Key Takeaways & Executive Findings
- •• Proposes a data-driven iterative learning fault estimation method that requires no model parameters, eliminating the need for system modeling. • Utilizes orthogonal triangular decomposition and singular value decomposition to construct a residual generator from input-output data. • Achieves superior transient and steady-state performance with enhanced real-time capability compared to existing methods. • Validates effectiveness on a mobile robot, demonstrating reduced manual intervention and operational complexity.
Abstract
For control systems with unknown model parameters, this paper proposes a data-driven iterative learning method for fault estimation. First, input and output data from the system under fault-free conditions are collected. By applying orthogonal triangular decomposition and singular value decomposition, a data-driven realization of the system's kernel representation is derived, based on this representation, a residual generator is constructed. Then, the actuator fault signal is estimated online by analyzing the system's dynamic residual, and an iterative learning algorithm is introduced to continuously optimize the residual-based performance function, thereby enhancing estimation accuracy. The proposed method achieves actuator fault estimation without requiring knowledge of model parameters, eliminating the time-consuming system modeling process, and allowing operators to focus on system optimization and decision-making. Compared with existing fault estimation methods, the proposed method demonstrates superior transient performance, steady-state performance, and real-time capability, reduces the need for manual intervention and lowers operational complexity. Finally, experimental results on a mobile robot verify the effectiveness and advantages of the method.
1. Introduction
With recent advances in information technology, the complexity of control systems in fields such as chemical engineering, mechanical engineering, automotive, and aerospace has also been steadily increasing. Complex control systems consist of numerous subsystems, each of which may experience failures due to issues such as component aging or damage. These failures can, in turn, impact the reliability and safety of the entire control system, potentially endangering personnel safety [1]. Under Industry 5.0's human-centric paradigm, which emphasizes a 'human-centered' approach, human needs and experiences have become a focal point. In the design and maintenance of control systems, fault detection and estimation methods based on human-machine collaboration play an indispensable role in enhancing system accuracy, ensuring personnel safety, and optimizing production efficiency, these methods are critical in modern control system design [2].
Current fault diagnosis approaches fall into two categories: model-based fault diagnosis methods [3] and data-driven fault diagnosis methods [4]. Typical model-based methods include detection filters [5] and diagnostic observers [6]. In addition, in Ref. [7], unknown fault inputs and process disturbances are expanded into new state variables of the system, and faults are estimated by designing extended state observers or intermediate observers [8]. Ref. [9] proposes an online reinforcement learning estimation strategy based on distributed intermediate observers, effectively ensuring fault estimation performance and achieving collaborative fault-tolerant control. All the above model-based fault estimation methods require precise system model parameters. Compared to model-based fault diagnosis methods, data-driven fault estimation methods do not require accurate system models; instead, they only need the system's input-output data to obtain residuals from the residual generator for fault estimation. Since system identification is unnecessary, the research on data-driven fault detection and estimation methods is of great significance for most complex industrial process systems.
In recent years, Odendaal and Jiang have conducted some research on data-driven fault detection [10, 11]. However, compared to fault detection, implementing data-driven fault estimation is much more complex, and fault estimation plays a critical role in fault-tolerant control of systems. Ref. [12] introduces an input–output (I/O) data model and establishes a residual analysis model by analyzing data, including fault signals, laying the foundation for the proposed method.
Loading authentic research manuscript (Pages 1–5)...
Fei Wang, Jie Sun, Junwei Zhu, Ruofeng Wei (2025). Data-Driven Human-in-the-Loop Iterative Learning Fault Estimation Method. Chinese Journal of Mechanical Engineering. https://doi.org/10.1186/s10033-025-01323-6
Research & Educational Purpose Only:The translations, structured abstracts, analytical annotations, and data reports provided by SinoTechIntel are intended exclusively for academic research, internal corporate R&D, and educational benchmarking. They do not constitute formal engineering, chemical safety, legal, or professional advice.
Copyright & Intellectual Property Notice: Original copyright of the underlying source articles and experimental data remains with the respective authors, institutions, and original publishing journals. SinoTechIntel claims intellectual property only over its proprietary translations, analytical syntheses, and AEO structured enhancements in accordance with international fair use and academic citation principles.
Frequently Asked Questions
What is the main contribution of this paper?
The paper proposes a data-driven iterative learning fault estimation method that does not require model parameters, eliminating the need for system modeling and enabling online fault estimation with improved accuracy and real-time performance.
