Comparative Analysis for Evaluating Wind Energy Resources Using Intelligent Optimization Algorithms and Numerical Methods

Statistical distributions are used to model wind speed,and the twoparameters Weibull distribution has proven its effectiveness at characterizing wind speed.Accurate estimation of Weibull parameters,the scale(c)and shape(k),is crucial in describing the actual wind speed data and evaluating the wind energy potential.Therefore,this study compares the most common conventional numerical(CN)estimation methods and the recent intelligent optimization algorithms(IOA)to show how precise estimation of c and k affects the wind energy resource assessments.In addition,this study conducts technical and economic feasibility studies for five sites in the northern part of Saudi Arabia,namely Aljouf,Rafha,Tabuk,Turaif,and Yanbo.Results exhibit that IOAs have better performance in attaining optimal Weibull parameters and provided an adequate description of the observed wind speed data.Also,with six wind turbine technologies rating between 1 and 3MW,the technical and economic assessment results reveal that the CN methods tend to overestimate the energy output and underestimate the cost of energy($/kWh)compared to the assessments by IOAs.The energy cost analyses show that Turaif is the windiest site,with an electricity cost of$0.016906/kWh.The highest wind energy output is obtained with the wind turbine having a rated power of 2.5 MW at all considered sites with electricity costs not exceeding$0.02739/kWh.Finally,the outcomes of this study exhibit the potential of wind energy in Saudi Arabia,and its environmental goals can be acquired by harvesting wind energy.

CALL FOR PAPERS Special Section on Intelligent Optimization and Scheduling

Due to various complexities in real-world systems,intelligent optimization algorithms have been successfully applied to complex problems in a variety of engineering fields which cannot be solved effectively by traditional methods.Inspired by the behavior,experience,and cognition from nature and society systems,data-driven intelligent optimization aims to achieve consistent results effectively and efficiently based on the mechanisms of computational intelligence.

Real-Time Optimization Model for Continuous Reforming Regenerator

An approach for the simulation and optimization of continuous catalyst-regenerative process of reforming is proposed in this paper.Compared to traditional method such as finite difference method,the orthogonal collocation method is less time-consuming and more accurate,which can meet the requirement of real-time optimization(RTO).In this paper,the equation-oriented method combined with the orthogonal collocation method and the finite difference method is adopted to build the RTO model for catalytic reforming regenerator.The orthogonal collocation method was adopted to discretize the differential equations and sequential quadratic programming(SQP)algorithm was used to solve the algebraic equations.The rate constants,active energy and reaction order were estimated,with the sum of relative errors between actual value and simulated value serving as optimization objective function.The model can quickly predict the fields of component concentration,temperature and pressure inside the regenerator under different conditions,as well as the real-time optimized conditions for industrial reforming regenerator.

Global Inverse Optimal Tracking Control of Underactuated Omni-directional Intelligent Navigators （ODINs）

This paper presents a design of optimal controllers with respect to a meaningful cost function to force an underactuated omni-directional intelligent navigator （ODIN） under unknown constant environmental loads to track a reference trajectory in two-dimensional space. Motivated by the vehicle＇s steering practice, the yaw angle regarded as a virtual control plus the surge thrust force are used to force the position of the vehicle to globally track its reference trajectory. The control design is based on several recent results developed for inverse optimal control and stability analysis of nonlinear systems, a new design of bounded disturbance observers, and backstepping and Lyapunov＇s direct methods. Both state- and output-feedback control designs are addressed. Simulations are included to illustrate the effectiveness of the proposed results.

Improved Quantum-Behaved Particle Swarm Optimization

To enhance the performance of quantum-behaved PSO, some improvements are proposed. First, an encoding method based on the Bloch sphere is presented. In this method, each particle carries three groups of Bloch coordinates of qubits, and these coordinates are actually the approximate solutions. The particles are updated by rotating qubits about an axis on the Bloch sphere, which can simultaneously adjust two parameters of qubits, and can automatically achieve the best matching of two adjustments. The optimization process is employed in the n-dimensional space [-1, 1]n, so this approach fits to many optimization problems. The experimental results show that this algorithm is superior to the original quantum-behaved PSO.

Adaptive Resource Planning for AI Workloads with Variable Real-Time Tasks

AI(Artificial Intelligence)workloads are proliferating in modernreal-time systems.As the tasks of AI workloads fluctuate over time,resourceplanning policies used for traditional fixed real-time tasks should be reexamined.In particular,it is difficult to immediately handle changes inreal-time tasks without violating the deadline constraints.To cope with thissituation,this paper analyzes the task situations of AI workloads and findsthe following two observations.First,resource planning for AI workloadsis a complicated search problem that requires much time for optimization.Second,although the task set of an AI workload may change over time,thepossible combinations of the task sets are known in advance.Based on theseobservations,this paper proposes a new resource planning scheme for AIworkloads that supports the re-planning of resources.Instead of generatingresource plans on the fly,the proposed scheme pre-determines resourceplans for various combinations of tasks.Thus,in any case,the workload isimmediately executed according to the resource plan maintained.Specifically,the proposed scheme maintains an optimized CPU(Central Processing Unit)and memory resource plan using genetic algorithms and applies it as soonas the workload changes.The proposed scheme is implemented in the opensourcesimulator SimRTS for the validation of its effectiveness.Simulationexperiments show that the proposed scheme reduces the energy consumptionof CPU and memory by 45.5%on average without deadline misses.

