Recurrent Adaptive Neuro-fuzzy Inference Research Articles

A chaos recurrent ANFIS optimized by PSO to predict ground vibration generated in rock blasting

The use of explosives is a common and economical method to fragment and/or displace hard rocks in tunnels and surface and underground mines. Ground vibration, as a side environmental effect induced by blast events, has detrimental impacts on nearby structures like dams and buildings. Therefore, an accurate and reliable estimation of ground vibration is imperative. The goal of this paper is to present a new hybrid model by combining chaos recurrent adaptive neuro-fuzzy inference system (CRANFIS) and particle swarm optimization (PSO) to predict ground vibration. To the best of our knowledge, this is the first research that predicts the ground vibration through a model integrating CRANFIS and PSO. To evaluate the efficiency of the proposed model, the results of CRANFIS-PSO were compared with those of the CRANFIS, RANFIS, ANFIS, artificial neural network (ANN), and several empirical methods. In other words, first, the empirical methods were developed; then, due to their unacceptable performance, the artificial intelligence methods were developed. The results clearly indicated the superiority of CRANFIS-PSO over the above-mentioned methods in terms of predicting ground vibration. The values of coefficient of determination (R2) obtained from CRANFIS-PSO, CRANFIS, RANFIS, ANFIS, and ANN models were 0.997, 0.967, 0.958, 0.822, and 0.775, respectively. Accordingly, the CRANFIS-PSO model could be employed as a reliable and accurate data intelligent model to solve highly-nonlinear problems such as the prediction of blast-induced flyrock and air-overpressure.

Neuro-Fuzzy Cognitive Temporal Models For Predicting Multidimensional Time Series With Fuzzy Trends

A new type of Neuro-Fuzzy Cognitive Temporal Models (NFCTM) is proposed for predicting multidimensional time series (MTS) taking into account the fuzzy trends of all MTS components. NFCTMS allow predicting the MTS in conditions of non-stochastic uncertainty, non-linearity of mutual influence, partial inconsistency and significant interdependence of the MTS components, as well as in conditions of small samples. This takes into account the direct and indirect mutual influence of all the MTS components with different time lags relative to each other. To carry out temporal changes in specific MTS components the original neuro-fuzzy models of RecANFIS (Recurrent Adaptive Neuro-Fuzzy Inference System/Model) type are applied, that: firstly, allow to save the predicted values of the MTS components in the range of “liding-window” time series; secondly, identify fuzzy trends of the components of the MTS in the range of “sliding-window” time series; thirdly, adaptively take into account fuzzy trends of the MTS components based on the fuzzy mappings. An original way of a coherent learning NFCTM is described, which lies: firstly, in training RecANFISs for each concept NFCTM (MTS component) taking into account fuzzy trends; secondly, in coherence of all RecANFISs between each other to maximize the prediction accuracy of each of the MTS components without compromising the prediction accuracy of at least one of the other MTS components. Experimental studies have been carried out and the results of using the proposed NFCTM for multidimensional forecasting of the urban environment state in Moscow in conditions of a complex epidemiological situation have been obtained.

Learning adaptive fuzzy droop of PV contribution to frequency excursion of hybrid micro-grid during parameters uncertainties

This paper improves the performance of contribution of photovoltaic (PV) panels to the stabilization of a stand-alone hybrid microgrid during frequency excursions. Recent growth in converter-interfaced topologies and using renewable energy resources has led to decrease in inertia of isolated micro-grids (MGs) and presented some challenges to the frequency control and strength of hybrid ones. One of the effective measures to resolve mentioned problems is to deploy some potential energy resources like PV by considering a small proportion of solar panels’ generation as headroom. In addition, other distributed energy resources (DERs), which can be regarded as ancillary resources, can play an active role in the resiliency of MG’s performance in the low-inertia power systems. To this effective contribution considering some uncertainties of the microgrid parameters like the time constant of the micro-turbine, the time constant of the governor, the speed droop regulation constant as well as the load damping coefficient, an adaptive fuzzy mechanism can be proposed to determine the level of active power injection. Recurrent Adaptive Neuro Fuzzy Inference System (ANFIS) technique trains this non-linear adaptable droop to deal with uncertainties in a reasonable way. In order to ascertain parameters of membership functions in an intelligent way, this droop takes advantage of Artificial Bee Colony (ABC) algorithm based on a multi-objective. The Simulation results verify the robustness and reliability of this flexible fuzzy droop during different operating conditions as well as intermittent nature of renewable energy resources such as wind turbine generators in hybrid micro-grid.

