Article
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

In this paper an evolutive algorithm is used to train an adaptative-network-based fuzzy inference system (ANFIS), particularly a genetic algorithm (GA). The GA is able to train the antecedent and consequent parameters of an ANFIS, which is used for energy load profile forecasting in an automated factory. This load forecasting is useful to support an intelligent energy management system (IEMS), which enables the user to optimize the energy consumptions by means of getting the optimal work points, scheduling the production according to these points, etc. The proposed training algorithm showed excellent results with complex plants like industrial energy consumers in the user-side, where the randomness of the loads is higher than in utility loads. Real data from an automated car factory were used to test the presented algorithms. Appropriated results were obtained.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... RMSE is a frequently-used method for forecasting technique evaluation. Machine learning models usually minimize RMSE to obtain [13]. Therefore, the fitness of the chromosome should be inversely proportional to RMSE. ...
... Table 2 summarises some studies which have used metaheuristic methods for training the ANFIS network. AWPSO EKF Sargolzaei et al. [38] PSO PSO Turki, Bouzaida [39] PSO PSO Rini, Shamsuddin [40] PSO PSO Karaboga, Kaya [41] ABC ABC Soto, Melin [42] GA LSE Cardenas, Garcia [43] GA GA ...
Article
Full-text available
This paper presents a control-oriented neuro-fuzzy model of brazed-plate evaporators for use in organic Rankine cycle (ORC) engines for waste heat recovery from exhaust-gas streams of diesel engines, amongst other applications. Careful modelling of the evaporator is both crucial to assess the dynamic performance of the ORC system and challenging due to the high nonlinearity of its governing equations. The proposed adaptive neuro-fuzzy inference system (ANFIS) model consists of two separate neuro-fuzzy sub-models for predicting the evaporator output temperature and evaporating pressure. Experimental data are collected from a 1-kWe ORC prototype to train, and verify the accuracy of the ANFIS model, which benefits from the feed-forward output calculation and backpropagation capability of the neural network, while keeping the interpretability of fuzzy systems. The effect of training the models using gradient-descent least-square estimate (GDLSE) and particle swarm optimisation (PSO) techniques is investigated, and the performance of both techniques are compared in terms of RMSEs and correlation coefficients. The simulation results indicate strong learning ability and high generalisation performance for both. Training the ANFIS models using the PSO algorithm improved the obtained test data RMSE values by 29% for the evaporator outlet temperature and by 18% for the evaporator outlet pressure. The accuracy and speed of the model illustrate its potential for real-time control purposes.
... An adaptive neuro-fuzzy inference system (ANFIS) is a fuzzy inference model that uses ANN to learn rules from dataset for predictions. Findings show that ANFIS has recorded successes in predicting outcomes of events, systems or processes in various areas of life [3][10] [12][13] [15][18] [19] [20][21] [22]. Despite all these recorded successes, criticisms arise as to the degree to which its predictions are reliable. ...
Article
Full-text available
Research findings established the usefulness of Adaptive neuro-fuzzy inference system (ANFIS) in predicting outcomes of events, processes or systems from their numerical representations (datasets). However, its effectiveness in terms of prediction accuracy and complexity depends on some factors specifically its set of rules that models data behavior, and tune-able parameters (antecedent, consequent and rule base). Although several promising non-heuristic and meta-heuristic optimization techniques like gradient descent, Ant Colony, Genetic Algorithm, Particle Swarm, and Invasive Weed have been proposed to improve on the prediction accuracy of ANFIS through proper optimization of its tune-able parameters, but absolutely no work has applied clustering ensemble to improving on ANFIS for better accuracy especially on rules improvement. Therefore, in this paper, we propose an improved ANFIS that uses an agglomerative-based clustering ensemble of fuzzy c-means to help extract rules from a given dataset that represent a process, event or system. The ensemble uses probability trajectories of random walk processes on base clustering partitions of the dataset so as to refine direct co-association relationships or links among data objects of the base partitions thereby improving on these relationships that later metamorphosed to fuzzy rules and finally, a better accuracy. Comparative analysis of our proposed improved ANFIS alongside with the conventional ANFIS using two standard petrophysical datasets for lithology prediction shows that the proposed ANFIS is better than the conventional ANFIS in terms of prediction model accuracy and stability although, both maintain the same architecture in terms of the number of rules and tune-able parameters as well as the training methods.
