Yazar "Celebi, Numan" seçeneğine göre listele

Listeleniyor 1 - 4 / 4
  • Küçük Resim
    Öğe
    Forecasting of daily natural gas consumption on regional basis in Turkey using various computational methods
    (Elsevier Science Sa, 2013) Taşpınar, Fatih; Celebi, Numan; Tutkun, Nedim
    It is widely accepted that natural gas is a clean energy source that can be used to meet energy demand for heating and industrial purpose among the fossil fuels and its usage remarkably increases in order to maintain a clean environment in many countries in the world. It is fact that this makes energy investment planning in a country or region highly important for suitable economic development as well as environmental aspect. Therefore, energy demand for various sectors should be estimated in the frame of short-term energy policy. For accurate estimation of short-term energy demand a limited number of computational methods are employed by using the 4 yearly measured natural gas consumption values. Among these methods, the ANN and time series are widely used for short-term estimation of natural gas consumption in Turkey's certain regions. In this study, multilayer perceptron the ANNs with time series approach is proposed to forecast short-term natural gas consumption. Meteorological data (moisture, atmospheric pressure, wind speed and ambient temperature) obtained from the regional gas distribution company and the local meteorology office in last 4 years to construct well-tuned algorithm. Although the number of data was small, the proposed algorithm works well to forecast the short-term natural gas consumption and produces encouraging and meaningful outcomes for future energy investment policy. (C) 2012 Elsevier B.V. All rights reserved.
  • Küçük Resim Yok
    Öğe
    An Improved Approach to Minimise Energy Cost in a Small Wind-Photovoltaic Hybrid System
    (Ieee, 2016) Tutkun, Nedim; Celebi, Numan
    Today recent developments in renewable technology such as wind and photovoltaic systems encourage people to generate electricity with low investment costs. In this manner, this may result in partial reduction in power demand hence import of primary sources may gradually decline as well as foreign dependence on energy. Furthermore this lets decrease power losses in transmission lines by reducing power generation capacity in power stations. The widely use of wind turbines and photovoltaic systems in residential and official buildings depends on being economic investment and operation costs. The unit investment cost shows no big difference in all the renewable systems but, to make operation cost lower the balance between production and consumption at any time interval must be maintained without shedding loads. For instance, in case of an hour time interval this balance should be maintained for 24 times in a day however this may not always be possible since sources of renewable energy are irregular. At some time intervals, a power boost is needed either by adding a power source to renewable system or by shedding loads according to their priorities but this is not desired case. In this study, the problem with 7 constraints was solved by a combinatorial optimization based the real-coded genetic algorithms to reduce operation cost to acceptable level and balance between generation and consumption at all the time intervals. The results are encouraging and meaningful for similar applications.
  • Küçük Resim
    Öğe
    A low-cost UAV framework towards ornamental plant detection and counting in the wild
    (Elsevier, 2020) Bayraktar, Ertugrul; Basarkan, Muhammed Enes; Celebi, Numan
    Object detection still keeps its role as one of the fundamental challenges within the computer vision territory. In particular, achieving satisfying results concerning object detection from outdoor images occupies a considerable space. In this study, in addition to comparing handcrafted feature detector/descriptor performance with deep learning methods over ornamental plant images at the outdoor, we propose a framework to improve the detection of these plants. Firstly, we take query images in the RGB format from the onboard UAV camera. Secondly, our model classifies the scene as a planting or an urban area. Thirdly, if the images are from planting area, thirdly, we filter the field according to the color and acquire only the green parts. Lastly, we feed the object detector model with the filtered area and obtain the category and localization of the plants as a result. In parallel, we also estimate the number of interested plants using the geometrical relations and predefined average plant size, then we verify the outputs of the object detector with this results. The conducted experiments show that deep learning based object detection methods overtake conventional feature detector/descriptor techniques in terms of accuracy, recall, precision, and sensitivity rates. The field classifier model, VGGNet, achieves a 98.17% accuracy for this task, whilst YoloV3 achieves 91.6% accuracy with 0.12 IOU for object detection as the best method. The proposed framework also improves the overall performance of these algorithms by 1.27% for accuracy and 0.023 for IOU. By specifying the limits thoroughly and developing task-dependent approaches, we reveal the great potential of our framework plant detection and counting in the wild consisting of basic image preprocessing techniques, geometrical operations, and deep neural network.
  • Küçük Resim
    Öğe
    Optimum Unit Sizing of Wind-PV-Battery System Components in a Typical Residential Home
    (Ieee, 2016) Tutkun, Nedim; Celebi, Numan; Bozok, Necati
    Optimal unit sizing of generation units in small offgrid systems is an important aspect to minimize total annual cost. This can be applied to wind, photovoltaic (PV), and hybrid wind-PV power generating systems to meet power demand for a residential home constructed at a specific site located in remote hilly areas in Turkey's northwest Black Sea coast where no grid extension is available. It is obvious that unit sizing of a standalone wind-PV system simply requires an optimization task to determine the optimal generation capacity and battery storage for a typical load profile. In optimization process, power demand, wind speed, and insolation rate are annually averaged hourly estimated values for the given site. The six combinations of PV alone, wind alone and hybrid wind-PV systems are optimally sized to both meet load demand and minimize total cost for 20-year lifetime projection using the real-coded genetic algorithms. The results indicated that a hybrid system with a 3 kW wind turbine and three 0.25 kW PV panel was the best configuration to reduce total annual cost to an affordable price for the specific example.