Optimization and Engineering最新文献

Countries around the world are gradually implementing the transition to a circular economy in waste management. This effort should be initiated already at the waste producers. It is necessary to plan and monitor waste management in as much detail as possible, i.e. at the level of micro-regions. At present, only indicators at the national level are analysed, as more detailed data at the micro-regional level are often not available or are burdened with significant errors and inconsistencies. The calculation of waste management indicators for micro-regions will allow to identify the potential for increasing material or energy recovery and to plan the necessary infrastructure directly to these locations instead of blanket and often ineffective legislative actions. This paper presents an approach for determining the producer-treatment linkage, i.e., provides information about each produced waste, where it was treated, and in what way. Such information is often not available based on historical waste management data as there are repeated waste transfers and often aggregated within a micro-region. The network flow approach is based on an iterative procedure combining a simulation with multi-criteria optimization. The chosen criteria replicate expert estimates in investigated issue such as minimum flow splitting, and minimum transfer micro-regions. A data reconciliation is performed where the deviation from all simulations is minimized, given that the capacity constraints of nodes and arcs resulting from the database must be satisfied. The approach is tested on a generated sample task to evaluate the precision and time complexity of the developed tool. Finally, the presented approach is applied to address a case study in the Czech Republic, within which it is possible to identify treatment location and methods for waste from individual regions.

The coupling problem between the structural field and temperature field is widely encountered in engineering applications and holds significant research importance. In the context of thermo-mechanical coupling topology optimization, the thermal stress resulting from the temperature field can cause structural deformation, consequently impacting the structural performance. Therefore, it is crucial to conduct rational optimization designs to ensure favorable mechanical behavior and heat dissipation. This study utilizes thermo-mechanical coupling theory to perform multi-objective topology optimization, aiming to minimize compliance and heat dissipation weakness concurrently, thereby obtaining a more comprehensive design scheme with enhanced overall performance. Initially, a topological optimization model for the coupled thermo-mechanical problem is established. Subsequently, the objective functions of structural compliance and heat dissipation weakness are normalized, and their sensitivities are derived. Next, a multi-load case and multi-objective optimization algorithm based on Floating Projection Topological Optimization (FPTO) is proposed to minimize both structural compliance and heat dissipation weakness. By comparing the topological configuration and objective function values obtained using the Solid Isotropic Material with Penalization (SIMP) method, it is evident that the FPTO method achieves clear and smooth boundaries in the topological configuration, while yielding smaller objective function values. Additionally, under appropriate trade-off factors, the FPTO method achieves a more balanced topological structure and optimizes material distribution without increasing the structural volume, thus enabling lightweight structures, providing novel ideas and methods for engineering applications.

A new bounds tightening algorithm for globally solving AC optimal power flow (ACOPF) problems is presented. Practical ACOPF instances are too large to be solved by conventional global optimization algorithms based on extensive search-space partitioning. However, tailored optimization-based bounds tightening (OBBT) algorithms using advanced relaxation techniques have been shown to achieve tight optimality gaps for many test cases with no partitioning at all. Unfortunately, OBBT is still costly because it requires solving two convex subproblems per decision variable in each iteration. We present a new OBBT algorithm, using a new SOCP based relaxation, that achieves tight optimality gaps while only solving subproblems for a small subset of variables. For PGLIB benchmarks up to 300 buses, the algorithm achieves the best gap on more test problems and is significantly faster on average than two existing OBBT algorithms chosen for comparison.

We address in this paper linear programming (LP) models in which it is desired to find a finite set of alternate optima. An LP may have multiple alternate solutions with the same objective value or with increasing objective values. For many real life applications, it can be interesting to have a pool of solutions to compare what operations should be executed and what is the cost/benefit of doing it. To obtain a specified number of these alternate solutions in the increasing order of objective values, we propose an iterative MILP algorithm in which we successively add integer cuts on inactive constraints. We demonstrate the application and effectiveness of this algorithm on a 2 dimensional LP and on small and large supply chain problems. The proposed iterative MILP algorithm provides an effective approach for finding a specified number of alternate optima in LP models, which provides a useful tool in a variety of applications as for instance in supply chain optimization problems.

