This paper investigates optimal procurement strategies for small and medium-sized enterprises (SMEs) under the dual constraints of financial limitations and supply disruption risks. Taking a construction engineering enterprise (Enterprise D) as a case study, we develop mathematical models to compare two procurement modes: dual-strategy sourcing and one-strategy-one-standby sourcing. The models consider order costs, disruption probability, loan interest rates, and limited funds. Results indicate that optimal order quantity decreases with higher disruption risk, while interest rates have little effect. Profitability persists up to a 0.7 disruption risk in the dual-strategy model but drops beyond 0.3 in the one-strategy-one-standby model. The study concludes that, under moderate risk levels, dual-strategy sourcing is more robust and should be prioritised, while financial efficiency can be enhanced by choosing banks with lower interest rates.

This paper presents a variable-gain sliding mode control (VGSMC) strategy based on a continuous barrier function to enhance the dynamic performance and disturbance rejection of aerospace electric fuel pumps. The approach integrates the robustness of conventional sliding mode control with adaptive gain tuning to reduce chattering while ensuring fast convergence. A comprehensive model of a surface-mounted permanent magnet synchronous motor (PMSM) and a nonlinear gear pump is developed. An extended sliding mode observer (ESMO) is designed to estimate and compensate for unknown, time-varying disturbances from complex fluid loads. Lyapunov analysis confirms finite-time convergence of both the controller and observer. Simulations under sudden load changes and variable-speed operations demonstrate that the proposed VGSMC achieves faster response, higher steady-state accuracy, and smoother control than PI and fixed-gain SMC. This method offers a robust and practical solution for high-performance aerospace fuel pump control.

This paper proposes a quantity-limited controller (QLC) for continuous-time Markov jump systems (MJSs) to make a good balance between mode-dependent and mode-independent controllers. It is handled by using a mode-separation method, whose conditions for the suboptimal boundary control strategy are given in terms of LMIs. The performance error between the mode-dependent controllers (MDCs) and QLCs is reduced by solving an optimisation problem. Another improved controller is presented by applying a multi-player non-cooperative game, whose existence conditions for a suboptimal boundary Nash equilibrium strategy are presented with coupled LMIs. A genetic-algorithm (GA)-based alternating optimisation procedure is given to overcome the computation difficulties of finding an optimal solution to the optimisation problem. A phenomenon having a directive significance is found that the pursuit of maximising the personal interests is a double-edged sword to maximise the overall interests. Finally, an example is used to demonstrate the utility and advantage of the methods.

Given the non-linear property of the belief rule-base (BRB) approach, attribute weights (AW), rule weights (RW), and belief degrees of rule consequences (BDRC) should be carefully optimised in advance for a better performance. However, the impact of optimisation on these three types of parameters has not been carefully analysed. In this paper, we analyse the effect of optimisation on three types of parameters, beginning with those calculated by the belief-based method. Specifically, we adopt an archive-based differential evolution algorithm with three objective functions (namely, arithmetic mean accuracy (AMA), cross-entropy (CE), and mean absolute difference (MAD)) for optimisation. Then, we leverage the Wilcoxon signed-rank tests to analyse the benefits obtained from the optimisation. Experiments indicate that optimising these parameters is necessary, and the fitness function should be carefully designed to achieve better performance. Furthermore, CE is the most effective objective function for optimising RW. And, MAD outperforms others for BDRC.

Aiming at the rendezvous and formation control problems of tanker aircraft and receiver aircraft during autonomous aerial refueling, the desired trajectory and flight control law of tanker aircraft are designed based on the Lyapunov navigation vector field, and the third-order consistency variable-structure formation controller is designed to keep the tanker aircraft and the receiver aircraft in the desired formation. Firstly, the UAV model is established; and the flight control law of the tanker aircraft is designed based on the Lyapunov navigation vector field; then, based on the position and speed information of the tanker aircraft, the third-order consistency formation controller is designed for the receiver aircraft, and combined with the distributed formation method, the variable-structure communication rule is designed in the case of a UAV is crashed within the formation; finally, the simulation proves the validity of the variable-structure formation controller based on the third-order consistency theory.

