MKMohammad Javad Khosrowjerdi

Mohammad Javad Khosrowjerdi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mohammad Javad Khosrowjerdi has a total of 32 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Optimizing SVM training for DCS fault detection: A novel ensemble‐based data selection algorithm

    by Seyed Hossein Ahmadi, Mohammad Javad Khosrowjerdi

    Traditional fault detection in distributed control systems (DCS) relies on manual inspections or model‐based methods, which are time‐consuming and struggle with complex industrial data. While support vector machines (SVM) offer promising alternatives, their practical adoption in industrial settings is constrained by two fundamental challenges: the requirement for large labelled datasets and inherent susceptibility to overfitting. This paper introduces powered SVM (PSVM), a novel ensemble‐based algorithm that identifies the optimal training dataset while detecting the overfitting threshold. PSVM employs a three‐stage mechanism. First, three complementary classifiers, like Cubic SVM, Quadratic SVM, and K‑nearest neighbours (KNN) are trained on an initial 10% dataset. Second, samples misclassified by all three methods are added, achieving rapid convergence. Third, the algorithm identifies the overfitting threshold by monitoring accuracy changes across iterations. PSVM was implemented in a gas refinery DCS via open platform communications (OPC) protocol and also validated on the dry bean dataset (13,611 samples, 7 classes, 6.8:1 imbalance). In the gas refinery, PSVM achieved 99.8% fault detection accuracy (95% CI: [99.2%, 100%]), significantly outperforming conventional SVM (94.8%, p < 0.01). On the dry bean dataset, PSVM reached 96.2% accuracy using only 1722 samples (13% of training data), compared to 93.1% for conventional SVM—a 3.1% improvement with 87% less data. The algorithm identified the overfitting threshold at 12.6% data utilization and achieves 95% accuracy in just 3 iterations. This work introduces three innovations: quantitative overfitting threshold detection, an ensemble‐based data selection mechanism achieving rapid convergence with minimal data, and experimental validation showing optimal accuracy with only 13% of training data.

  • Open Access

    Nonlinear control of coaxial double rotor magnetic gear based on high gain observer in wind turbine

    by Hamideh Habibi, Amanj Aminnzhad, Mohammad Javad Khosrowjerdi

    This paper proposes a nonlinear control law with a dynamic controller based on high gain observer (HGO) for a co‐axial double rotor magnetic gear (CADRMG) based on Halbach array used in wind turbines. The generalized canonical form (GCF) is used to normalize the nonlinear augmented system to achieve the dynamic control signal with a chain of integrator of system output. In addition, a tracking problem is defined to track high speed rotor (HSR) with respect to GCF. On the other hand, the HGO is used to estimate the error tracking at the expense of nonlinear terms. Furthermore, the nonlinear system observability of the augmented system is evaluated. A dynamic control law is used to solve the tracking problem based on HGO. This controller can eliminate the effect of load torque as a constant and unknown disturbance in the closed‐loop system. Moreover, the closed‐loop stability of the system under dynamic signal control is guaranteed. The system states estimation is calculated by recursive equations from tracking error estimations. Finally, numerical simulations are given to illustrate the theoretical results of proposed system.


  • Roll and Pitch Estimation From IMU Data Using an LPV H∞ Filter

    by Ahmad Akbari, Fatemeh Rahemi, Mohammad Javad Khosrowjerdi, Saeed Ebadollahi

    Fusion schemes used for estimating roll and pitch angles from Inertial Measurement Unit (IMU) data typically suffer from one or more of the following limitations: sensitivity to inaccurate initial estimate, computational burden and sensitivity to inaccurate noise model. This paper introduces a new fusion algorithm called the Linear Parameter Varying (LPV) H ∞ filter, which aims to effectively address these limitations and strike a balance between them. The proposed LPV H ∞ filter is a weighted sum of a number of local linear H ∞ filters. Each local filter corresponds to a specific set of parameters, allowing for adaptability to different situations. By combining these local filters with their validity weights, the overall filter achieves robustness in the estimation process. The local filters and the overall filter are designed to be stable with fast dynamics, making the scheme less sensitive to inaccurate initial estimates. Moreover, the synthesis of the local filter gains is performed offline (beforehand), which reduces the computational burden during online implementation, making the proposed scheme more efficient and practical for real-time applications. To evaluate the effectiveness of the proposed scheme, a comparative study is conducted against two benchmark and most popular fusion schemes, namely, the highly accurate but computationally demanding fusion scheme of Extended Kalman Filter (EKF) as well as the adequately accurate yet computationally simple algorithm of Explicit Complementary Filter (ECF). Through this comparative study, the performance of the LPV H ∞ filter, using synthetic and experimental data, is assessed in terms of estimation accuracy/ robustness against varying noise characteristics, computational efficiency and sensitivity to inaccurate initial estimate.


