-scale linear optimization model with the ability to evaluate the system’s performance against a reduced inflow scenario in future. Results showed an acceptable match between predicted and observed outflows from the system at selected hydrometric stations. Moreover, an efficient operating policy was determined for Sefidroud dam leading to a minimum water shortage in the reduced inflow scenario.

Keywords: auto-calibration, Gilan, large-scale water resources, simulation

Procedia PDF Downloads 306

Authors: Ming Liu, Chong Wu, Bingquan Liu, Lei Chen

Abstract:

Internet users have to face the massive amount of textual data every day. Organizing texts into categories can help users dig the useful information from large-scale text collection. Clustering, in fact, is one of the most promising tools for categorizing texts due to its unsupervised characteristic. Unfortunately, most of traditional clustering algorithms lose their high qualities on large-scale text collection. This situation mainly attributes to the high- dimensional vectors generated from texts. To effectively and efficiently cluster large-scale text collection, this paper proposes a vector reconstruction based clustering algorithm. Only the features that can represent the cluster are preserved in cluster’s representative vector. This algorithm alternately repeats two sub-processes until it converges. One process is partial tuning sub-process, where feature’s weight is fine-tuned by iterative process. To accelerate clustering velocity, an intersection based similarity measurement and its corresponding neuron adjustment function are proposed and implemented in this sub-process. The other process is overall tuning sub-process, where the features are reallocated among different clusters. In this sub-process, the features useless to represent the cluster are removed from cluster’s representative vector. Experimental results on the three text collections (including two small-scale and one large-scale text collections) demonstrate that our algorithm obtains high quality on both small-scale and large-scale text collections.

Keywords: vector reconstruction, large-scale text clustering, partial tuning sub-process, overall tuning sub-process

Procedia PDF Downloads 401

Authors: Yihao Kuang, Bowen Ding

Abstract:

With the increasing scale and complexity of knowledge graph, modern knowledge graph contains more and more types of entity, relationship, and attribute information. Therefore, in recent years, it has been a trend for knowledge graph inference to use reinforcement learning to deal with large-scale, incomplete, and noisy knowledge graphs and improve the inference effect and interpretability. The Proximal Policy Optimization (PPO) algorithm utilizes a near-end strategy optimization approach. This allows for more extensive updates of policy parameters while constraining the update extent to maintain training stability. This characteristic enables PPOs to converge to improved strategies more rapidly, often demonstrating enhanced performance early in the training process. Furthermore, PPO has the advantage of offline learning, effectively utilizing historical experience data for training and enhancing sample utilization. This means that even with limited resources, PPOs can efficiently train for reinforcement learning tasks. Based on these characteristics, this paper aims to obtain a better and more efficient inference effect by introducing PPO into knowledge inference technology.

Keywords: reinforcement learning, PPO, knowledge inference

Procedia PDF Downloads 193

Authors: Belloufi Mohammed, Sellami Badreddine

Abstract:

Conjugate gradient methods have played a special role for solving large scale optimization problems due to the simplicity of their iteration, convergence properties and their low memory requirements. In this work, we propose a new class of conjugate gradient methods which ensures sufficient descent. Moreover, we propose a new search direction with the Wolfe line search technique for solving unconstrained optimization problems, a global convergence result for general functions is established provided that the line search satisfies the Wolfe conditions. Our numerical experiments indicate that our proposed methods are preferable and in general superior to the classical conjugate gradient methods in terms of efficiency and robustness.

Keywords: unconstrained optimization, conjugate gradient method, sufficient descent property, numerical comparisons

Procedia PDF Downloads 368

Authors: Yihao Kuang, Bowen Ding

Abstract:

With the increasing scale and complexity of knowledge graph, modern knowledge graph contains more and more types of entity, relationship, and attribute information. Therefore, in recent years, it has been a trend for knowledge graph inference to use reinforcement learning to deal with large-scale, incomplete, and noisy knowledge graph and improve the inference effect and interpretability. The Proximal Policy Optimization (PPO) algorithm utilizes a near-end strategy optimization approach. This allows for more extensive updates of policy parameters while constraining the update extent to maintain training stability. This characteristic enables PPOs to converge to improve strategies more rapidly, often demonstrating enhanced performance early in the training process. Furthermore, PPO has the advantage of offline learning, effectively utilizing historical experience data for training and enhancing sample utilization. This means that even with limited resources, PPOs can efficiently train for reinforcement learning tasks. Based on these characteristics, this paper aims to obtain better and more efficient inference effect by introducing PPO into knowledge inference technology.

