عنوان مقاله
ماژول ارزیابی امنیت استاتیک آنلاین با استفاده از شبکه‌های عصبی مصنوعی
فهرست مطالب
مقدمه
شاخص امنیت ترکیبی
ماژول ارزیابی امنیت استاتیک آنلاین با استفاده از ANN
سیستم تست و نتایج شبیه سازی
نتیجه گیری
بخشی از مقاله
شبکه تابع بنیادی شعاعی
RBFN یک کلاس از شبکه‌های عصبی پیشخور و شامل یک لایه ورودی ، یک لایه مخفی، و یک لایه خروجی بودند. شبکه قادر به اجرای نگاشت غیرخطی خصیصه‌های ورودی در خروجی بود. لایه مخفی شامل نورون‌هایی با تابع فعال‌سازی گاووس بود. در طول آموزش، همه متغییرهای ورودی به نورن‌ها در طول لایه‌مخفی به طور مستقیم از طریق ارتباطات درونی با وزن واحد خورانده می‌شود و تنها وزن‌ها بین لایه مخفی و لایه یاد گرفته می‌شدند. بنابراین، RBFN همگرایی سریعتری را نسبت به MLFFN معمولی ارائه می‌دهد.
کلمات کلیدی:
Online Static Security Assessment Module Using Artificial Neural Networks Sunitha R, Member, IEEE, R. Sreerama Kumar, Senior Member, IEEE, and Abraham T. Mathew, Senior Member, IEEE Abstract—Fast and accurate contingency selection and ranking method has become a key issue to ensure the secure operation of power systems. In this paper multi-layer feed forward artificial neural network (MLFFN) and radial basis function network (RBFN) are proposed to implement the online module for power system static security assessment. The security classification, contingency selection and ranking are done based on the composite security index which is capable of accurately differentiating the secure and non-secure cases. For each contingency case as well as for base case condition, the composite security index is computed using the full Newton Raphson load flow analysis. The proposed artificial neural network (ANN) models take loading condition and the probable contingencies as the input and assess the system security by screening the credible contingencies and ranking them in the order of severity based on composite security index. The numerical results of applying the proposed approach to IEEE 118-bus test system demonstrate its effectiveness for online power system static security assessment. The comparison of the ANN models with the model based on Newton Raphson load flow analysis in terms of accuracy and computational speed indicate that the proposed model is effective and reliable in the fast evaluation of the security level of power systems. The proposed online static security assessment (OSSA) module realized using the ANN models are found to be suited for online application. Index Terms—Composite security index, contingency screening and ranking, multi-layer feed forward neural network, online static security assessment, radial basis function network. I. INTRODUCTION MAINTAINING system security is an important requirement in the operation of a power system. Power system security assessment is the analysis performed to determine whether, and to what extent, a power system is reasonably safe from serious interference to its operation [1]. Three major functions involved in power system security assessment are system monitoring, contingency analysis and security control. System monitoring provides up-to-date information of bus voltages, currents, power flows and the status of circuit breaker through the telemetry system so that operators can easily identify the system in the normal state or in abnormal condition. Manuscript received November 11, 2012; revised February 13, 2013 and April 25, 2013; accepted May 28, 2013. Date of publication June 26, 2013; date of current version October 17, 2013. Paper no. TPWRS-01264-2012. S. R and A. T. Mathew are with the Department of Electrical Engineering, National Institute of Technology Calicut, Kerala 673601, India (e-mail: rsunitha@nitc.ac.in; atm@nitc.ac.in). R. S. Kumar is with King Abulaziz University Jeddah, Jeddah 21589, Saudi Arabia (e-mail: sreeram@nitc.ac.in). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TPWRS.2013.2267557 On the other hand, contingency analysis is carried out to evaluate the outage events in power system and it is a critical part in security assessment and involves critical contingency screening and ranking [2]. If the system is found to be in insecure, security control will take the preventive or corrective control actions to ensure the system back to secure condition. Static security assessment checks the degree of satisfaction for all relevant static constraints of post contingency steady states and is needed to solve a large set of nonlinear algebraic equations [3] for N and N-1 system conditions. The security analysis becomes more complex and difficult, as these studies need to be performed online for it to be more effective. Conventionally these analysis are performed offline, since the simulation take significant computation time. The large computational burden has been the main impediment in preventing the security assessment from online use [4]. On the other hand there is a pressing need for more accurate and powerful tool for security assessment [5]. The work presented in this paper was motivated by the attempt to significantly reduce the computation time required for security assessment so that the analysis can be converted from offline to online use, in order to assist the grid operators in their real time controller analysis. The trend towards deregulation has forced the modern utilities to operate their systems closer to security boundaries. This has fueled the need for faster and more accurate methods of security assessment [6]. The overall computational speed and accuracy of an online security assessment depends on the effectiveness of contingency screening and ranking method, the objective of which is to identify the critical contingencies among a list of possible contingencies.The contingency selection and rankingis conventionally performed by various schemes by computing a scalar performance index (PI) derived from DC or fast decoupled load flow solution for each contingency [7]. These methods generally employ a quadratic function as the performance index. This makes the contingency ranking prone to masking problems, where a contingency with many small limit violations is ranked equally with the one in which there are only a few large limit violations. Also, the selection of weighting factors in the performance index is found to be a difficult task, as it should be chosen based on both the relative importance of buses and branches and the power system operating practice [8], [9]. In addition, majority of the performance indices do not provide an exact differentiation between the secure and non-secure states. The performance indices were traditionally calculated separately for line flows and bus voltages, as the overall performance index defined as the sum or weighted-sum of the scalar performance indices for bus voltages and the line flows could not provide accurate results [10].