  • Open Access

    A data‐driven sensor fault‐tolerant control scheme based on subspace identification

    by Mina Salim, Saeed Ahmed, Mohammad Javad Khosrowjerdi

    We study the sensor fault estimation and accommodation problems in a data‐driven setting, leading to a data‐driven sensor fault‐tolerant control scheme. First, we formulate the fault estimation problem as a finite‐horizon minimax ‐optimization problem in a data‐driven setup, whose solution yields the fault estimate. The estimated fault is then used for output compensation. This compensated output and the experimental input are used to achieve certain control objectives in a data‐driven setting. Next, the data‐driven fault estimation and control problems are solved using a subspace predictor‐based approach. Finally, the proposed algorithm is applied to the steering subsystem of the remotely operated underwater vehicle.


  • Simultaneous Output Selection and Observer Design for Vehicle Suspension System With Unknown Road Profile

    by Taha Falahati, Mehdi Mirzaei, Mohammad Javad Khosrowjerdi

    In this paper, a reduced-order observer is designed using an unknown input decoupling approach to estimate the road profile and the state variables of vehicle suspension system. During the observer design process, a new concept called the degree of observability (DO) is introduced to quantitatively describe the strengths and weaknesses of the system observability in terms of different outputs. This concept integrated with the decoupling conditions lead to a systematic method for selecting the suitable outputs of the suspension system for the observer design. In this method, among all output combinations that satisfy the decoupling requirements, the combination of sprung mass displacement and unsprung mass velocity is found to have the highest DO containing the most information of the system. The performance of the designed observer with the selected outputs is compared with the Kalman filter (KF) developed in the literature by assuming the initial random walk model for the road profile. To have a nearly real situation for simulation studies, a MacPherson suspension model constructed in the ADAMS multibody software is employed. The results obtained by deterministic and stochastic road excitations indicate much more accurate estimations for the designed observer with minimum outputs compared with the KF. The estimation errors are quantitatively compared with the KF using different set of outputs to confirm the positive effect of simultaneous DO analysis and decoupling approach. Meanwhile, the proposed observer is free of assumptions, and its structure is more simple and easier for implementation.


  • Neural Adaptive Fault Tolerant Control of Nonlinear Fractional Order Systems Via Terminal Sliding Mode Approach

    by Bijan Hashtarkhani, Mohammad Javad Khosrowjerdi

    This article proposes an adaptive neural output tracking control scheme for a class of nonlinear fractional order (FO) systems in the presence of unknown actuator faults. By means of backstepping terminal sliding mode (SM) control technique, an adaptive fractional state-feedback control law is extracted to achieve finite time stability along with output tracking for an uncertain faulty FO system. The unknown nonlinear terms are approximated by radial-basis function neural network (RBFNN) with unknown approximation error upper bound. Using convergence in finite time and fractional Lyapunov stability theorems, the finite time stability and tracking achievement are proved. Finally, the proposed fault tolerant control (FTC) approach is validated with numerical simulations on two fractional models including fractional Genesio–Tesi and fractional Duffing's oscillator systems.


  • Adaptive actuator failure compensation for uncertain nonlinear fractional order strict feedback form systems

    by Bijan Hashtarkhani, Mohammad Javad Khosrowjerdi

    In this paper, adaptive fractional control design is established for uncertain nonlinear fractional order strict feedback form systems with unknown actuator failures. The fractional actuator failure compensation problem is considered in the sense of two actuation shapes, and output matching conditions for these shapes are employed. By means of a fractional backstepping control method, two fractional adaptive state feedback control laws are designed to accomplish output tracking and to guarantee closed-loop stability in the presence of unknown actuator failures and unknown system parameters. A fractional order filter is proposed to avoid the problem of computational explosion of the backstepping design. The stability is proved via fractional order analysis method for the whole closed-loop system. Finally, simulation results for the control of fractional Chua’s circuit and fractional Genesio–Tesi systems demonstrate the effectiveness of the proposed actuator failure compensation method and output tracking property along with fast convergence of unknown parameters estimations.