Search

The aim of this study is to investigate the most appropriate constraint-handling mechanism in the sewer hydraulic design optimization problem. For this purpose, Dandelion Optimizer, which has not been used before to solve this problem, was used. Since the considered problem is a constrained optimization problem, constraint-handling mechanisms are required to solve this problem. In this study, the effects of various constraint-handling mechanisms on the performance of the algorithm were investigated in two different sized sewer networks. The constraint-handling mechanisms used are static penalty method, dynamic penalty method, dynamic penalty method with the superiority of feasible points, exponential dynamic penalty method, exponential dynamic penalty method with the superiority of feasible points, eclectic penalty method, adaptive penalty method and inverse tangent method. Solution quality, solution time and progress ratio were used as performance criteria. The most consistent result among the constraint-handling mechanisms discussed, was the adaptive penalty method. [ABSTRACT FROM AUTHOR]

This paper discusses the adaptive detection of extended radar targets buried in Gaussian clutter, assuming a diffuse multipath environment. The target return signal from each range cell is modeled as the sum of a deterministic data vector, which includes an unknown scaling factor representing the direct path component, and a randomly distributed data vector in a Gaussian distribution with unknown covariance matrix representing multipath echoes. During the design phase, it is assumed that the primary data covariance matrix falls within the vicinity of a sample covariance matrix that is devised from the secondary data set. The paper proposes a constraint Generalized Likelihood Ratio Test (GLRT) for the adaptive detection problem of extended radar targets in diffuse multipath environments, and conducts a performance analysis comparing the developed algorithm with well-known adaptive detectors in the literature. The results and performance analysis demonstrate that the proposed approach enhances the detection performance of extended radar targets in environments with diffuse multipath. Overall, this article provides valuable insights for improving the adaptive detection of extended targets in challenging environments, with potential applications in radar and sensing technologies. [ABSTRACT FROM AUTHOR]

The electrical energy transmission system lasted to expand in order to compensate for growing energy demand, leads to a variety of security problems. The decreasing system equivalent impedance in the result of the new investments integrated into the network causes to raise the short-circuit currents. It is a crucial problem, which taking remedial action is a necessity, for the short-circuit currents to be occurred exceed the circuit-breaker capacity. One method used to contain short-circuit currents is to perform the bus splitting technique. The bus splitting method is implemented through distributing feeders to appropriate buses and disconnect the zero-impedance line between buses in substations. Even if the short-circuit currents are restricted thanks to alterations in the network configuration by using bus splitting, the violation of the N-1 security and the growth of the power losses might occur. Increasing the system equivalent impedance in order to limit the short-circuit currents results in greater power losses. A balance between power losses and short-circuit currents should be found out due to the conflict of interest between these objectives. The objective function is created with short-circuit currents and active power losses and N-1 security is implemented as a constraint in this study. The constrained NSGA-II algorithm integrating Pareto-optimality concept into Genetic Algorithm is performed to solve the multi-objective bus splitting optimization problem. Mathematical model and algorithm created are implemented to the RTS 96-bus test system. The results show that the proposed method is successive in obtaining the secure topological frameworks of the network in terms of restricting the short-circuit currents, preventing the over increase of the power losses and maintaining the N-1 security. [ABSTRACT FROM AUTHOR]

Hospitals are among the most important institutions of today. For hospitals, efficient use of operating rooms is of great importance. Efficient use of operating rooms is a problem that needs to be solved. The operating room scheduling problem is a very complex problem with large number of constraints. This type of problem called as NP-Hard type problem. NP-Hard type problems do not consist of polynomial values. Therefore, the solution of these problems is very complex and difficult. Solutions consisting of polynomial values can be solved effectively with existing mathematical methods. However, more effective algorithms were needed to solve NP-hard type problems. As a result of the studies, many heuristic, meta-heuristic algorithms such as Genetic Algorithm, Particle Swarm Optimization, Simulated Annealing, Taboo Search Algorithm have been developed to solve the complexity of NP-Hard problems. In this article, the operating room scheduling problem solved with a hybrid genetic algorithm. In this solution, it shows how the algorithm affects the solution area in the changes in the number of surgeons, operating rooms and operating room reservations, which are among the operating room parameters. In the developed software, C# programming language has been preferred in order to provide comfortable use of the end user.