Keywords: reinforcement learning, PPO, knowledge inference, supervised learning

Procedia PDF Downloads 31

Authors: Mohammad H. Almomani

Abstract:

In this paper, we argue the effect of the initial sample size, and the increment in simulation samples on the performance of a sequential approach that used in selecting the top m designs when the number of alternative designs is very large. The sequential approach consists of two stages. In the first stage the ordinal optimization is used to select a subset that overlaps with the set of actual best k% designs with high probability. Then in the second stage the optimal computing budget is used to select the top m designs from the selected subset. We apply the selection approach on a generic example under some parameter settings, with a different choice of initial sample size and the increment in simulation samples, to explore the impacts on the performance of this approach. The results show that the choice of initial sample size and the increment in simulation samples does affect the performance of a selection approach.

Keywords: Large Scale Problems, Optimal Computing Budget Allocation, ordinal optimization, simulation optimization

Procedia PDF Downloads 323

Authors: Serges Mendomo Meye, Li Guowei, Shen Zhenzhong, Gan Lei, Xu Liqun

Abstract:

Given the limitations of the original whale optimization algorithm (WAO) in local optimum and low convergence accuracy in truss structure optimization problems, based on the fundamental whale algorithm, an improved whale optimization algorithm (SWAO) based on information entropy is proposed. The information entropy itself is an uncertain measure. It is used to control the range of whale searches in path selection. It can overcome the shortcomings of the basic whale optimization algorithm (WAO) and can improve the global convergence speed of the algorithm. Taking truss structure as the optimization research object, the mathematical model of truss structure optimization is established; the cross-sectional area of truss is taken as the design variable; the objective function is the weight of truss structure; and an improved whale optimization algorithm (SWAO) is used for optimization design, which provides a new idea and means for its application in large and complex engineering structure optimization design.

Keywords: information entropy, structural optimization, truss structure, whale algorithm

Procedia PDF Downloads 208

Authors: Akshay Paranjape, Nils Plettenberg, Robert Schmitt

Abstract:

Producing faulty products can be costly for manufacturing companies and wastes resources. To reduce scrap rates in manufacturing, process parameters can be optimized using machine learning. Thus far, research mainly focused on optimizing specific processes using traditional algorithms. To develop a framework that enables real-time optimization based on a predictive model for an arbitrary production process, this study explores the application of reinforcement learning (RL) in this field. Based on a thorough review of literature about RL and process parameter optimization, a model based on maximum a posteriori policy optimization that can handle both numerical and categorical parameters is proposed. A case study compares the model to state–of–the–art traditional algorithms and shows that RL can find optima of similar quality while requiring significantly less time. These results are confirmed in a large-scale validation study on data sets from both production and other fields. Finally, multiple ways to improve the model are discussed.

Keywords: reinforcement learning, production process optimization, evolutionary algorithms, policy optimization, actor critic approach

Procedia PDF Downloads 59

Authors: Bernard Doherty, Andrew Jelfs, Aveek Dasgupta, Patrick Holden

Abstract:

Many companies have moved to agile and hybrid agile methodologies where portions of the Software Design Life-cycle (SDLC) and Software Test Life-cycle (STLC) can be time boxed in order to enhance delivery speed, quality and to increase flexibility to changes in software requirements. Despite widespread proliferation of agile practices, implementation often fails due to lack of adequate project management support, decreased motivation or fear of increased interaction. Consequently, few organizations effectively adopt agile processes with tailoring often required to integrate agile methodology in large scale environments. This paper provides an overview of the challenges in implementing an innovative large-scale tailored realization of the agile methodology termed the Agile Factory Model (AFM), with the aim of comparing and contrasting issues of specific importance to organizations undertaking large scale agile development. The conclusions demonstrate that agile practices can be effectively translated to a globally distributed development environment.

