• Korean Journal of Ecology and Environment
• /
• v.33 no.3 s.91
• /
• pp.230-243
• /
• 2000

A large-scale community data were patterned by utilizing an unsupervised learning algorithm in artificial neural networks. Data for benthic macroinvertebrates in streams of South Korea reported in publications for 12 years from 1984 to 1995 were provided as inputs for training with the Kohonen network. Taxa included for the training were 5 phylum, 10 class, 26 order, 108 family and 571 species in 27 streams. Abundant groups were Diptera, Ephemeroptera, Trichoptera, Plecoptera, Coleoptera, Odonata, Oligochaeta, and Physidae. A wide spectrum of community compositions was observed: a few tolerant taxa were collected at polluted sites while a high species richness was observed at relatively clean sites. The trained mapping by the Kohonen network effectively showed patterns of communities from different river systems, followed by patterns of communities from different environmental disturbances. The training by the proposed artificial neural network could be an alternative for organizing community data in a large-scale ecological survey.

• Journal of Korean Society on Water Environment
• /
• v.33 no.3
• /
• pp.291-301
• /
• 2017

This study was carried out to investigate distribution on the alien plants in the Gyeong-in Ara Waterway. The alien plants were a total of 82 taxa: 17 families, 63 genera, 80 species, and 2 varieties. This number corresponds to 25.5% of alien plants identified in Korea. The proportion of alien plants in every year was increased from upstream to downstream. As the result of the analysis on vegetation stratification, bank of waterside was covered with artificial materials preventing existence of vegetation, and had step-type cross section. Floodplain was composed of waterfront area. An array of vegetation was not typical dispersion, and terrestrial and alien plants were dominated the Gyeong-in Ara Waterway. Evaluation of naturalness based on the vegetation stratification showed grade 3 or 4. In order to solve a problem, method and level of restoration should be decided based on the result of diagnostic assessment. Therefore, we need to restore the step-type cross section as pool type one. From waterside to bank in this waterway, we recommend to introduce natural plants by imitating reference species composition. Since, an invasion of alien plants is expected to be accelerated due to the continuous artificial disturbance, we recommend to quantitative investigation on the invasion of alien plants and monitoring on the change of distribution.

• Journal of Korean Society on Water Environment
• /
• v.32 no.2
• /
• pp.205-215
• /
• 2016

In order to improve water quality in the Paldang Lake, a riparian vegetation belt, treatment wetland, and artificial floating island were designed for introduction in the upland field, the estuary of tributaries, and the section of water facing mountainous land, respectively. We synthesized vegetation information collected from a reference river and found that herbaceous, shrubby, and tree vegetation zones tended to be dominated by Phragmites japonica, Phalaris arundinacea, etc.; Salix gracilistyla, S. integra, etc.; and S. koreensis, S. subfragilis, and Morus alba, respectively. In our plan, the herbaceous vegetation zone, which is established on floodplains with a high frequency of disturbance, will be left in its natural state. A shrubby vegetation zone will be created by imitating the species composition of the reference river in the ecotone between floodplain and embankment. A tree vegetation zone will be created by imitating species composition on the embankment slope. In the treatment wetland, we plan to create emerged and softwood plant zones by imitating the species composition of the Zizania latifolia community, the Typha orientalis community, the P. communis community, the S. integra community, and the S. koreensis community. The floating island will be created by restoring Z. latifolia and T. orientalis for water purification purposes.

• 유체기계공업학회:학술대회논문집
• /
• 2002.12a
• /
• pp.258-263
• /
• 2002

A numerical analysis using 2-D unsteady compressible program is conducted to explain characteristics of rotating stall such as rotating speed and number of stall cells in an one-stage axial compressor. Unlike an axial compressor which has only a rotor, in one-stage axial compressor a rotating stall is generated by rotor/stator interaction and tack pressure rising without any artificial disturbance and modeling. As a back pressure is raised, the separation of suction side at blades is increased uniformly, but because of the discrepancy of blockage effect by stator, the disturbances are generated to form a stall cell. Once the stall cell is formed, regularly the stall cell are rotating through rotor blades. When the speed of rotor is design speed the rotating speed of stall cell is $83.3\%$ of rotor rotating speed. When the speed of rotor is $80\%$ of design speed, the speed of rotating stall is $88.2\%$ of rotor speed. The number of generated stall cell are also varied for rotor speed and back pressure.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.58 no.3
• /
• pp.249-256
• /
• 2009

The widely used control theory based design of PI family controllers fails to perform satisfactorily under parameter variation, nonlinear or load disturbance. In high performance applications, it is useful to automatically extract the complex relation that represent the drive behaviour. The use of learning through example algorithms can be a powerful tool for automatic modelling variable speed drives. They can automatically extract a functional relationship representative of the drive behavior. These methods present some advantages over the classical ones since they do not rely on the precise knowledge of mathematical models and parameters. The paper proposes high performance speed and current control of synchronous reluctance motor(SynRM) drive using adaptive learning mechanism-fuzzy neural network (ALM-FNN) and fuzzy logic control (FLC) controller. The proposed controller is developed to ensure accurate speed and current control of SynRM drive under system disturbances and estimation of speed using artificial neural network(ANN) controller. Also, this paper proposes the analysis results to verify the effectiveness of the ALM-FNN, FLC and ANN controller.

