Stochastic sampled-data multi-objective control of active suspension systems for in-wheel motor driven electric vehicles

This paper addresses the sampled-data multi-objective active suspension control problem for an in-wheel motor driven electric vehicle subject to stochastic sampling periods and asynchronous premise variables.The focus is placed on the scenario that the dynamical state of the half-vehicle active suspension system is transmitted over an in-vehicle controller area network that only permits the transmission of sampled data packets.For this purpose,a stochastic sampling mechanism is developed such that the sampling periods can randomly switch among different values with certain mathematical probabilities.Then,an asynchronous fuzzy sampled-data controller,featuring distinct premise variables from the active suspension system,is constructed to eliminate the stringent requirement that the sampled-data controller has to share the same grades of membership.Furthermore,novel criteria for both stability analysis and controller design are derived in order to guarantee that the resultant closed-loop active suspension system is stochastically stable with simultaneous𝐻2 and𝐻∞performance requirements.Finally,the effectiveness of the proposed stochastic sampled-data multi-objective control method is verified via several numerical cases studies in both time domain and frequency domain under various road disturbance profiles.

Deep Neural Network Architecture Search via Decomposition-Based Multi-Objective Stochastic Fractal Search

Deep neural networks often outperform classical machine learning algorithms in solving real-world problems.However,designing better networks usually requires domain expertise and consumes significant time and com-puting resources.Moreover,when the task changes,the original network architecture becomes outdated and requires redesigning.Thus,Neural Architecture Search(NAS)has gained attention as an effective approach to automatically generate optimal network architectures.Most NAS methods mainly focus on achieving high performance while ignoring architectural complexity.A myriad of research has revealed that network performance and structural complexity are often positively correlated.Nevertheless,complex network structures will bring enormous computing resources.To cope with this,we formulate the neural architecture search task as a multi-objective optimization problem,where an optimal architecture is learned by minimizing the classification error rate and the number of network parameters simultaneously.And then a decomposition-based multi-objective stochastic fractal search method is proposed to solve it.In view of the discrete property of the NAS problem,we discretize the stochastic fractal search step size so that the network architecture can be optimized more effectively.Additionally,two distinct update methods are employed in step size update stage to enhance the global and local search abilities adaptively.Furthermore,an information exchange mechanism between architectures is raised to accelerate the convergence process and improve the efficiency of the algorithm.Experimental studies show that the proposed algorithm has competitive performance comparable to many existing manual and automatic deep neural network generation approaches,which achieved a parameter-less and high-precision architecture with low-cost on each of the six benchmark datasets.

Scheduling an Energy-Aware Parallel Machine System with Deteriorating and Learning Effects Considering Multiple Optimization Objectives and Stochastic Processing Time

Currently,energy conservation draws wide attention in industrial manufacturing systems.In recent years,many studies have aimed at saving energy consumption in the process of manufacturing and scheduling is regarded as an effective approach.This paper puts forwards a multi-objective stochastic parallel machine scheduling problem with the consideration of deteriorating and learning effects.In it,the real processing time of jobs is calculated by using their processing speed and normal processing time.To describe this problem in a mathematical way,amultiobjective stochastic programming model aiming at realizing makespan and energy consumption minimization is formulated.Furthermore,we develop a multi-objective multi-verse optimization combined with a stochastic simulation method to deal with it.In this approach,the multi-verse optimization is adopted to find favorable solutions from the huge solution domain,while the stochastic simulation method is employed to assess them.By conducting comparison experiments on test problems,it can be verified that the developed approach has better performance in coping with the considered problem,compared to two classic multi-objective evolutionary algorithms.

Optimized Selection Method of Air Combat Course of Action under Stochastic Uncertainty

Aiming at the design problem of aviation swarm combat course of action(COA),considering the influence of stochastic parameters in the causal relationship model and optimization problem model,according to the dynamic influence net(DIN)theory,stochastic simulation technique,feedforward neural network(FNN)function approximation technique and multi-objective artificial fish school algorithm(MOAFSA),this paper proposed a COA optimized method based on DIN and multi-objective stochastic chance constraint optimization for aviation swarm combat.First,on the basis of establishing the overall framework of the model and defining the elements of causal relationship modeling,the static and dynamic causal relationship modeling and optimization problem modeling were carried out respectively.Second,the probability propagation mechanism of DIN was established,which mainly included two aspects,i.e.,the overall process and the specific algorithm.Then,input and output data were generated based on stochastic simulation.According to these data,FNN was adopted for function approximation,and MOAFSA was adopted for iterative optimization.Finally,the rationality of the model,and the effectiveness and superiority of the algorithm were verified through multiple sets of simulation cases.

REAL CODED GENETIC ALGORITHM FOR STOCHASTIC HYDROTHERMAL GENERATION SCHEDULING

The intent of this paper is to schedule short-term hydrothermal system probabilistically considering stochastic operating cost curves for thermal power generation units and uncertainties in load demand and reservoir water inflows. Therefore, the stochastic multi-objective hydrothermal generation scheduling problem is formulated with explicit recognition of uncertainties in the system production cost coefficients and system load, which are treated as random variable. Fuzzy methodology has been exploited for solving a decision making problem involving multiplicity of objectives and selection criterion for best compromised solution. A real-coded genetic algorithm with arithmetic-average-bound-blend crossover and wavelet mutation operator is applied to solve short-term variable-head hydrothermal scheduling problem. Initial feasible solution has been obtained by implementing the random heuristic search. The search is performed within the operating generation limits. Equality constraints that satisfy the demand during each time interval are considered by introducing a slack thermal generating unit for each time interval. Whereas the equality constraint which satisfies the consumption of available water to its full extent for the whole scheduling period is considered by introducing slack hydro generating unit for a particular time interval. Operating limit violation by slack hydro and slack thermal generating unit is taken care using exterior penalty method. The effectiveness of the proposed method is demonstrated on two sample systems.