How does the proposed method work?
The method collects input-output data under fault-free conditions, applies orthogonal triangular decomposition and singular value decomposition to derive a kernel representation, constructs a residual generator, and then estimates actuator faults online by analyzing dynamic residuals and iteratively optimizing a performance function.
What are the advantages of the proposed method over existing approaches?
Compared to model-based methods, it does not require precise system models, reducing modeling time and complexity. It also demonstrates superior transient and steady-state performance, better real-time capability, and reduced manual intervention.
What experimental validation was performed?
The method was validated on a mobile robot, confirming its effectiveness and advantages in practical applications.
What is the significance of the human-in-the-loop aspect?
The human-in-the-loop aspect allows operators to focus on system optimization and decision-making rather than time-consuming modeling, aligning with Industry 5.0's human-centric paradigm.
Related Technical Papers & Translations
Direct Repair of the Crystal Structure and Coating Surface of Spent LiFePO4 Materials Enables Superfast Li-Ion Migration
The rapid accumulation of spent LiFePO4 (LFP) cathodes from retired lithium-ion batteries necessitates the development of effective and environmental-friendly recycling strategies. In this context, direct regeneration has emerged as a promising approach for reclaiming LFP cathode materials, offering a streamlined pathway to restore their electrochemical functionality. We report an integrated regeneration protocol that simultaneously repairs the degraded crystal structure and reconstructs the damaged carbon coating in spent LFP. The regenerated cathode material had superfast lithium-ion diffusion kinetics and a stable cathode–electrolyte interface, giving a remarkable rate capability with specific capacities of 122 mAh g−1 at 5C and 106 mAh g−1 at 10C (1C = 170 mA g−1). It also maintained capacities of 110.7 mAh g−1 (5C) and 84.1 mAh g−1 (10C) after 400 cycles. It could be used in harsh environments and could be stably cycled at subzero temperatures (−10 and −20 °C) and in solid-state electrolyte batteries. Life cycle assessment combined with economic evaluation using the EverBatt model reveals that this direct regeneration approach has high economic and environmental benefits.
Oxide Semiconductor for Advanced Memory Architectures: Atomic Layer Deposition, Key Requirement and Challenges
Oxide semiconductors (OSs), introduced by the Hosono group in the early 2000s, have evolved from display backplane materials to promising candidates for advanced memory and logic devices. The exceptionally low leakage current of OSs and compatibility with three-dimensional (3D) architectures have recently sparked renewed interest in their use in semiconductor applications. This review begins by exploring the unique material properties of OSs, which fundamentally originate from their distinct electronic band structure. Subsequently, we focus on atomic layer deposition (ALD), a core technique for growing excellent OS films, covering both basic and advanced processes compatible with 3D scaling. The basic surface reaction mechanisms—adsorption and reaction—and their roles in film growth are introduced. Furthermore, material design strategies, such as cation selection, crystallinity control, anion doping, and heterostructure engineering, are discussed. We also highlight challenges in memory applications, including contact resistance, hydrogen instability, and lack of p-type materials, and discuss the feasibility of ALD-grown OSs as potential solutions. Lastly, we provide an outlook on the role of ALD-grown OSs in memory technologies. This review bridges material fundamentals and device-level requirements, offering a comprehensive perspective on the potential of ALD-driven OSs for next-generation semiconductor memory devices.
Laser powder bed fusion of biodegradable Zn-4Cu alloy: Processing, microstructure and properties
Zn's natural degradability and biocompatibility make it a promising candidate for implants, however, its mechanical properties remain insufficient for bone applications. In this study, the performance of Zn was enhanced by developing Zn-Cu alloys via laser powder bed fusion (LPBF). Optimal LPBF parameters for forming stable tracks were achieved by adjusting laser power and scanning speed. Under optimized conditions of 100 W and 100 mm/s, high-density (99.58%) Zn-Cu alloys with improved hardness (68.2HV) and yield strength (160 MPa) were achieved. These improvements are attributed to solid solution strengthening, segregation strengthening, and grain refinement. The Zn-Cu alloys also demonstrated favorable degradation behavior, with a rate of 0.16 mm/year. This degradation is primarily driven by micro-galvanic corrosion between the CuZn5 phase and Zn matrix, along with refined grains and increased grain boundary density. This work demonstrates a viable strategy for fabricating Zn-based implants with enhanced structural integrity and mechanical performance via LPBF.