Bu çalışmada, ürünlerin dayanım ve performanslarının belirlenmesi için kullanılan hidrolik test sistemlerinin kontrolü için Model Öngörülü Kontrol (Model Predictive Control- MPC) tasarımı yapılmıştır Sistem kısıtlarının optimal kontrol kuralına dahil edilmesiyle test sistemi uygulamalarında karşılaşılan doyum problemleri için performans artımı sağlanması hedeflenmiştir. Bu amaçla, ilk olarak örnek sistem için sistemin doğrusal olmayan dinamik denklemleri oluşturulmuştur. Model çalışma noktası etrafında doğrusallaştırılarak, ivme durum değişkeni olacak şekilde durum uzayı modeli oluşturulmuştur. Elde edilen model, örnek sisteme ait model parametreleri kullanılarak, MPC içerisinde kullanılmak üzere ayrıklaştırılmıştır. MPC kuralı, yığın metodu (batch method) yardımı ile oluşturularak, kısıtlamalı optimal kontrol problemi arama algoritması yardımı ile çözülmüştür. Kontrol performansının tespiti amacı ile LQR ile karşılaştırmalı sayısal benzetim sonuçları sunulmuştur. Ayrıca sayısal benzetim testleri model belirsizliği ve ölçüm gürültüsü koşulları altında tekrarlanmış ve sonuçlar sunulmuştur. Elde edilen sonuçlar yorumlanmıştır ve gelecek çalışmalar için önerilerde bulunulmuştur.

Bu çalışmada, ürünlerin dayanım ve performanslarının belirlenmesi için kullanılan hidrolik test sistemlerinin kontrolü için Model Öngörülü Kontrol (Model Predictive Control- MPC) tasarımı yapılmıştır Sistem kısıtlarının optimal kontrol kuralına dahil edilmesiyle test sistemi uygulamalarında karşılaşılan doyum problemleri için performans artımı sağlanması hedeflenmiştir. Bu amaçla, ilk olarak örnek sistem için sistemin doğrusal olmayan dinamik denklemleri oluşturulmuştur. Model çalışma noktası etrafında doğrusallaştırılarak, ivme durum değişkeni olacak şekilde durum uzayı modeli oluşturulmuştur. Elde edilen model, örnek sisteme ait model parametreleri kullanılarak, MPC içerisinde kullanılmak üzere ayrıklaştırılmıştır. MPC kuralı, yığın metodu (batch method) yardımı ile oluşturularak, kısıtlamalı optimal kontrol problemi arama algoritması yardımı ile çözülmüştür. Kontrol performansının tespiti amacı ile LQR ile karşılaştırmalı sayısal benzetim sonuçları sunulmuştur. Ayrıca sayısal benzetim testleri model belirsizliği ve ölçüm gürültüsü koşulları altında tekrarlanmış ve sonuçlar sunulmuştur. Elde edilen sonuçlar yorumlanmıştır ve gelecek çalışmalar için önerilerde bulunulmuştur.

In this study, solutions for multi-objective constrained problems and their fixed weight linearly aggregated single-objective variants were obtained using the Pareto based multi-objective feasibility enhanced particle swarm optimization and single-objective approaches respectively. Comparisons involving three problems (two of which were highly constrained) revealed that optimizations performed using the multi-objective approach resulted in solutions that were also suitable for all single-objective criteria. With the multi-objective approach, objectives can be weighted after the optimization run and trade-offs can be performed without repeating the run. [ABSTRACT FROM AUTHOR]

In this study, Model Predictive Control (MPC) is designed for the control of hydraulic test systems that are used for determining the strength and performance of the product. It is aimed to increase the performance of the saturation problems faced during the test system applications while including the system constraints in the optimal control rule. For this purpose, the nonlinear dynamic equations are first obtained for the considered test system. The state space model is obtained by linearizing the model around the equilibrium point in a way that the design variable is considered to be acceleration. The obtained model is discretized for employing it in MPC by using the model parameters of the considered system. MPC rule is solved via constituting batch approach method through constrained optimal control problem search algorithm. The simulation result of the comparisons with LQR is presented with the aim of examining the control performance. In addition, numerical simulations are repeated under parametric model uncertainty and measurement noise conditions and results are presented. The obtained results are discussed, and future studies are suggested. [ABSTRACT FROM AUTHOR]