In complex underwater environments, Autonomous Underwater Vehicles (AUVs) often suffer from limited onboard perception and low search efficiency in complete coverage path planning (CCPP). This paper leverages both local observations and global task statistics and proposes a Proximal Policy Optimization (PPO) algorithm with a dual-branch state representation and heuristic search to accelerate training, referred to as DB-HPPO. To better match practical scenarios and improve training efficiency, the state space integrates local environmental information perceived by the AUV and a compact global statistic of coverage progress. These semantically distinct inputs are processed in parallel by a dual-branch network and then fused for decision making. To enhance search efficiency, we design a dynamic reward function that combines coverage progress with action efficiency, and introduce a heuristic search mechanism to guide exploration when the agent is trapped in repetitive coverage. Simulation results show that, compared with baseline methods including merged-state PPO and multi-scale local-state representations, DB-HPPO converges faster and produces higher-quality coverage trajectories, improving training efficiency while reducing redundant coverage.

In this work, an adaptive learning robust controller is proposed to suppress the vibration of offshore platforms, which are subject to waves, winds, varying control delays and parametric perturbations. To realize nonlinear uncertainty approximation under the bounded H-infinity performance, the H-infinity controller incorporates both an online adaptive part and an offline fixed part. The adaptive part constructed by neural networks adjusts online, while the fixed part is obtained by regulating the H-infinity performance. Importantly, adaptive updating strategy does not require accurate values or upper bounds for real-time control delay or uncertainty. Several comparable experiments demonstrate the feasibility and effectiveness in vibration-suppression of the designed adaptive controller in shallow/deep water. This scheme significantly reduces system response variations due to structural and hydrodynamic uncertainty, as well as additional random environmental forces caused by winds.

In this article, an optimal bipartite consensus control (OBCC) scheme is proposed for heterogeneous multiagent systems (MASs) with input delay by reinforcement learning (RL) algorithm. A directed signed graph is established to construct MASs with competitive and cooperative relationships, and model reduction method is developed to tackle input delay problem. Then, based on the Hamilton–Jacobi–Bellman (HJB) equation, policy iteration method is utilized to design the bipartite consensus controller, which consists of value function and optimal controller. Further, a distributed event‐triggered function is proposed to increase control efficiency, which only requires information from its own agent and neighboring agents. Based on the input‐to‐state stability (ISS) function and Lyapunov function, sufficient conditions for the stability of MASs can be derived. Apart from that, RL algorithm is employed to solve the event‐triggered OBCC problem in MASs, where critic neural networks (NNs) and actor NNs estimate value function and control policy, respectively. Finally, simulation results are given to validate the feasibility and efficiency of the proposed algorithm.

Adaptive control methods are suitable for offshore steel structures subject to harmful vibrations, as they employ reference models to adapt to coastal and nearshore physics. To decrease the dependence on the accurate characteristics of the offshore platform, a compensating measure containing the ocean environment is proposed in the adaptive control scheme. With incomplete states as the driving input, external loads are approximated using a wavelet neural network frame. Numerical experiments are conducted on a platform model with varying parameters to test the performance of the proposed adaptive controller. It is shown that the adaptive weights derived from the chosen Lyapunov function are qualified both theoretically and practically. The system-output-based adaptive controller overcomes the disadvantage of state loss. The compensated disturbance environment guarantees the reliability of the restored reference system based on mismatched physics. The designed estimator as a part of the adaptive controller compensates for the deviations of the environment between the reference and the practical, resulting in a desirable reduction in the excessive vibration.

Damage detection plays a key role for the safety of offshore platforms. In this paper, a detection method is proposed based on one dimensional CNN to extract damage related sensitive features automatically from the acceleration data. The existence, location, number, severity of damages can be detected by our method and need no handcrafted feature engineering. The damage cases considered in the paper include both single and multiple damages with different severities. By using of ANSYS, we get the offshore platform model and make the finite element analysis. The validity of this method is verified and the results show that the proposed method can make damage detection from vibration signal data effectively.

Marine structures are inevitably influenced by parametric perturbations as well as multiple external loadings. Among these loadings, earthquake is generally more destructive and unpredictable than others. It is significant to develop effective active control schemes to guarantee the safety, stability, and integrity of marine structures subject to earthquakes and parametric perturbations. In this paper, the problem of networked [Formula: see text] robust damping control is addressed to stabilize a marine structure subject to earthquakes. First, in consideration of perturbations of the structure parameters, an uncertain model of the networked marine structure under earthquakes is presented. Second, a robust networked [Formula: see text] control scheme is presented to suppress seismic responses of the structure. By using stability theory of time-delay systems, several sufficient conditions on robust stability of the networked marine structure system are obtained, and the linear matrix inequality methods are utilized to solve the gain matrix of the controller. Finally, simulation indicates that compared with the traditional robust [Formula: see text] control and the proposed networked [Formula: see text] control, the seismic responses amplitudes of the marine structure under the two controllers are almost the same, while the latter is more economic than the former.

