Journal of Intelligent & Fuzzy Systems - Volume 7, issue 3

Authors: Dash, P.K. | Liew, A.C. | Satpathy, H.P.

Article Type: Research Article

Abstract: This paper presents a functional-link network based short-term electric load forecasting system for real-time implementation. The load and weather parameters are modelled as a nonlinear autoregressive moving average (ARMA) process and parameters of this model are obtained using the functional approximation capabilities of an auto-enhanced Functional Link net. Numerous and significant advantages accrue from using a flat net, including rapid quadratic optimisation in the learning of weights, simplification in the hardware as well as in computational procedures. The functional link net based load forecasting system accounts for seasonal and daily load characteristics as well as abnormal conditions, holidays and other …conditions. It is capable of forecasting load with a lead time of one hour to seven days. The adaptive mechanism with a nonlinear learning rule is used to train the network on-line. The results indicate that the functional link net based load forecasting system produces robust and more accurate load forecasts in comparison to simple adaptive neural network or statistical based approaches. Testing the algorithm with load and weather data for a period of two years reveals satisfactory performance with mean absolute percentage error (MAPE) mostly less than 2% for a 24-hour ahead forecast and less than 2.5% for a 168-hour ahead forecast. Show more

Citation: Journal of Intelligent and Fuzzy Systems, vol. 7, no. 3, pp. 209-221, 1999

Authors: Eng, Y.W. | Elangovan, S.

Article Type: Research Article

Abstract: A new type of fuzzy logic power system stabilizer is proposed in this paper. It is constructed using a five-layered neural fuzzy network architecture based on α-level fuzzy sets. The workability of this neural fuzzy power system stabilizer is first demonstrated using regularly spaced and triangular fuzzy sets. Then, it is shown that the fuzzy sets can be tuned so as to improve the damping performance of the stabilizer. To obtain the desired output for backpropagation to be applied, the network output is altered at a randomly chosen time instant. The altered output is then taken as the desired output …if the stabilizer performs better than without the alteration. Simulation results show that the performance of the neural fuzzy network can be improved within 30 training cycles. Show more

Citation: Journal of Intelligent and Fuzzy Systems, vol. 7, no. 3, pp. 223-238, 1999

Authors: Sawaragi, Tetsuo | Tani, Naoki | Katai, Osamu

Article Type: Research Article

Abstract: This paper presents a method for concept formation of a personal learning apprentice (PLA) system that attempts to capture users' internal conceptual structure by observing interactions between user and system. The primary goal of a PLA system is to identify the users' cognition that underlies the taking of action. This is based on the capability to reconstruct internal concepts as behavior-shaping constraints by observing operations as well as the information presented by the system. Our proposed algorithm comprises two processes; adaptive feature selection and GA-based feature discovery. The former selects the essential attributes out of a provided set of …attributes that may initially be either relevant or irrelevant, and the latter constructs new attributes using genetic algorithms applied to a set of elementary features logically represented in a disjunctive normal form. Our method can be applied to artificial data as well as to a data set obtained from human-machine interactions observed during operation of a simulator of a generic dynamic production process. Show more

Citation: Journal of Intelligent and Fuzzy Systems, vol. 7, no. 3, pp. 239-256, 1999

Authors: Senjyu, Tomonobu | Higa, Shuzo | Uezato, Katsumi

Article Type: Research Article

Abstract: Load forecasting has always been an essential part of efficient power system planning and operations. In this paper, we propose the fuzzy logic approach for future load curve forecasting based on similarity. The proposed approach has the advantage of dealing with not only the nonlinear part of forecasted load curve but also with load curve forecasting without distinction of day type. In addition to the above mentioned advantages, the proposed method is useful in situations where accurate forecasting models are difficult to design. The fuzzy logic approach is used to modify the load curves on similar days in order …to shape the next day load curve. The suitability of the proposed approach is illustrated through an application to actual load data of the Okinawa Electric Power Company in Japan. Show more

Citation: Journal of Intelligent and Fuzzy Systems, vol. 7, no. 3, pp. 257-265, 1999

Authors: Cho, Young-Bin | Gweon, Dae-Gab

Article Type: Research Article

Abstract: Aritificial neural networks may be used for a function approximator which includes not only deterministic but also probabilistic model. Conditional variance estimation using a neural network is a good example of probabilistic model approximation, because conditional variance, which is a function of input variable, is an important parameter to describe a Gaussian probabilistic model. The majority of learning algorithms are based on a concept of likelihood maximization or expectation maximization method. This article presents an alternative learning algorithm based on a different concept for a multilayer perceptron. The proposed variance learning algorithm can be regarded as a kind of modified …delta rule, where delta is determined by an iterative estimation algorithm, which is also proposed in this article. The proposed learning algorithm has stochastic property because the delta is stochastically determined by the estimation algorithm. Relationships of delta to the transient and steady state of the learning process are also stochastic. First, the iterative variance estimation algorithm is explained. Second, the transient state behavior is investigated to have an insight into convergence and stability properties with respect to delta. Third, the steady state analysis is described to show the relationship of delta to steady state error bound. Theoretical analysis on steady state behavior produces analytic formula for steady state error bound of the variance learning algorithm in terms of the delta. Finally, multilayer perceptron using the proposed learning algorithm is simulated for the demonstration of variance estimation. Show more

Citation: Journal of Intelligent and Fuzzy Systems, vol. 7, no. 3, pp. 267-282, 1999

Authors: Diab, Hassan B. | Saade, Jean J.

Article Type: Research Article

Abstract: This paper presents the use of fuzzy inference methodologies in the prediction of the weather, as a viable alternative to the classical complex, costly and time-consuming weather prediction techniques. Basic weather elements which affect weather characteristics and which underlie the factors used in advanced weather prediction methods are fuzzified. They are then used in fuzzy weather prediction models based on fuzzy inferences. These models are simulated and the crisp results obtained using different existing and recently developed defuzzification strategies are compared with the actual weather data. Further, conclusive comments are provided regarding the effectiveness of the introduced models and the …avenues which need to be explored in order to make them increasingly efficient. Show more

Citation: Journal of Intelligent and Fuzzy Systems, vol. 7, no. 3, pp. 283-305, 1999

Authors: Falkner, C.M. | Heck, B.S.

Article Type: Research Article

Abstract: This article compares the performance of six output feedback control schemes used as power system stabilizers. The power system provides a good testbed for a comparison of modern control techniques since the power system has many nonlinearities and can be subjected to many perturbations. In addition, the power system model (even for a single generator-infinite bus system) can be of a size large enough to allow for nontrivial control design. The control schemes used in this article are fuzzy control, sliding mode control, model reference adaptive control, two Lyapunov-based robust control methods (one linear, the other nonlinear) and a standard …linear PSS. Every effort was made to tune the controllers for similar nominal behavior. This was done so the controllers' ability to handle perturbations could be examined objectively. It is found in this article that the fuzzy and linear robust controllers had the best overall behavior with the least amount of control effort. Show more

Citation: Journal of Intelligent and Fuzzy Systems, vol. 7, no. 3, pp. 307-324, 1999