Coordination between Demand Response Programming and Learning-Based FOPID Controller for Alleviation of Frequency Excursion of Hybrid Microgrid

In recent years, residential rate consumptions have increased due to modern appliances which require a high level of electricity demands. Although mentioned appliances can improve the quality of consumers’ lives to a certain extent, they suffer from various shortcomings including raising the electricity bill as well as serious technical issues such as lack of balance between electricity generation and load disturbances. This imbalance can generally lead to the frequency excursion which is a significant concern, especially for low-inertia microgrids with unpredictable parameters. This research proposes an intelligent combination of two approaches in order to alleviate challenges related to the frequency control mechanism. Firstly, a learning-based fractional-order proportional-integral-derivative (FOPID) controller is trained by recurrent adaptive neuro-fuzzy inference (RANFIS) in the generation side during various operational conditions and climatic changes. In the following, a decentralized demand response (DR) programming in the load side is introduced to minimize consumption rate through controllable appliances and energy storage systems (ESSs). Furthermore, parameters uncertainties and time delay, which are generally known as two main concerns of isolated microgrids, are regarded in the frequency plan of a low-inertia microgrid including renewable energy sources (RESs), and energy storage systems (ESSs). Simulation results are illustrated in three different case studies in order to compare the performance of the proposed two methods during various operational conditions. It is obvious that the frequency deviation of microgrid can be improved by taking advantage of intelligent combination of both DR program and modern control mechanism.

A new practical feed-forward cascade analyze for close loop identification of combustion control loop system through RANFIS and NARX

Boiler-turbine is a multi-variable and complex system in steam power plants including combustion control loop, temperature control loop and drum water level control loop, where each of them are separate control loops. Selecting control loops as a unique loop in order to identify and control the boiler as a whole unit is a difficult and complex task, because of nonlinear time variant dynamic characteristics of the boiler. In the present article, a realistic and efficient model of the combustion control loop of the 320 MW steam power plant of Islam Abad, Isfahan is identified separately from other control loops of the boiler, unlike similar articles. Another critical point about this article is being superior to the others, due to identifying the combustion control loop system by feed-forward cascade analysis. Due to the sensitivity and complexity of the system, with respect to its nonlinear and closed loop characteristics, the combustion control loop system is considered as a feed-forward cascade and the identification of the system is carried out during the load changes and conducted through intelligent procedures like recurrent adaptive neuro-fuzzy inference system (RANFIS) and nonlinear autoregressive model with exogenous input (NARX). The obtained results through actual data collected from the plant are presented and the accuracy of the procedures is determined.

Recurrent ANFIS-Coordinated Controller Design for Multimachine Power System with FACTS Devices

This paper proposes the design of auxiliary-coordinated controller for static VAR compensator (SVC) and thyristor-controlled series capacitor (TCSC) devices by adaptive fuzzy optimized technique for oscillation damping in multimachine power systems. The performance of the coordinated control of SVC and TCSC devices based on feedforward adaptive neuro fuzzy inference system (F-ANFIS) is compared with that of the adaptive neuro fuzzy inference system (ANFIS) structure based on recurrent adaptive neuro fuzzy inference system (R-ANFIS) network architecture. The objective of the coordinated controller design is to tune the parameters of SVC and TCSC fuzzy lead lag compensator simultaneously to minimize the deviation of rotor angle and rotor speed of the generators. The performance of the system is enhanced by optimally tuning the membership functions of fuzzy lead lag controller parameter of the flexible AC transmission system (FACTS) by R-ANFIS controller. The training data for F-ANFIS and R-ANFIS are generated by conventional linear control technique under various operating conditions. The offline trained controller tunes the parameter of lead lag controller in online. The oscillation damping ability of the system is analyzed for three-machine test system by calculating the standard deviation and cost function. The superior performance of R-ANFIS controller is compared with various particle swarm optimization-based feedforward ANFIS controllers available in literature.