Generation of fuzzy rules with subtractive clustering

Jurnal Teknologi, 2005

Real-time road traffic data analysis is the cornerstone for the modern transport system. The real-time adaptive traffic signal control system is an essential part for the system. This analysis is to describe a traffic scene in a way similar to that of a human reporting the traffic status and the extraction of traffic parameters such as vehicle queue length, traffic volume, lane occupancy and speed measurement. This paper proposed the application of two-stage neural network in real-time adaptive traffic signal control system capable of analysing the traffic scene detected by video camera, processing the data, determining the traffic parameters and using the parameters to decide the control strategies. The two-stage neural network is used to process the traffic scene and decide the traffic control methods: optimum priority or optimum locality. Based on simulation in the traffic laboratory and field testing, the proposed control system is able to recognise the traffic pattern and enhance the traffic parameters, thus easing traffic congestion more effectively than existing control systems.

This paper presents the application of Adaptive Network Based Fuzzy Inference System ANFIS on speech recognition. The primary tasks of fuzzy modeling are structure identification and parameter optimization, the former determines the numbers of membership functions and fuzzy if-then rules while the latter identifies a feasible set of parameters under the given structure. However, the increase of input dimension, rule numbers will have an exponential growth and there will cause problem of “rule disaster”. Thus, determination of an appropriate structure becomes an important issue where subtractive clustering is applied to define an optimal initial structure and obtain small number of rules. The appropriate learning algorithm is performed on TIMIT speech database supervised type, a pre-processing of the acoustic signal and extracting the coefficients MFCCs parameters relevant to the recognition system. Finally, hybrid learning combines the gradient decent and least square estimation LSE of parameters network. The results obtained show the effectiveness of the method in terms of recognition rate and number of fuzzy rules generated.

Fuzzy logic is widely used in intelligent control because it has the advantage that it does not require a clear input. But in some applications, fuzzy logic has a weakness especially in choosing the most appropriate output. Ability enjin inferens owned fuzzy logic is not able to resolve quickly if the amount of data they handle a lot and require a process of rapid output with high accuracy. Therefore, the required help to other devices to engine inferens can make choices quickly and accurately. Here the role of genetic algorithm is needed. Development of parallel evolutionary algorithm for multi-objective problems involving the analysis of different paradigms in parallel processing and the appropriate parameters. However, not many publications that discuss this topic. Has done in-depth research on a new version of hybrid parallel genetic algorithm for optimization of the control system. Hybrid genetic algorithm approach is based on the simultaneous evolution of the two populations to find a solution that is focused on the goal to complete a separate constraint local optimum. While the first population is used to minimize the total distance traveled and the time it takes and the second population aims to minimize constraints in generating viable solutions. The hybrid algorithm combines the advantages of a genetic algorithm with clonal system in order not to get trapped into local optimum and rigor local search process. This paper presents a parallel hybrid genetic algorithm which combines genetic algorithm and clonal systems to optimize the fuzzy logic inference to intelligent control systems .

This paper presents applied research on fuzzy logic modeling to forecast the distribution of salinity in the coastal region of southern Tamil Nadu, India. Geoelectrical resistivity data has been used in this research, apart from nominal approach of salinity forecasting using geochemical data analysis. The data collected using vertical electrical sounding (VES) method was fed as an input for fuzzy inverting algorithm to evaluate true resistivity and thickness of subsurface layers. Inverted resistivity values have been subjected to fuzzy rule-based approach for salinity forecasting. Classifications have been made on the basis of linguistic variables with five linguistic terms of resistivity range using a triangular membership function of fuzzy logic. The purpose is to find the saltwater intrusion in the coastal region of Tuticorin district, Tamil Nadu, India. This research work reveals that fuzzy logic would be the effective tool for solving complex problems as well as enhancement in integrating multiple features necessary for the study. The results overlain in the surrounding regions which were mapped the threatening zones; hence, to mark pre-awareness in the regions, more rainwater harvesting system and avoidance of human anthropogenic activities need to be implemented.

NAFIPS 2006 - 2006 Annual Meeting of the North American Fuzzy Information Processing Society, 2006

In this paper, a subtractive clustering identification algorithm is introduced to model type-2 Takagi-Sugeno-Kang (TSK) fuzzy logic systems (FLS). The type-2 TSK FLS identification algorithm is an extension of the type-1 TSK FLS modeling algorithm proposed in [1, 2]. In the type-2 algorithm, subtractive clustering method is combined with least squares estimation algorithms to pre-identify a type-1 FLS form input/output data. Then using type-2 TSK FLS theory [3], expand the type-1 FLS to a type-2 TSK FLS. Minimum error models are obtained through enumerative search of optimum values for spreading percentage of cluster centers and consequence parameters. By doing so, fuzzy modeling of type-2 TSK FLS is found to be more effective than that of type-1 TSK FLS. Experimental results confirm the effectiveness of this method. A comparison of the Type-1 and -2 TSK FLSs is presented and the limitations of this method are discussed.

Information Sciences, 2014

Please cite this article in press as: Q. Ren et al., Type-2 fuzzy tool condition monitoring system based on acoustic emission in micromilling, Inform. Sci. (2013), http://dx.

Engineering Applications of Artificial Intelligence, 2011

This paper presents an experimental study for turning process in machining by using Takagi-Sugeno-Kang (TSK) fuzzy modeling to accomplish the integration of multi-sensor information and tool wear information. It generates fuzzy rules directly from the input-output data acquired from sensors, and provides high accuracy and high reliability of the tool wear prediction over a wide range of cutting conditions. The experimental results show its effectiveness and satisfactory comparisons relative to other artificial intelligence methods.

NAFIPS 2008 - 2008 Annual Meeting of the North American Fuzzy Information Processing Society, 2008

This paper presents the generalized type-2 Takagi-Sugeno-Kang (TSK) fuzzy logic system (FLS) in which the antecedent or consequent membership functions are type-2 fuzzy sets and the consequent part a first or higher order polynomial function. The architecture of the generalized type-2 TSK FLS and its inference engine are based on the Mendel's first order type-2 TSK FLS. The design method of high order system is an extension of the subtractive clustering based type-2 TSK FLS identification algorithm.

2006

Computer networks have experienced an explosive growth over the past few years and have become the targets for hackers and intruders. An intrusion detection system's main goal is to classify activities of a system into two major categories: normal activity and suspicious or intrusive activity. The objective of this paper is to expose ANFIS as a neuro-fuzzy classifier to detect intrusions in computer networks. Our experiments and evaluations were performed with the KDD Cup 99 intrusion detection dataset which is a version of the 1998 DARPA intrusion detection evaluation dataset prepared and managed by MIT Lincoln Laboratories. This paper shows that our proposed method can be effective in detecting various intrusions.

— Adaptive Neuro-Fuzzy Inference System (ANFIS) has become a popular tool in neuro-fuzzy modeling. However, since it includes many parameters needed to be set, its designing process is a complicated and time-intensive task for experimenters. To tackle this problem, in this paper we implement the Design of Experiment (DOE) technique to identify the significant parameters of ANFIS when it applies to the problem of stock price prediction. Using full factorial design, nine factors are considered as independent variables. Results identify six factors as statistically significant parameters, as well as four significant interactions between some independent variables.