Search

The efficacy of photovoltaic systems is significantly impacted by electrical production losses attributed to faults. Ensuring the rapid and cost-effective restoration of system efficiency necessitates robust fault detection and diagnosis (FDD) procedures. This study introduces a novel interval-gated recurrent unit (I-GRU) based Bayesian optimization framework for FDD in grid-connected photovoltaic (GCPV) systems. The utilization of an interval-valued representation is proposed to address uncertainties inherent in the systems, the GRU is employed for fault classification, while the Bayesian algorithm optimizes its hyperparameters. Addressing uncertainties through the proposed approach enhances monitoring capabilities, mitigating computational and storage costs associated with sensor uncertainties.

In this paper, we present an innovative strategy for nonlinear model predictive control by employing a discrete-time NARX-Laguerre model. This latter model is crafted through the expansion of discrete-time NARX model parameters using a set of five independent Laguerre bases. A notable benefit of this approach is a substantial reduction in the number of parameters compared to the classical NARX model. However, the realization of this reduction depends on the careful selection of optimal Laguerre poles that define these bases. The parameters of the NARX-Laguerre model are determined through a recursive methodology. This resulting model is subsequently applied in the implementation of nonlinear model predictive control. To formulate the optimization problem, we incorporate a performance criterion that takes into account both process input and output constraints. We assess the effectiveness of this novel approach to nonlinear model predictive control through experimentation on the Twin Rotor System.

This paper proposes effective Random Forest (RF)-based fault detection and diagnosis for wind energy conversion (WEC) systems. The proposed technique involved two major steps: feature selection and fault classification. Feature selection pre-processing is an important step to increase the accuracy of the classification algorithm and decrease the dimensionality of a dataset. Therefore, a hybrid feature selection based diagnosis technique, that can preserve the advantages of wrapper and filter algorithms as well as RF model, is proposed. In the first phase, the neighborhood component analysis (NCA) filter algorithm is used to reduce and select only the pertinent features from the original raw data. This phase helps in improving data by removing redundant and unimportant features. In the second step, we applied a wrapper technique called equilibrium optimizer to get optimized features and better classification accuracy. The main idea behind using a hybrid feature selection step is to select a small subset from original data that can achieve maximum classification accuracy and reduce the computational complexity of the RF technique. Then, the sensitive and significant characteristics are transmitted to the RF model for classification purposes. The presented results prove that the proposed methods offer enhanced diagnosis accuracy when applied to WEC systems.

This paper introduces a pioneering fault diagnosis technique termed Interval Ensemble Learning based on Sine Cosine Optimization Algorithm (IEL- SCOA), tailored to tackle uncertainties prevalent in wind energy conversion (WEC) systems. The approach unfolds in three integral phases. Firstly, the establishment of interval centers and ranges, employing upper and lower bounds, effectively manages the inherent uncertainties arising from noise and measurement errors intrinsic to the wind system. Subsequently, the dataset undergoes processing via the Sine-Cosine Optimization Algorithm (SCOA), enabling the extraction of the most pertinent attributes. The culmination of predictive precision and classification performance is achieved through the integration of the refined dataset into an ensemble learning paradigm, harmonizing bagging, boosting techniques, and an artificial neural network classifier. The principal aim of the IEL-SCOA approach is to discern the spectrum of operational conditions within WEC systems, encompassing a healthy mode alongside six distinct faulty modes. These anomalies, encompassing short circuits, open circuits, and wear-out incidents, are deliberately induced at diverse locations and facets of the system, notably the generator and grid sides. Empirical results underscore the robustness and efficiency of the proposed methodology, showcasing an exceptional accuracy rate of 99.76 %. These outcomes definitively establish the IEL-SCOA approach as a potent and efficacious tool for precise fault diagnosis in uncertain WEC systems.

The diagnosis of faults in grid-connected photovoltaic (GCPV) systems is a challenging task due to their complex nature and the high similarity between faults. To address this issue, we propose a wrapper approach called the salp swarm algorithm (SSA) for feature selection. The main objective of SSA is to extract only the most important features from the raw data and eliminate unnecessary ones to improve the classification accuracy of supervised machine learning (SML) classifiers. Subsequently, the selected features are used to train supervised machine learning (SML) techniques in distinguishing between various operating modes. To evaluate the efficiency of the technique, we used healthy and faulty data from GCPV systems that have been injected with frequent faults, 20 different types of faults were introduced, including line-to-line, line-to-ground, connectivity faults, and those affecting the operation of bay-pass diodes. These faults present diverse conditions, such as simple and multiple faults in the PV arrays and mixed faults in both arrays. The performances of the developed SSA-SML are compared with those using principal component analysis (PCA) and kernel PCA (KPCA) based SML techniques through different criteria (i.e., accuracy, recall, precision, F1 score, and computation time). The experimental findings demonstrated that the proposed diagnosis paradigm outperformed the other techniques and achieved a high diagnostic accuracy (an average accuracy greater than 99%) while significantly reducing computation time.

The major concept of the future Industrial 4.0 framework is the integration of artificial intelligence (AI) and the implementation of digital twin (DT), which avoids serious economic losses caused by unexpected equipment failures and significantly improves system reliability. DT is an emerging technology in the context of digital transformation that enables the monitoring, diagnosis, energy efficiency, and optimization of different systems. Numerous initiatives have shown how AI can enhance the performance of DT for industrial applications. This paper describes a data-based DT architecture for the monitoring, and predictive maintenance (PdM) in manufacturing. This new concept is based on deep learning, specifically the autoencoder model. The system was tested on a real industry example, by developing the data collection, data system analysis, and applying the deep learning approach. The data were collected from a Profinet communication network installed on an automated system. This approach enables better quality results and more efficient management of the weaver's workshop. Lastly, to prove the efficiency and the accuracy of the newly developed approach, an example is shown.