Keywords: agile, agile factory model, globally distributed development, large-scale agile

Procedia PDF Downloads 264

Authors: Supriya Manikrao Ovhal

Abstract:

In India, the summer monsoon rainfall occurs owing to large scale convection with reference to continental ITCZ. It was found that convection over tropical ocean increases with SST from 26 to 28 degree C, and when SST is above 29 degree C, it sharply decreases for warm pool areas of Indian and for monsoon areas of West Pacific Ocean. The reduction in convection can be influenced by large scale subsidence forced by nearby or remotely generated deep convection, thus it was observed that under the influence of strong large scale rising motion, convection does not decreases but increases monotonically with SST even if SST value is higher than 29.5 degree C. Since convection is related to SST gradient, that helps to generate low level moisture convergence and upward vertical motion in the atmosphere. Strong wind fields like cross equatorial low level jet stream on equator ward side of the warm pool are produced due to convection initiated by SST gradient. Areas having maximum SST have low SST gradient, and that result in feeble convection. Hence it is imperative to mention that the oceanic role (other than SST) could be prominent in influencing large Scale Atmospheric convection. Since warm oceanic surface somewhere or the other contributes to penetrate the heat radiation to the subsurface of the ocean, and as there is no studies seen related to oceanic subsurface role in large Scale Atmospheric convection, in the present study, we are concentrating on the oceanic subsurface contribution in large Scale Atmospheric convection by considering the SST gradient, mixed layer depth (MLD), thermocline, barrier layer. The present study examines the probable role of subsurface ocean parameters in influencing convection.

Keywords: sst, d20, olr, wind

Procedia PDF Downloads 61

Authors: Akshay Paranjape, Nils Plettenberg, Robert Schmitt

Abstract:

Traditional optimization methods like evolutionary algorithms are widely used in production processes to find an optimal or near-optimal solution of control parameters based on the simulated environment space of a process. These algorithms are computationally intensive and therefore do not provide the opportunity for real-time optimization. This paper utilizes the Deep Reinforcement Learning (DRL) framework to find an optimal or near-optimal solution for control parameters. A model based on maximum a posteriori policy optimization (Hybrid-MPO) that can handle both numerical and categorical parameters is used as a benchmark for comparison. A comparative study shows that DRL can find optimal solutions of similar quality as compared to evolutionary algorithms while requiring significantly less time making them preferable for real-time optimization. The results are confirmed in a large-scale validation study on datasets from production and other fields. A trained XGBoost model is used as a surrogate for process simulation. Finally, multiple ways to improve the model are discussed.

Keywords: reinforcement learning, evolutionary algorithms, production process optimization, real-time optimization, hybrid-MPO

Procedia PDF Downloads 73

Authors: B. Sellami, M. Belloufi

Abstract:

Conjugate gradient methods are an important class of methods for unconstrained optimization, especially for large-scale problems. Recently, they have been much studied. In this paper, we propose a new two-parameter family of conjugate gradient methods for unconstrained optimization. The two-parameter family of methods not only includes the already existing three practical nonlinear conjugate gradient methods, but also has other family of conjugate gradient methods as subfamily. The two-parameter family of methods with the Wolfe line search is shown to ensure the descent property of each search direction. Some general convergence results are also established for the two-parameter family of methods. The numerical results show that this method is efficient for the given test problems. In addition, the methods related to this family are uniformly discussed.

Keywords: unconstrained optimization, conjugate gradient method, line search, global convergence

Procedia PDF Downloads 413

Authors: Thomas Weber, Nina Strobel, Thomas Kohne, Eberhard Abele

Abstract:

In context of the energy transition in Germany, the importance of so-called virtual power plants in the energy supply continues to increase. The progressive dismantling of the large power plants and the ongoing construction of many new decentralized plants result in great potential for optimization through synergies between the individual plants. These potentials can be exploited by mathematical optimization algorithms to calculate the optimal application planning of decentralized power and heat generators and storage systems. This also includes linear or linear mixed integer optimization. In this paper, procedures for reducing the number of decision variables to be calculated are explained and validated. On the one hand, this includes combining n similar installation types into one aggregated unit. This aggregated unit is described by the same constraints and target function terms as a single plant. This reduces the number of decision variables per time step and the complexity of the problem to be solved by a factor of n. The exact operating mode of the individual plants can then be calculated in a second optimization in such a way that the output of the individual plants corresponds to the calculated output of the aggregated unit. Another way to reduce the number of decision variables in an optimization problem is to reduce the number of time steps to be calculated. This is useful if a high temporal resolution is not necessary for all time steps. For example, the volatility or the forecast quality of environmental parameters may justify a high or low temporal resolution of the optimization. Both approaches are examined for the resulting calculation time as well as for optimality. Several optimization models for virtual power plants (combined heat and power plants, heat storage, power storage, gas turbine) with different numbers of plants are used as a reference for the investigation of both processes with regard to calculation duration and optimality.

Keywords: CHP, Energy 4.0, energy storage, MILP, optimization, virtual power plant

Procedia PDF Downloads 145