The optimization of pump operations has been widely studied, as it can decrease operational and maintenance costs and can reduce greenhouse gas emissions caused by the energy consumption from fossil fueled electricity...
详细信息
The optimization of pump operations has been widely studied, as it can decrease operational and maintenance costs and can reduce greenhouse gas emissions caused by the energy consumption from fossil fueled electricity sources. However, only the optimization of pump scheduling (where pumps are controlled based on times) and the optimization of simple controls (where pumps are controlled based on one condition only, e.g. the level of one tank) were previously able to be used in the EPANET2 toolkit. This paper uses a modified version of the hydraulic solver EPANET2 that enables rule-based controls (i.e. controls based on more than one condition) to be automatically changed by an optimization algorithm. This modification is particularly useful in cases where the pump operations need to be decided based on multiple conditions: typical examples are the cases where the pumps are controlled according to the water levels of multiple tanks or when both tank levels and time of day are taken into account to reduce the pumping in the peak tariff period. The new toolkit, called ETTAR (EPANET2 Toolkit to Alter Rules), is applied to a large case study, where different types of pump operations will be tested. Results show that the optimization of rule-based controls can decrease operational costs while guaranteeing robust pump controls.
Additive Metal Deposition (AMD) is an additive manufacturing process building parts based on a nozzle-fed powder by laser assisted solidification. The AMD technology offers unique advantages for the production of near...
详细信息
Additive Metal Deposition (AMD) is an additive manufacturing process building parts based on a nozzle-fed powder by laser assisted solidification. The AMD technology offers unique advantages for the production of near net-shape parts. In contrast to the powder bed-based technologies it provides a high productivity grade. Today AMD lacks reproducible process strategies manufacturing large parts in narrow tolerances. The building height of a single layer and the geometrical shape of a whole part alter progressively with increasing part dimensions - consecutively leading to a higher effort in the manufacturing-process development for such parts. To reduce this effort, in this paper first an iterative identification of optimal process parameters is performed by following an evolutionary algorithm under varied BC. Based on the geometry-related parameter sets, tolerances are defined. The process strategies and tolerances are validated for a prototype application considering the defined quality aims. Finally the results are discussed and summarized in an a-priori process design guideline for AMD Ti6Al4V-parts. (C) 2016 The Authors. Published by Elsevier B.V.
The assessment of a building's energy performance as a design factor in the early design stages is a very demanding and complex procedure. Over the last decades, a number of tools and methods have been developed t...
详细信息
The assessment of a building's energy performance as a design factor in the early design stages is a very demanding and complex procedure. Over the last decades, a number of tools and methods have been developed to address performance-related design questions, mostly using Multi-Objective Optimization algorithms. Parametric modelling offers dynamic control over geometry and components, allowing the designer to assess multiple variants at the same time. In this paper, a new design workflow methodology is proposed, integrating evolutionary algorithms and energy simulation through Grasshopper for Rhinoceros 3d, for a comprehensive exploration of performance-based design alternatives in the building scale.
Integration of non-dispatchable renewable energy sources such as wind and solar into the grid is challenging due to the stochastic nature of energy sources. Hence, electrical hubs (EH) and virtual power plants that co...
详细信息
ISBN:
(纸本)9780791850220
Integration of non-dispatchable renewable energy sources such as wind and solar into the grid is challenging due to the stochastic nature of energy sources. Hence, electrical hubs (EH) and virtual power plants that combine non-dispatchable energy sources, energy storage and dispatchable energy sources such as internal combustion generators and micro gas turbines are getting popular. However, designing such energy systems considering the electricity demand of a neighborhood, curtailments for grid interactions and real time pricing (RTP) of the main utility grid (MUG) is a difficult exercise. Seasonal and hourly variation of electricity demand, potential for each non-dispatchable energy source and RTP of MUG needs to be considered when designing the energy system. Representation of dispatch strategy plays a major role in this process where simultaneous optimization of system design and dispatch strategy is required. This study presents a bi-level dispatch strategy based on reinforced learning for simultaneous optimization of system design and operation strategy of an EH. Artificial Neural Network (ANN) was combined with a finite state controller to obtain the operating state of the system. Pareto optimization is conducted considering, lifecycle cost and system autonomy to obtain optimum system design using evolutionary algorithm.
Optimization plays a key role in MEMS design. However, most MEMS design optimization (exploration) methods either depend on ad-hoc analytical / behavioural models or time consuming numerical simulations. Surrogate mod...
详细信息
ISBN:
(纸本)9783981537079
Optimization plays a key role in MEMS design. However, most MEMS design optimization (exploration) methods either depend on ad-hoc analytical / behavioural models or time consuming numerical simulations. Surrogate modeling techniques have been introduced to integrate generality and efficiency, but the number of design variables which can be handled by most existing efficient MEMS design optimization methods is often less than 5. To address the above challenges, a new method, called Adaptive Gaussian Process-Assisted Differential Evolution for MEMS Design Optimization (AGDEMO) is proposed. The key idea is the proposed ON-LINE adaptive surrogate model assisted optimization framework. In particular, AGDEMO performs global optimization of MEMS using numerical simulation and the differential evolution (DE) algorithm, and a Gaussian process surrogate model is constructed ONLINE to predict the results of expensive numerical simulations. AGDEMO is tested by two actuators (both with 9 design variables). Comparisons with state-of-the-art methods verify advantages of AGDEMO in terms of efficiency, optimization capacity and scalability.