... Hasan et al. (2011a) trained ANFIS by GA to predict the mislaid reading data. Cárdenas et al. (2011) determined the premise and consequence parameters of ANFIS by GA for energy load profile forecasting. Suja and Raglend (2012) optimized the membership function of fuzzy system by GA. ...
Article
Full-text available
In the structure of ANFIS, there are two different parameter groups: premise and consequence. Training ANFIS means determination of these parameters using an optimization algorithm. In the first ANFIS model developed by Jang, a hybrid learning approach was proposed for training. In this approach, while premise parameters are determined by using gradient descent (GD), consequence parameters are found out with least squares estimation (LSE) method. Since ANFIS has been developed, it is used in modelling and identification of numerous systems and successful results have been achieved. The selection of optimization method utilized in training is very important to get effective results with ANFIS. It is seen that derivate based (GD, LSE etc.) and non-derivative based (heuristic algorithms such us GA, PSO, ABC etc.) algorithms are used in ANFIS training. Nevertheless, it has been observed that there is a trend toward heuristic based ANFIS training algorithms for better performance recently. At the same time, it seems to be proposed in derivative and heuristic based hybrid algorithms. Within the scope of this study, the heuristic and hybrid approaches utilized in ANFIS training are examined in order to guide researchers in their study. In addition, the final status in ANFIS training is evaluated and it is aimed to shed light on further studies related to ANFIS training.
... The prior research used ANFIS for solving classification problem, while the later one chose ANFIS to predict the values of Longitude and Altitude. Cardenas et al. [23] trained the parameters of ANFIS using GA alone for energy load forecast. ...
... The prior research used ANFIS for solving classification problem, while the later one chose ANFIS to predict the values of Longitude and Altitude. Cardenas et al. [23] trained the parameters of ANFIS using GA alone for energy load forecast. ...
Conference Paper
Full-text available
Fuzzy Neural Networks (FNNs) with the integration of fuzzy logic, neural networks and optimization techniques have not only solved the issue of " black box " in Artificial Neural Networks (ANNs) but also have been effective in a wide variety of real-world applications. Despite of attracting researchers in recent years and outperforming other fuzzy inference systems, Adaptive Neuro-Fuzzy Inference System (ANFIS) still needs effective parameter training and rule-base optimization methods to perform efficiently when the number of inputs increase. Moreover, the standard gradient based learning via two pass learning algorithm is prone slow and prone to get stuck in local minima. Therefore many researchers have trained ANFIS parameters using metaheuristic algorithms however very few have considered optimizing the ANFIS rule-base. Mostly Particle Swarm Optimization (PSO) and its variants have been applied for training approaches used. Other than that, Genetic Algorithm (GA), Firefly Algorithm (FA), Ant Bee Colony (ABC) optimization methods have been employed for effective training of ANFIS networks while solving various problems in the field of business and finance.
... The fuzzy rules consist of several if-then This work was supported in part by the Seventh Framework Programme under the FP7-ICT-2011-7 / 288102 Research Project. cases, which define how the output must be calculated for a specific value of its inputs [11]. The particular type of the AFNIS' fuzzy system is based on Takagi-Sugeno rules, despite the fact that the fuzzy systems can have different kind of inference methods. ...
Conference Paper
Full-text available
This paper presents an energy optimization methodology applied on industrial plants with multiple energy carriers. The methodology combines an adaptive neuro-fuzzy inference system to calculate the short-term load forecasting of a plant, and the sequential quadratic programming algorithm to optimize its energy flow. Furthermore, the mathematical models of the plant's equipment are considered into the optimization process, in order to calculate the dynamic system response and the equipment's inertias. The final algorithm optimizes the operation of the plant in order to satisfy the energy demand, minimizing several optimization criteria. The methodology has been tested and evaluated in an automotive factory plant using real production and consumption data. Keywords—energy optimization; energy hub; energy prediction; adaptive neuro-fuzzy inference system; sequential quadratic programming algorithm.