In this paper, we investigate the vibration control problem in an offshore platform control structure. A model predictive controller is designed under internal model principle (IMP) and model predictive control (MPC) on the basis of linear quadratic optimal theory, where a rolling-optimized observer is taken to observe and estimate mixed external disturbance. Firstly, a steel jacket offshore platform is modelled as a single-degree-of-freedom (SDOF) vibration system subjected to varying waves and winds containing sensing delay. And the process of finding a vibration suppression controller is summarized as a global optimization problem. Secondly, an optimal quadratic regulator is proposed to attenuate the structure vibration, which naturally adopts IMP considering the varying dynamics of marine disturbance. Data-driven MPC method is then adopted to deal with leading disturbance items. Thirdly, a rolling-horizon optimal algorithm is applied to the proposed disturbance observer so that desired predictive states in deriving the optimal control law are obtained regardless of disturbance sensing delay. Lyapunov stability of the proposed control strategy is proved and comparable simulation experiments are conducted with other controllers.

Abstract In this paper, an adaptive learning H∞ controller is proposed to attenuate the wave-wind-induced vibration of a jacket offshore platform with varying control delay and structured uncertainty. A novel scheme is proposed regarding bounded H∞ performance and nonlinear uncertainty approximation online, incorporated with adaptive and fixed parts respectively. The adaptive part is self-adjusting based on neural network updating laws, and the fixed part is derived through minimizing the generalized H∞ disturbance attenuation index. In the presented scheme, accurate values of real-time control delay and upper-bounds of uncertainties are not necessarily required. The stability of controller is proved by Lyapunov functions following the simulation results. Comparable experiments efficiently illustrate feasibility and vibration-attenuation effectiveness of the proposed approach in both shallow and deep water.

A neural network-based fuzzy controller is proposed to attenuate the irregular wave-induced vibration of a steel-jacket offshore platform. Firstly, the offshore platform is modeled as a system with time-varying control delay under random wave forces. Secondly, disturbance rejection measures are taken in designing an optimal controller containing delayed states. Thirdly, neural networks are adopted to observe and restore the controlled system, in order to learn the delayed control law using instant state. Finally, fuzzy models are constructed for reducing the complexity in data collecting and neural network training. Trained with sample data from fuzzy models, the neuro-fuzzy observation system is able to reconstruct the control system, and the generalized neural network-based controller works efficiently in different delayed cases. It achieves better vibration-attenuating performance under uncertain control delay and random waves, when compared to existed optimal control laws and fuzzy controllers without neural networks. The main contributions of this paper are: 1) obtaining a neural network-based observer in state approximation; 2) designing a neural network-based controller based on fuzzy rules to cope with random control delay.

This paper investigates the vibration suppression and fault diagnosis problem of jacket-type offshore platforms with uncertainties and time delay. Firstly, through a model transformation, the original continuous-time model is transformed into an identifiable differential form. Secondly, according to the model-based theory, the fault diagnosis problem can be regarded as a parameter estimation issue. The Self-adjusting forgetting factor system identification algorithm is adopted to estimate the time- varying system parameter. By computing the residuals of the parameters in the fault-free model and the estimated parameters, we can detect the faults. Thirdly, we propose an adaptive controller to overcome the parameter uncertainty by using system identification algorithm. The controller parameters can converge to a new value as the physical parameters of offshore platforms change. The simulation results demonstrate the feasibility and efficiency of the design presented in this paper for eliminating vibration and fault diagnosis.

This paper deals with the optimal bipartite consensus control (OBCC) for heterogeneous multi-agent systems (MASs) with time-delay based on reinforcement learning (RL) scheme. MASs, denoted by communication network, is modeled as a signed graph, and the competitive (cooperative) relationship is represented by negative (positive) edges. Firstly, Heterogeneous MASs is depicted by cooperative and competitive (hereinafter referred to as coopetition) network, and local neighborhood bipartite consensus errors and their performance index function (PIF) are put forward to formulate OBCC problem. Secondly, a proper model reduction method is developed to tackle time-delay in coopetition network, and then MASs with time-delay can be transformed into a delay-free MASs. Thirdly, a RL algorithm is employed to solve the solution of Hamilton-Jacobi-Bellman (HJB) equations under OBCC laws, and substantially we adopt neural networks (NNs) to calculate the control policy and PIF online. Finally, the pivotal simulation validates availability of proposed algorithm.

The paper studies the damage detection of jacket-type offshore platform and presents a new detection method based on Hilbert-Huang transform (HHT) and BP neural network (BPNN). We use HHT to decompose dynamical signals based on ANSYS model of offshore platform and get three damage indicators to determine the existence, location and severity of damage. For the damaged offshore platform, to detect damage location and severity, two kinds of neural networks are established using the damage indicators based on ANSYS model. The damage severity indicator is built from part layers of offshore platform instead of all layers. In the simulation part, we utilize a jacket-type offshore platform to verify the designed method. Simulation results show that the method can effectively determine the location and severity of damage when the noise exist. Compared with the method based on wavelet package and BPNN, the proposed method is more robust and efficient.