In this paper we are continuing in our research to show mutual intersection of two different areas of research: complex network and evolutionary computation. This research parer is focused on possibility to convert ru...
详细信息
ISBN:
(纸本)9781509041237
In this paper we are continuing in our research to show mutual intersection of two different areas of research: complex network and evolutionary computation. This research parer is focused on possibility to convert run of evolution algorithm to a complex network inspired by ants. Such network can then be analyzed to get useful information about algorithm dynamics. In this paper we focused on one global network property, average network strength. This property is described and used for different types of cost functions and different representations of network conversion.
Energy management system (EMS) is crucial to a plug-in hybrid electric vehicle (PHEV) in reducing its fuel consumption and pollutant emissions. The EMS determines how energy flows in a hybrid powertrain should be mana...
详细信息
ISBN:
(纸本)9781450343237
Energy management system (EMS) is crucial to a plug-in hybrid electric vehicle (PHEV) in reducing its fuel consumption and pollutant emissions. The EMS determines how energy flows in a hybrid powertrain should be managed in response to a variety of driving conditions. In the development of an EMS, the battery state-of-charge (SOC) control strategy plays a critical role. This paper proposes a novel evolutionary algorithm (EA)-based EMS with a self-adaptive SOC control strategy for PHEVs, which can significantly improve the fuel efficiency without knowing the trip length (in time). Numerical studies show that this proposed system can save up to 13% fuel, compared to other on-line EMS with different SOC control strategies. Further analysis indicates that the proposed system is less sensitive to the errors in predicting propulsion power demand in real-time, which is favorable for online implementation. Original publication: X. Qi, G. Wu, K. Boriboonsomsin and M. J. Barth, evolutionary algorithm based online PHEV energy management system with self-adaptive SOC control, Intelligent Vehicles Symposium (IV), 2015 IEEE, Seoul, 2015, pp. 425-430.
evolutionary Computing for Educational Data Mining is a research field which with the applications of evolutionary algorithms (EAs) to mine, analyze and modify educational data. This paper presents the most relevant s...
详细信息
ISBN:
(纸本)9781509006120
evolutionary Computing for Educational Data Mining is a research field which with the applications of evolutionary algorithms (EAs) to mine, analyze and modify educational data. This paper presents the most relevant studies conducted in this research area. The paper also describes different EAs used for implementing different data mining techniques. It goes on to list how these algorithms are utilized by different educational users to carry out different tasks. Finally, a new combination of EA, Educational User and data mining technique is suggested for implementation. As a part of that a personalized courseware construction technique is proposed and a sample courseware is constructed using the proposed technique. The details about the rule construction and the data mining process involved in the courseware construction techniques are also explained.
Most improvements for Naive Bayes (NB) have a common yet important flaw - these algorithms split the modeling of the classifier into two separate stages - the stage of preprocessing (e.g., feature selection and data e...
详细信息
ISBN:
(纸本)9781509006199
Most improvements for Naive Bayes (NB) have a common yet important flaw - these algorithms split the modeling of the classifier into two separate stages - the stage of preprocessing (e.g., feature selection and data expansion) and the stage of building the NB classifier. The first stage does not take the NB's objective function into consideration, so the performance of the classification cannot be guaranteed. Motivated by these facts and aiming to improve NB with accurate classification, we present a new learning algorithm called evolutionary Local Instance Weighted Naive Bayes or ELWNB, to extend NB for classification. ELWNB combines local NB, instance weighted dataset extension and evolutionary algorithms seamlessly. Experiments on 20 UCI benchmark datasets demonstrate that ELWNB significantly outperforms NB and several other improved NB algorithms.
Decision tree induction is inherently a multi-objective task. However, most of the conventional learning algorithms can only deal with a single-objective that may possibly aggregate multiple objectives. This paper pro...
详细信息
ISBN:
(纸本)9783319490014;9783319490007
Decision tree induction is inherently a multi-objective task. However, most of the conventional learning algorithms can only deal with a single-objective that may possibly aggregate multiple objectives. This paper proposes the multi-objective evolutionary approach to Pareto optimal model trees. We developed a set of non-dominated model trees for a Global Model Tree framework using efficient sort and specialized selection. Performed study covers variants with two and three objectives that relate to the tree error and the tree comprehensibility. Pareto front generated by the GMT system allows the decision maker to select desired output model according to his preferences on the conflicting objectives. Experimental evaluation of the proposed approach is performed on three real-life datasets and is confronted with competitive model tree inducers.
暂无评论