... The training inputs are also called energy drivers and are variables that can affect the output, such as, in case of the energy consumptions: the daily production, the climatic data, the day of the week, etc [14]. The membership functions of the system are the functions that define the fuzzy sets [15][16]. The fuzzy rules have a form of if-then rule and define how the output must be for a specific value of membership of its inputs. ...
Article
Full-text available
This document presents an energy forecast methodology using Adaptive Neuro-Fuzzy Inference System (ANFIS) and Genetic Algorithms (GA). The GA has been used for the selection of the training inputs of the ANFIS in order to minimize the training result error. The presented algorithm has been installed and it is being operating in an automotive manufacturing plant. It periodically communicates with the plant to obtain new information and update the database in order to improve its training results. Finally the obtained results of the algorithm are used in order to provide a short-term load forecasting for the different modeled consumption processes.
Article
Full-text available
In this paper, a new version of IPO called (VLIPO- Variable Length Inclined Planes System Optimization Algorithm) has been provided primarily. Then, an efficient tool for simultaneous design and training of an ANFIS (adaptive neuro-fuzzy inference system) has been proposed using the mentioned algorithm. It should be noted that till the present time, related research has been only dealing with finding type and location of membership functions or proposing a method of training such networks. Length of standard versions of heuristic algorithms has been mainly the reason for not specifying the type and location of membership and training an ANFIS network simultaneously. For the same reason, first, a new version of the IPO is introduced with such factors being variable. Then, such capability is used for specifying the type and location of membership functions and simultaneous training of an ANFIS classifier. It goes without saying that the idea of making variations in search factors could be also implemented in other heuristic methods. Some of them have been reported in the research. Therefore, the presented idea in the paper may be applied to other heuristic methods and used in the design and simultaneous training of an ANFIS. So, the results from the comparison made between implementing the proposed method and other methods of which a version with a variable length has been previously reported (PSO, ACOR, DE, and GA) have been presented in several well-known databases. The results showed the better execution of ANFIS classifier designed by VLIPO compared to other heuristic methods.
Article
Premise and consequent parameters of ANFIS are optimized by an optimization algorithm in its training process. A successful optimization algorithm should be utilized for an effective training process. In this study, an adaptive and hybrid artificial bee colony (aABC) algorithm, which is one of the variants of ABC algorithm, is employed in ANFIS training. aABC algorithm uses arithmetic crossover and adaptive neighborhood radius in the solution generating mechanism. aABC algorithm has gained the ability to obtain fast convergence and quality solution with these two control parameters. ANFIS is trained using aABC algorithm to obtain better solutions according to standard ABC algorithm. Firstly, five nonlinear static test systems are utilized for performance analysis of aABC algorithm. With aABC algorithm, performance increases up to about 16% compared to standard ABC algorithm. At the same time, better convergence is obtained in all examples. Wilcoxon signed rank test is applied to determine significance of the results. In addition, the results reached by aABC algorithm are compared with GA, PSO, HS algorithms and more effective results are found with aABC algorithm. As a result, it is seen that aABC algorithm is more successful than ABC, GA, PSO and HS in ANFIS training for identification of nonlinear static systems. Secondly, ANFIS is also trained by utilizing aABC algorithm for solving a real-world problem. Estimating number of foreign visitors coming to Turkey is selected as a real-world problem. The results obtained are compared standard with standard ABC algorithm, and more successful results are found by aABC algorithm.
Chapter
The intermittency of wind remains the greatest challenge to its large scale adoption and sustainability of wind farms. Accurate wind power predictions therefore play a critical role for grid efficiency where wind energy is integrated. In this paper, we investigate two hybrid approaches based on the genetic algorithm (GA) and particle swarm optimisation (PSO). We use these techniques to optimise an Adaptive Neuro-Fuzzy Inference system (ANFIS) in order to perform one-hour ahead wind power prediction. The results show that the proposed techniques display statistically significant out-performance relative to the traditional backpropagation least-squares method. Furthermore, the hybrid techniques also display statistically significant out-performance when compared to the standard genetic algorithm.
Conference Paper
This document presents an energy forecast methodology using Adaptive Neuro-Fuzzy Inference System (ANFIS) and Genetic Algorithms (GA). The GA has been used for the selection of the training inputs of the ANFIS in order to minimize the training result error. The presented algorithm has been installed and it is being operating in an automotive manufacturing plant. It periodically communicates with the plant to obtain new information and update the database in order to improve its training results. Finally the obtained results of the algorithm are used in order to provide a short-term load forecasting for the different modeled consumption processes.