In this work, we give a generalized formulation of the Black–Scholes model. The novelty resides in considering the Black–Scholes model to be valid on ’average’, but such that the pointwise option price dynamics depends on a measure representing the investors’ ’uncertainty’. We make use of the theory of non-symmetric Dirichlet forms and the abstract theory of partial differential equations to establish well posedness of the problem. A detailed numerical analysis is given in the case of self-similar measures.

In this paper we define a fractional Sturm–Liouville problem (FSLP) on [0, 1] subject to dirichlet boundary condition. First we discretize FSLP to obtain the corresponding matrix eigenvalue problem (MEP) of finite order N. In direct problem we give an efficient numerical algorithm to make good approximations for eigenvalues of FSLP by adding a correction term to eigenvalues of MEP. For inverse problem, using the idea of correction technique, we propose an algorithm for recovering the symmetric potential function using one given spectrum. Finally, we give some numerical examples to show the efficiency of the proposed algorithm.

Abstract

We propose an ensemble artificial neural network (EANN) methodology for predicting the day ahead energy demand of a district heating operator (DHO). Specifically, at the end of one day, we forecast the energy demand for each of the 24 h of the next day. Our methodology combines three artificial neural network (ANN) models, each capturing a different aspect of the predicted time series. In particular, the outcomes of the three ANN models are combined into a single forecast. This is done using a sequential ordered optimization procedure that establishes the weights of three models in the final output. We validate our EANN methodology using data obtained from a A2A, which is one of the major DHOs in Italy. The data pertains to a major metropolitan area in Northern Italy. We compared the performance of our EANN with the method currently used by the DHO, which is based on multiple linear regression requiring expert intervention. Furthermore, we compared our EANN with the state-of-the-art seasonal autoregressive integrated moving average and Echo State Network models. The results show that our EANN achieves better performance than the other three methods, both in terms of mean absolute percentage error (MAPE) and maximum absolute percentage error. Moreover, we demonstrate that the EANN produces good quality results for longer forecasting horizons. Finally, we note that the EANN is characterised by simplicity, as it requires little tuning of a handful of parameters. This simplicity facilitates its replicability in other cases.

Conventional knowledge graph completion methods are effective for completing knowledge graphs (KGs), but they face significant challenges when dealing with relations with only a limited number of associative triples. To address the issue of incompleteness and long-tail distribution of relations in KGs, few-shot knowledge graph completion emerges as a promising solution. This approach predicts new triplets about a relation by leveraging only a handful of associated triples. Previous methods have focused on aggregating neighbor information and imposing sequential dependency assumptions. However, these methods can be counterproductive when they involve unrelated neighbors and rely on unrealistic assumptions, which hinders the learning of meta-representations. This paper proposes a simple and effective meta relational learning model (SMetaR) for few-shot knowledge graph completion that maintains the complete feature information of few-shot relations through a linear model. This approach effectively learns the meta-representation of few-shot relations and enhances meta-relational learning capabilities. Extensive experiments on two public datasets reveal that the model outperforms existing few-shot knowledge graph completion methods, demonstrating its effectiveness.

Abstract

Dry bulk shipping plays a crucial role in intercontinental bulk cargo transport, with operators managing fleets to meet shippers’ transportation demand. A primary challenge for these operators is making optimal operational decisions about ship scheduling, routing, and sailing speed in the face of stochastic demand. We address this problem by developing a stochastic integer programming model designed to maximize revenue while maintaining high service levels for shippers. We quantify service levels for shippers using the probability of demand being fully satisfied. To solve this model, we introduce an innovative offline–online Lagrange relaxation framework. This framework leverages training data to determine the optimal Lagrange multiplier, which subsequently guides decision-making with test data. Numerical experiments show that our method closely matches the performance of Sampling Average Approximation (SAA) solutions while reducing computational time.

In this paper, we present a regularized dynamical system method for solving monotone inverse variational inequalities (IVIs) in infinite dimensional Hilbert spaces. It is shown that the corresponding Cauchy problem admits a unique strong global solution, whose limit at infinity exists and solves the given monotone IVI. Then by discretizing the dynamical system, we obtain a class of iterative regularization algorithms with relaxation parameters, which are strongly convergent under quite mild assumptions on the cost operator. Some simple numerical examples, including an infinite dimensional one, are given to illustrate the performance of the proposed algorithms.