International Journal of Soft Computing and Engineering (IJSCE) ISSN: 2231-2307, Volume-1 Issue-5, November 2011 169 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: E0181091511/2011©BEIESP Abstract— This paper reports the Typical Load Profile of different types of consumers of distribution feeder which is based on clustering methods. Among many clustering methods, fuzzy c-means has been examined for determination of representative clusters because fuzzy logic is conceptually easy to understand. It is a well known clustering algorithm for typical load profiles determination. The results demonstrate that the proposed method is efficient for assigning Typical Load Profile (TLP) to the consumers. Moreover, the finding shows that the energy consumption can be clustered not only based on the load pattern but also load value. The results demonstrate that the proposed method is efficient for assigning TLP to the consumers. Key Words— Deregulation, Load Profile, Fuzzy Clustering Algorithm, Optimum Cluster, Probability, Neural network. I. INTRODUCTION Utilities would have a better trading and marketing strategies and the ability to design specific tariff options for the various classes of consumers in tune with real operation cost. Load profile of consumer makes more reliable and acceptable in this regards. Consumers can participate in the retail market and keep track of their actual power consumption and distributors could use such information for load management, distribution system planning and developed electricity tariffs, etc.The multiple participants of the electricity market need new business strategies for serving in competitive environments. So, accurate consumer’s information is badly needed to fulfillment their electricity demand. Detailed knowledge on consumer’s load consumption can facilitate distribution companies in determining specific tariff options for different type of consumers. Initially, the most efficient method of determining electricity consumption would be the direct monitoring. This can be achieved by installing time interval meters, quarter-hourly, half-hourly or hourly at each point of consumption. However, this approach is cost-prohibitive due to the equipment and processing costs. Furthermore, a significant amount of time would be needed to develop such a system. An alternative to this approach is by determining load profiles for consumers. Manuscript Received October 18, 2011. Md. Jahangir Hossain, Department of Electrical & Electronic Engineering, Khulna University of Engineering & Technology(KUET), Khulna, Bangladesh, Phone: +88041725375, Mobile No: +8801710862308 (e-mail: jhossain2k2@gmail.com ) A.N.M. Enamul Kabir, Department of Electrical & Electronic Engineering, Khulna University of Engineering & Technology, Khulna, Bangladesh, Mobile No.+8801714087197, (e-mail: anmenamulkabir@eee.kuet.ac.bd ) Md. Rafiqul Islam Department of Electrical & Electronic Engineering, Khulna University of Engineering & Technology, Khulna, Bangladesh, Mobile No.+8801714087330, (e-mail: rafiq043@yahoo.com ) Small eligible consumers have the possibility to change supplier even they don’t have appropriate metering equipment. Therefore, the formation of a feasible and cost effective way to determine consumers’ electricity consumption and deviations without installing new time interval meters is an important issue. A load profile based settlement can replace expensive installations of time-interval meters and it ensures a "fair" and accurate billing system and access to the retail market. Typical Load Profiles (TLP) representing coherent groups of consumers Moreover, the load profiles can be also very helpful in dealing with load management, distribution system planning, and state estimation, distribution transformer loss of life evaluation or distribution system restoration. There are two different approaches can be used for load profile based settlement; the area model and the category model [1]. The area model includes all those customers that are not metered on time interval basis within the geographic region covered by a network. On the other hand, the category model grouped customers with a similar load pattern into categories. Application of category model requires typical load profiles (TLP) representing coherent groups of consumers. Several papers present work regarding the establishment load profiles for a group of consumers [1], [2], [3]-[7], which are based on field measurements and can be divided into two groups. Typical feature for the first group is that TLPs are derived from load survey systems according to some predefined consumers groups [1], [2], [7]. The second group is obtained by identifying TLPs depending on the shape of the load curves [7], [9]. For this purpose, the use of pattern recognition methods is proposed. Limitations of the first TLP-determination approach are measurements performed over long time period and analyses how to define characteristic groups. Disadvantage of the second approach is formation of typical customer groups represented by TLP. The aim of the paper is to analyze the Measured Load Profiles (MLP) obtained from the Distribution System Operators (DSO), determine the TLPs and their allocation to the particular group of eligible consumers. Among the popular method are fuzzy clustering, artificial neural network (ANN) and self-organizing map (SOM) as described in [5-7]. Typical load profile has great potential for further improvement in distribution system applications. Md. Jahangir Hossain, A. N. M. Enamul Kabir, Md Mostafizur Rahman, Borhan Kabir, Md Rafiqul Islam Determination of Typical Load Profile of Consumers using Fuzzy C-Means Clustering Algorithm