Conference Paper
A model of power demand represents the foundation of any intelligent Energy Management System, and its accuracy is the key factor determining the performance of such system. In order to improve the accuracy of the modeling process, a multi-model approach based on a Hierarchical Clustering of similar load behaviors is presented. The clustering algorithm joins similar data subsets in groups that are modelled separately using Adaptive Neuro-Fuzzy Inference Systems. Thus, each of the obtained models addresses only the characterization of one behavior, which provides better accuracy than classical approaches based on a single model, in addition to being easier and faster to train. During the training process of the models, an input selection technique based on Genetic Algorithms is proposed to search and select the best combination of inputs. The use of search algorithms allows to reduce the complexity of this task while maintaining the system performance, which represents a significant time saving of expert staff. The proposed approach is validated by means of experimental data from an automotive manufacturing plant. In addition to improving the forecasting accuracy, this methodology automates the segmentation of the load profiles into models and the selection of their inputs, as well as improving parallelization to effectively reduce the computation time.
Conference Paper
It is a fact that the short-term load forecasting (STLF) in the user side is growing interest. Consequently, intelligent energy management systems (iEMSs) are including this capability in order to take autonomous decisions. In this context, this paper presents a new STLF scheme based on Adaptative Networks Fuzzy Inference Systems (ANFIS). This ANFIS has an exponential output membership functions (e-ANFIS) and has been trained by means of a novel evolutionary training algorithm (ETA). Due to the computational burden required by ETA, parallel computing was used to eliminate this problem especially for embedded applications. This new scheme has been tested with real data from an automotive factory and it shows better results in comparison with typical adaptative network structures (neural network and ANFIS).
Conference Paper
The improvement of the forecasting accuracy for prediction of future loads has been object of exhaustive study in the recent years, to the point that a wide variety of methodologies which have been proved to be valid and practical exists. However, most methodologies for demand forecasting do not handle uncertainties of the resulting model, which leads to a nonproper interpretation of the forecasted outcomes. In this context, this work presents a novel load forecasting methodology in order to quantify the model uncertainties and complement the resulting information by means of adaptive confidence intervals. First, an input selection technique based on Genetic Algorithms is used to select the best combination of inputs in order to obtain a state-of-the-art model by means of Adaptive Neuro-Fuzzy Inference Systems. Then the data space is analyzed in terms of error probability of the model outcomes. The principal component analysis is used to visualize the error probability in a 2-D map. Finally, an Artificial Neural Network is used to perform the identification of the error probability associated to new measurements. In conjunction with the forecasting model, the proposed classifier extends the resulting information with an adaptive confidence intervals and its probability distribution. The effectiveness of this enhanced load forecasting methodology has been verified by experimental data obtained from an automotive plant.
Conference Paper
Full-text available
This paper introduces a new approach for training the adaptive network based fuzzy inference system (ANFIS). The previous works emphasized on gradient base method or least square (LS) based method. In this study we apply one of the swarm intelligent branches, named particle swarm optimization (PSO) with some modification in it to the training of all parameters of ANFIS structure. These modifications are inspired by natural evolutions. Finally the method is applied to the identification of nonlinear dynamical system and is compared with basic PSO and showed quite satisfactory results.
Article
Full-text available
This paper presents a new approach to short-term load forecasting in a deregulated and price-sensitive environment. A real-time pricing type scenario is envisioned where energy prices could change on an hourly basis with the consumer having the ability to react to the price signal through shifting electricity usage from expensive hours to other times, when possible. The load profile under this scenario would have different characteristics compared to that of the regulated, fixed-price era. Consequently, short-term load forecasting models customized on price-insensitive (PIS) historical data of regulated era would no longer be able to perform well. A price-sensitive (PS) load forecaster is developed. This forecaster consists of two stages, an artificial neural network based PIS load forecaster followed by a fuzzy logic (FL) system that transforms the PIS load forecasts of the first stage into PS forecasts. The first stage forecaster is a widely used forecaster in the industry known as ANNSTLF. For the FL system of the second stage, a genetic algorithm based approach is developed to automatically optimize the number of rules and the number and parameters of the fuzzy membership functions. Another FL system is developed to simulate PS load data from the PIS historical data of a utility. This new forecaster termed NFSTLF is tested on three PS databases, and it is shown that it produces superior results to the PIS ANNSTLF.