The design of decentralised robust event-triggered non-fragile dissipative controllers for a fractional-order linear uncertain interconnected system is considered in this paper. The subsystem's uncertainty is assumed to be norm-bounded and time-varying. First, a new concept of event-triggered dissipativity is introduced for fractional-order interconnected systems. Then, decentralised event-triggered non-fragile controllers are designed to guarantee that the closed-loop interconnected systems are asymptotically stable and satisfy a dissipative performance index. Third, a sufficient condition for the existence of the proposed controller is established based on the linear matrix inequalities and the fractional-order Lyapunov method. Finally, an example with simulation results is presented to verify the effectiveness of the obtained results.

To improve the parameter identification accuracy of vessel steering machine (VSM), a multivariate evidential reasoning-based particle swarm optimisation (MER-PSO) method is proposed. Firstly, multivariate evidential reasoning (MER), a framework designed to directly model multiple-input multiple-output (MIMO) problems without cascading multiple MISO models, is proposed and applied to the joint dynamic modelling and collaborative inference of multiple PSO hyperparameters, which overcomes the inherent limitations of traditional ER in handling MIMO scenarios. Subsequently, the integrated MER-PSO is proposed and used for VSM parameter identification, which provides a new adaptive optimisation framework for parameter identification of complex nonlinear systems. Finally, the significant advantages of MER-PSO (with a 53.7% reduction in mean square error compared to PSO) are demonstrated via the VSM parameter identification experiments, which include the simulation/actual data-based parameter identification experiments and provide a new paradigm for fine-grained modelling.

A nonlinear model predictive control (NMPC) with a feedback fault compensator, using an adaptive-bias radial basis function neural network (RBFNN), is proposed to achieve high precision control of nonlinear processes with dynamic modelling deviations or fault disturbances. The stability control is realised by NMPC with prediction optimisation and feedback correction. An RBFNN is constructed to compensate for the lumped deviations caused by modelling errors, fault disturbances and random noise. A particle swarm optimisation algorithm with Q-reinforcement learning (QPSO) is established to obtain the optimal control parameters, and its fitness function is designed by cumulative absolute tracking error (CATE). The closed-loop stability of the proposed algorithm is analysed by constructing a Lyapunov function. The superiority of the proposed control algorithm is verified through comparative testing. The average coupling error (MCE) of the control algorithm proposed is 0.14, which has been reduced by three times compared to other algorithms.

Solving dynamic traveling salesman problems (DTSPs) is a challenging task due to the constantly changing freight volumes and demands between different city nodes. To alleviate this issue, a deep reinforcement learning-assisted large neighborhood dynamic robust algorithm (LNDRA-DRL) is proposed in the present study. In the LNDRA-DRL, an end-to-end method is used to produce a high-quality initial robust individual, which is then refined using a large neighborhood dynamic robust algorithm to find final robust solutions. To demonstrate the performance of the proposed LNDRA-DRL, five novel dynamic robust traveling salesman problems (DRTSPs) are constructed based on the TSPLIB benchmark test suite, and three competitive algorithms are used for comparison in experiments. Experimental results demonstrate that the proposed LNDRADRL can find a set of satisfactory robust solutions that are nearly optimal in different dynamic environments. Furthermore, it can reduce the switching times of solutions within an acceptable threshold when environmental conditions change.

This paper studies distributed nonlinear estimation with fixed communication delays over a connected undirected graph. Agents take independent and identically distributed noisy measurements of an unknown parameter and update estimates via a delayed consensus + innovations algorithm, combining local information, delayed neighbour estimates, and local innovations. Under mild assumptions, proper stepsizes ensure consistent estimation, with asymptotic performance matching delay-free estimators. Finally, simulation results are provided to validate the proposed algorithm.