This paper investigates the problem of incipient fault detection and diagnosis (FDD) in wind energy conversion systems (WECS) using an innovative and effective approach called the ensemble learning-sine cosine optimization algorithm (EL-SCOA). The evolved strategy involves two primary steps: first, a sine-cosine algorithm is used to extract and optimize features in order to only select the most descriptive ones. Second, to further improve the capability, thereby providing the highest accuracy performance, the newly gathered dataset is introduced as input to an ensemble learning paradigm, which merges the benefits of boosting and bagging techniques with an artificial neural network classifier. The essential goal of the developed proposal is to discriminate between the diverse operating conditions (one healthy and six faulty conditions). Three potential and frequent types of faults that can affect the system behaviors including short-circuit, open-circuit, and wear-out are considered and thereby injected at diverse locations and sides (grid and generator sides) in order to evaluate the availability and performance of the proposed technique when compared to the conventional FDD methods. The diagnosis performance is analyzed in terms of accuracy, recall, precision, and computation time. The acquired outcomes demonstrate the efficiency of the suggested diagnostic paradigm compared to conventional FDD techniques (accuracy rate has been successfully achieved 98.35%).

This paper proposes a new method to estimate, from input/output measurements, the structure parameters (order and memory) of Volterra models used for describing nonlinear systems. For each structure parameter (order and memory), the identification method is based on the definition, for increasing values of such parameter, of a specific matrix the components of which are lagged inputs and lagged outputs. This matrix becomes singular once the parameter value exceeds its exact value. The proposed method is tested in numerical examples, then it is used for modelling a chemical reactor and the results were successful.

In this article, we explore the use of the Laguerre functions for the estimation of multivariable autoregressive with exogenous (ARX) input model. Each polynomial function of the MIMO ARX model associated to the inputs and to the outputs is expanded on independent Laguerre orthonormal bases. The resulting model is entitled MIMO ARX-Laguerre model. The optimal approximation of which is ensured once the poles characterising each Laguerre orthonormal basis are set to their optimal values. In this paper, a minimal recursive representation realisations for MIMO discrete-time linear systems are derived using Laguerre functions. Further we propose, from input/output measurements, an iterative optimisation algorithm for the free parameters (Laguerre poles). Simulation results show the effectiveness of the proposed optimal modelling method.

This paper proposes a new approach for synthesising a predictive control for non-linear uncertain process based on a proposed reduced complexity discrete-time Volterra model known as GOBF-Volterra model. This model, provided by expanding each Volterra kernel on independent generalised orthonormal basis functions (GOBF), is efficient for the synthesis of non-linear model-based predictive control (NMBPC) which copes with physical constraints and geometrical constraints due to parameter uncertainties. A quadratic criterion is optimised and a new optimisation algorithm, formulated as a quadratic programming (QP) under linear and non-linear constraints, is proposed. Simulation results on a chemical reactor are presented to illustrate the performance of the proposed NMBPC strategy for uncertain process. This reveals that the stability performance of the resulting closed-loop system depends on the choice of the tuning parameters.

In this paper, we propose a new dynamic non-linear MISO system model using discrete-time Volterra series. To provide a reduced complexity model, each Volterra kernel is expanded on independent generalised orthonormal bases (GOBs) associated to the inputs to develop a new black-box non-linear MISO-GOB-Volterra model. However, this reduction is ensured once the poles characterising each independent generalised orthonormal basis (GOB) are set to their optimal values. For the selection of optimal GOBs poles, we develop two new general approaches based on Gauss-Newton and exhaustive algorithms, the performances of which are illustrated and compared in simulation.

• Distributed model predictive control based on game theory framework is proposed for nonlinear systems with nonlinearly coupled dynamics. • We consider the control of a set of sub-systems as a group of players in a cooperative situation, where a decision made by each individual player affects the decisions of the other players. • The formulation of the distributed MPC as a coalitional game is done using a dual decomposition approach where the interconnection between subsystems is relaxed using lagrange multipliers. • The subsystems cooperate and decide which coalition to join according to the benefit that each can gain from that coalition. • The approach is tested to a four tanks system where robustness is also proved. In this paper, a distributed model predictive control based on game theory framework is proposed for nonlinear systems with nonlinearly coupled dynamics. In our approach we consider the control of a set of sub-systems as a group of players in a cooperative situation, where a decision made by each individual player affects the decisions of the other players. The formulation of the distributed MPC as a coalitional game is done using a dual decomposition approach where the interconnection between subsystems is relaxed using Lagrange multipliers, the subsystems cooperate and decide which coalition to join according to the benefit that each can gain from that coalition. The approach is tested to a four tanks system where robustness is also proved. [ABSTRACT FROM AUTHOR]

This paper is devoted to the issue of a robust predictive control for linear discrete-time systems by using Meixner-like model. The Meixner-like functions are an extension of Laguerre functions and convenient when the system has a slow start or delay. To ensure the reduction of the parameter number in the Meixner-like model, the optimization of parameters characterizing the Meixner-like functions is proposed. This proposed robust predictive control copes with physical constraints and geometrical constraints due to parameter uncertainties, which are estimated by using the Unknown But Bounded Error (UBBE) approach, and leads to the min-max optimization problem. [ABSTRACT FROM AUTHOR]