Article
Full-text available
The architecture and learning procedure underlying ANFIS (adaptive-network-based fuzzy inference system) is presented, which is a fuzzy inference system implemented in the framework of adaptive networks. By using a hybrid learning procedure, the proposed ANFIS can construct an input-output mapping based on both human knowledge (in the form of fuzzy if-then rules) and stipulated input-output data pairs. In the simulation, the ANFIS architecture is employed to model nonlinear functions, identify nonlinear components on-line in a control system, and predict a chaotic time series, all yielding remarkable results. Comparisons with artificial neural networks and earlier work on fuzzy modeling are listed and discussed. Other extensions of the proposed ANFIS and promising applications to automatic control and signal processing are also suggested
This paper presents the architecture and learning procedure underlying ANFIS (Adaptive-Network-based Fuzzy Inference System), a fuzzy inference system implemented in the framework of adaptive networks. By using a hybrid learning procedure, the proposed ANFIS can construct an input-output mapping based on both human knowledge (in the form of fuzzy if-then rules) and stipulated input-output data pairs. In our simulation, we employ the ANFIS architecture to model nonlinear functions, identify nonlinear components on-linely in a control system, and predict a chaotic time series, all yielding remarkable results. Comparisons with artificail neural networks and earlier work on fuzzy modeling are listed and discussed. Other extensions of the proposed ANFIS and promising applications to automatic control and signal processing are also suggested. I. Introduction System modeling based on conventional mathematical tools (e.g., differential equations) is not well suited for dealing with ill-defin...
A mathematical tool to build a fuzzy model of a system where fuzzy implications and reasoning are used is presented. The premise of an implication is the description of fuzzy subspace of inputs and its consequence is a linear input-output relation. The method of identification of a system using its input-output data is then shown. Two applications of the method to industrial processes are also discussed: a water cleaning process and a converter in a steel-making process.
Conference Paper
This paper introduces a new hybrid approach for training the adaptive network based fuzzy inference system (ANFIS). The previous works emphasized on gradient base method or least square (LS) based method. In this study we apply one of the swarm intelligent branches, named particle swarm optimization (PSO). The hybrid method composes PSO with gradient decent (GD) for training. We use PSO with some changes for training procedure parameters in antecedent part. These changes are inspired from genetic algorithm (GA) method. The simulation results show that in comparison with current GD training, the novel training can have a better adaptation to complex plants. Also, the results show this new hybrid approach optimizes ANFIS parameters faster and better parameters than gradient base method
Conference Paper
A methodology for building the rule-base of a fuzzy logic controller (FLC), using the iterative learning algorithm and ANFIS training is tested both in simulation and practical conditions. The methodology is aiming to bring in the intelligent characteristic to controller design procedure, by implying methods that simulates human actions as learning and adapting. The iterative self-learning algorithm is used to gather useful and trustful control data. These are subsequently used as training data for the ANFIS structure. The presented methodology is verified in two steps: i) by running simulations using Matlab environment, and ii) by constructing the rule-base of a fuzzy controller for a DC drive. The reason for developing a DC drive control structure is to analyze method's viability by comparing the results with some already known.
Article
Short term load forecasting (STLF) is an essential tool for efficient power system planning and operation. We propose in this paper the use of Bayesian techniques in order to design an optimal neural network based model for electric load forecasting. The Bayesian approach to modelling offers significant advantages over classical neural network (NN) learning methods. Among others, one can cite the automatic tuning of regularization coefficients, the selection of the most important input variables, the derivation of an uncertainty interval on the model output and the possibility to perform a comparison of different models and, therefore, select the optimal model. The proposed approach is applied to real load data.
Article
The aim of this work is to forecast total electric demand of Turkey one day in advance using neural networks. Starting with random weights and getting unrealistic forecasts could not be acceptable for a real-time operation; therefore, available past data, which is the actual load data obtained from Turkish Electricity Authority, is used off-line and the model is prepared for on-line forecasts. Moreover, a method based on hourly electric consumption is proposed to cluster input data. In order to have an idea about the success of the model, several alternate models are formed. The proposed model shows considerably better results.