In this paper, the fault diagnosis of offshore platform is studied, and a method combining Support Vector Machine (SVM) and Hilbert-Huang Transform (HHT) is developed. The response signal of the structure is processed by HHT as the index of damage identification, which is used as the input of SVM. Then the damage is judged according to the output of SVM. In addition, Principal Component Analysis (PCA) is used to process the data before the damage degree examination, in order to reduce the number of data dimensions and improve the identification accuracy. This method can diagnosis the location and degree of fault. Meanwhile, diagnosis for the compound faults is also diagnosed. And the feasibility of this method is proved by the jacket platform simulated by ANSYS.

For steel jacket-type offshore platforms under irregular wave forces, we study the networked predictive vibration controller with inevitable random time delays, packet dropouts, disordering, and disturbance. First, we present a model of networked control system with two buffers, and the model is applied in the vibration control of offshore platforms. The buffers are designed to solve the above problems and located in the sensor-to-controller and controller-to-actuator channels of the networked control systems (NCSs). Second, we design networked predictive feedforward and feedback controller based on the received packets of past time. A new Algorithm is presented to simplify the computation of control law and reduce the storage required. Therefore the designed controller is physically realizable and easy to complete. Third, we made stability analysis of the controller by Lyapunov function. Finally, example of steel jacket-type offshore platform is applied to verify the feasibility and efficiency of controller. The simulation results show that the networked predictive vibration controller can compensate random big delays, large packet dropouts and disordering efficiently. Compared with different controllers, the presented predictive controller can decrease oscillation of offshore platform more significantly and the required control force can be kept in ideal small scale simultaneously.

In the paper, we study the problem of vibration control for offshore platform in nonideal network environment under wave and current forces based on the modified transformation. We consider the wave and current synchronously and give a dynamical system for the outside loads. The problem of time delay, packet dropouts and disordering due to nonideal network are taken into account. In the networked control model, two buffers output the received packets according to time stamps and upper bound of time delay. Based on a transformation, we reduce the time delay system into an equivalent delay-free system. Taking the nonideal network communication conditions into account, we design a modified transformation based on the past time data, wave and current loads, and received packets through network at current time. The networked controller is presented to depress the vibration of structure in nonideal network communication environment based on the modified transformation. An algorithm is given to simplify the computation of control law by recurrence. The system stability under presented networked controller is analyzed based on Lyapunov functional. Through the simulation results and comparison with other networked controllers, we can see that the presented controller can reduce the oscillation of offshore platform effectively.

In this paper, the network-based delayed H ∞ control with uncertainties for jacket-type offshore platform subjected to irregular wave forces is investigated. First, a dynamical model of network-based offshore platform with time delay and parameters uncertainties is established. Then, the sufficient condition of existence of controller for the offshore platform system is derived based on Lyapunov functional, and control scheme is presented to reduce the effect of wave-induced vibration on offshore platform. As shown in simulation, networked-based delayed H ∞ control (NDHC) can reduce the amplitudes of vibration of jacket type offshore platform system.

In this paper, we investigate the optimization of BP neural network. An optimized BP neural network model, adaptive BP neural network based on PSO and PCA algorithms (TPPMA) is proposed to improve the training speed and increase prediction accuracy. By introducing Momentum backpropagation and adaptive learning rate into BP Neural Network, one can reduce the possibility of local optimization. For the architecture of BP Neural Network, the initial connection weight is determined by Particle Swarm Optimization (PSO) and the number of hidden nodes is decided by three-division method. In the model, Principal Component Analysis (PCA) is used to reduce the dimension of the sample. Simulation result demonstrates that TPPMA method is efficient and can get promising results with less time compared with other results.

This paper is concerned with the damping control problem for linear systems with sinusoidal disturbances. For the linear systems with sinusoidal disturbances, a state feedback control algorithm with second-order dynamic characteristics is proposed. The dynamic feedback control law is constructed in two parts. First using the internal model principle, the mode of the sinusoidal disturbance is embedded in the controller such that the disturbances rejected to zero steady-state error. Then by designing parameters of the controller, the poles of the closed-loop system can be arbitrarily assigned, so that the exponential asymptotic stability of the closed-loop system is guaranteed. Numerical simulation examples show the effectiveness of the control strategy.

Machine vision is widely used in the detection of surface defects in industrial products. However, traditional detection algorithms are usually specialized and cannot be generalized to detect all types of defects. Object detection algorithms based on deep learning have powerful learning ability and can identify various types of defects. This paper applied object detection algorithm to defects detection of paper dish. We first captured the images with different shapes of defects. Then defects in these images were annotated and integrated for model training. Next, the model Mask R-CNN were trained for defects detection. At last, we tested the model on different defects categories. Not only the category and the location of the defect in the image could be got, but also the pixel segmentation were given. The experiments show that Mask R-CNN is a successful approach for defect detection task, which can quickly detect defects with a high accuracy.