• Journal of Environmental Science International
• /
• v.18 no.2
• /
• pp.153-167
• /
• 2009

Riparian vegetation gives basic habitats for animal and other biological organisms. It is one of essential landscapes in river ecosystem. This floral study was conducted to give basic information for management of the Geum river ecosystem. We discovered total 69 plant communities out of 40 sites studied and mean 8 plant communities were analysed per study site. Salix koreensis and Salix subfragilis community were dominant of all communities at tree layer and Salix gracilistyla and Rubus parvifolius were dominant communities at shrub layer. Phalaris arundinacea, Phragmites japonica, Miscanthus sacchariflorus and Artemisia princeps var. orientalis appeared most as perennial grasses and Humulus japonicus, Persicaria thunbergii and Persicaria hydropiper appeared most as annual grasses. Robinia pseudoacacia, Amorpha fruticosa and Bidens frondosa were frequently present as exotic species. Grass communities were higher than tree communities because of human disturbance by artificial banks. The number of species was 182 appeared as 53 families 144 species, 34 varieties and 4 forma and mean species number was 25.5 per site. The most abundant site was Yanggang bridge site (14 vegetation association, 49 species) and the least abundant site was Namdaechon, Muju site (3 vegetation association, 12 species). We suggest that continuous riparian vegetation monitoring should be carried out to manage water environment ecologically.

• Proceedings of the KIEE Conference
• /
• 1992.07a
• /
• pp.225-230
• /
• 1992

In this paper, a new learning fuzzy logic controller(LFLC) is presented. The proposed controller is composed of the main control part and the learning part. The main control part is a fuzzy logic controller(FLC) based on linguistic rules and fuzzy inference. For the learning part, artificial neural network(ANN) is added to FLC so that the controller may adapt to unknown plant and environment. According to the output values of the ANN part, which is learned using error back-propagation algorithm, scale factors of the FLC part are determined. These scale factors transfer the range of values of input variables into corresponding universe of discourse in the FLC part in order to achieve good performance. The effectiveness of the proposed control strategy has been demonstrated through simulations involving the control of an unknown robot manipulator with load disturbance.

• Korean Journal of Environment and Ecology
• /
• v.12 no.4
• /
• pp.317-330
• /
• 1999

To study the actual vegetation and vegetation structure of Tongyoung district and Yokjido (Island) in Hallyo-Haesang National Park, thirty seven plots were surveyed in July,1998. The vegetation structure of the surveyed plots were dissimilar in virtue of geographical isolation and artificial disturbance. The actual vegetation were classified into five group: Pinus thunbergii forest, P. thunbergii-P. densiflora- deciduous broad-leaved forest, deciduous broad-leaved forest, P. thunbergii-P. densiflora-evergreen broad-leaved forest, evergreen broad-leaved forest. P. thunbergii forest was a major vegetaion type of the surveyed area and evergreen broad-leaved forests were distributed at Ch'ubongdo, Yongch'odo, Pijindo, Yokjido and Youjado as a small patch. Successional sere shall be assumed from Pinus thunbergii, Pinus densiflora, and deciduous broad leaved trees to Neolitsea sericea, Cinnamomum japonicum and Persea thunbergii via Castanopsis sieboldii.

• Korean Journal of Environmental Biology
• /
• v.27 no.2
• /
• pp.176-182
• /
• 2009

We surveyed the total 30 sites in Gilgol, Nae Sorak for the patterns of the natural regeneration of the korean fir in connection with the characteristic of the stand. The natural regeneration in a forestrical meaning is not found in natural forest but mainly in the anthropogenically already disturbed area. It may be inferred from this fact that the anthropogenic disturbance can give more better condition for a natural regeneration. The regeneration phase in the natural forest is not so distinctive as the artificial forest, and shows itself transiently. And it may be also said that a shift in generation is only a part of the change of the stand phase. The concept of the natural regeneration will lose its meaning, in so far as the patch dynamic determines the phase of the stand, and the main tree species are shade-tolerant abies.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.21 no.3
• /
• pp.83-98
• /
• 2018

This study was conducted on rare plants distributed in Chungnam area from April, 2012 to November, 2017. The rare plant populations that are emerging are: Utricularia pilosa, Jeffersonia dubia, Iris ruthenica, Tipularia japonica, Calanthe discolor, Parasenecio pseudotaimingasa, Ranunculus kazusensis, Berchemia racemosa, Pogonia minor and Glehnia littoralis. According to the criteria of the International Union for Conservation of Nature (IUCN), plant resources in the rare species category were one important type (CR), three hazardous species (EN) and six vulnerable species (VU). As a result of analyzing characteristics of ecological environment and threats, five types of habitat types, two aquatic areas, two types of flatland and coastal sand dune distribution types were analyzed. The decreasing tendency of the rare plant populations in the surveyed area is judged to be artificial disturbance and habitat destruction rather than climate or environmental change. Considering the characteristics of habitat, conservation measures should be prepared for each population.