Article
In this paper, a new artificial intelligent peak power load forecaster constructed by non-fixed neural networks (NNs) is developed. Several techniques, including gray analysis and stochastic back-propagation learning rule with dynamic learning rate and momentum, are used in this forecaster in order to attain more accurate prediction in forecasting operation. As a comparison, several models including recursive time series model, ANNSTLF module and fixed size NNs with constant learning rate and momentum are also performed for demonstrating the advantages of our proposed forecaster.
Article
Precise Short term load forecasting (STLF) plays a significant role in the management of power system of countries and regions on the grounds of insufficient electric energy for increased need. This paper presents an approach of back propagation neural network with rough set (RSBP) for complicated STLF with dynamic and non-linear factors to develop the accuracy of predictions. Through attribute reduction based on variable precision with rough set, the influence of noise data and weak interdependency data to BP is avoided so the time taken for training is decreased. Using load time series from a practical power system, we tested the performance of RSBP by comparing its predictions with that of BP network.
Conference Paper
The achievement of approximation models may constitute a complex computational task, in the cases of models with non-linear relation between parameters and data. This problem becomes even harder when the system to be modeled is subject to noisy data, since the simple minimization of error over a training data set can give rise to misleading models that fit both the system structure and the noise (the phenomenon of model overfit ). This paper proposes a multiobjective genetic algorithm for guiding the training of ANFIS fuzzy networks. This algorithm considers the complexity of network jointly with the error over the training set as relevant objectives, that should be minimized. Results obtained in three regression problems are presented to show the generalization capacity of models constructed with the proposed methodology.
Article
David Goldberg's Genetic Algorithms in Search, Optimization and Machine Learning is by far the bestselling introduction to genetic algorithms. Goldberg is one of the preeminent researchers in the field--he has published over 100 research articles on genetic algorithms and is a student of John Holland, the father of genetic algorithms--and his deep understanding of the material shines through. The book contains a complete listing of a simple genetic algorithm in Pascal, which C programmers can easily understand. The book covers all of the important topics in the field, including crossover, mutation, classifier systems, and fitness scaling, giving a novice with a computer science background enough information to implement a genetic algorithm and describe genetic algorithms to a friend.
Article
The adaptation of enterprise energy management (EEM) technology by businesses to control costs, optimize processes, and prevent downtime, is discussed. The system can provide facility managers with the information to make informed decisions, from a functional perspective and a financial one. It consists of a network of intelligent energy meters linked to a centrally located server running the enterprise management software. The system logs the information in an historical database, responds to any alarm conditions by relaying notifications to operations personnel, and displays the real-time statues of each monitored area on the screens of one or more networked workstations.
Article
Many power system short-term load forecasting techniques use forecast hourly temperatures in generating a load forecast. Some utility companies, however, do not have access to a weather service that provides these forecasts. To fill this need, a temperature forecaster, based on artificial neural networks, has been developed that predicts hourly temperatures up to seven days ahead. The prediction is based on forecast daily high and low temperatures and other information that would be readily available to any electric utility. The forecaster has been evaluated using data from eight utilities in the USA. The mean absolute error of one day ahead forecasts for these utilities is 1.48°F. The forecaster is implemented at several electric utilities and is being used in production environments
Article
This paper describes a modular artificial neural network (ANN) based hourly load forecaster which has already been implemented at 20 electric utilities across the US and is being used on-line by several of them. The behavior or the load and its correlation with parameters affecting it (e.g. weather variables) are decomposed into three distinct trends of weekly, daily, and hourly. Each trend is modeled by a separate module containing several multi-layer feed-forward ANNs trained by the back-propagation learning rule. The forecasts produced by each module are then combined by adaptive filters to arrive at the final forecast. During the forecasting phase, the parameters of the ANNs within each module are adaptively changed in response to the system's latest forecast accuracy. The performance of the forecaster has been tested on data from these 20 utilities with excellent results. The on-line performance of the system has also been quite satisfactory and superior to other forecasting packages used by the utilities. Moreover, the forecaster is robust, easy to use, and produces accurate results in the case of rapid weather changes