Distributed Intelligent Lighting System by Performing New Model for Illuminance and Color Temperature in the Workplace

A new approach has been proposed to improve the performance of the in-telligent lighting system by estimating personal illuminance and desired color temperature at the workplace. We are considering the problem of using the sensing devices manually for the intelligent lighting system. The lighting control system has not become useful without sensing devices to measure the provided illuminance and color temperature. In this paper, we have used the property of light for the color temperature to estimate the level of color temperature for each user at the workplace. The new method will give personal illuminance for each user at the workplace and decrease the power consumption of the environment as well. As a result, the proposed method of the intelligent lighting system has realized the target of illuminance and color temperature for each user at the workplace by adapting dimming levels using illuminance sensing information for each user. Thus, the energy of the workplace has reduced by using a distributed luminance to realize the target for each user.

Recent Advances in Particle Swarm Optimization for Large Scale Problems

Accompanied by the advent of current big data ages,the scales of real world optimization problems with many decisive design variables are becoming much larger.Up to date,how to develop new optimization algorithms for these large scale problems and how to expand the scalability of existing optimization algorithms have posed further challenges in the domain of bio-inspired computation.So addressing these complex large scale problems to produce truly useful results is one of the presently hottest topics.As a branch of the swarm intelligence based algorithms,particle swarm optimization (PSO) for coping with large scale problems and its expansively diverse applications have been in rapid development over the last decade years.This reviewpaper mainly presents its recent achievements and trends,and also highlights the existing unsolved challenging problems and key issues with a huge impact in order to encourage further more research in both large scale PSO theories and their applications in the forthcoming years.

Novel intelligent reasoning system for tool wear prediction and parameter optimization in intelligent milling

Accurate intelligent reasoning systems are vital for intelligent manufacturing.In this study,a new intelligent reasoning system was developed for milling processes to accurately predict tool wear and dynamically optimize machining parameters.The developed system consists of a self-learning algorithm with an improved particle swarm optimization(IPSO)learning algorithm,prediction model determined by an improved case-based reasoning(ICBR)method,and optimization model containing an improved adaptive neural fuzzy inference system(IANFIS)and IPSO.Experimental results showed that the IPSO algorithm exhibited the best global convergence performance.The ICBR method was observed to have a better performance in predicting tool wear than standard CBR methods.The IANFIS model,in combination with IPSO,enabled the optimization of multiple objectives,thus generating optimal milling parameters.This paper offers a practical approach to developing accurate intelligent reasoning systems for sustainable and intelligent manufacturing.

Intelligent methods for the process parameter determination of plastic injection molding

Injection molding is one of the most widely used material processing methods in producing plastic products with complex geometries and high precision. The determination of process parameters is important in obtaining qualified products and maintaining product quality. This article reviews the recent studies and developments of the intelligent methods applied in the process parameter determination of injection molding. These intelligent methods are classified into three categories： Case-based reasoning methods, expert sys- tem-based methods, and data fitting and optimization methods. A framework of process parameter determination is proposed after comprehensive discussions. Finally, the conclusions and future research topics are discussed.

A Chance Constrained Optimal Reserve Scheduling Approach for Economic Dispatch Considering Wind Penetration

The volatile wind power generation brings a full spectrum of problems to power system operation and management, ranging from transient system frequency fluctuation to steady state supply and demand balancing issue. In this paper, a novel wind integrated power system day-ahead economic dispatch model, with the consideration of generation and reserve cost is modelled and investigated. The proposed problem is first formulated as a chance constrained stochastic nonlinear programming U+0028 CCSNLP U+0029, and then transformed into a deterministic nonlinear programming U+0028 NLP U+0029. To tackle this NLP problem, a three-stage framework consists of particle swarm optimization U+0028 PSO U+0029, sequential quadratic programming U+0028 SQP U+0029 and Monte Carlo simulation U+0028 MCS U+0029 is proposed. The PSO is employed to heuristically search the line power flow limits, which are used by the SQP as constraints to solve the NLP problem. Then the solution from SQP is verified on benchmark system by using MCS. Finally, the verified results are feedback to the PSO as fitness value to update the particles. Simulation study on IEEE 30-bus system with wind power penetration is carried out, and the results demonstrate that the proposed dispatch model could be effectively solved by the proposed three-stage approach. © 2017 Chinese Association of Automation.

Segmented Real-time Dispatch Model and Stochastic Robust Optimization for Power-gas Integrated Systems with Wind Power Uncertainty

This paper develops a segmented real-time dispatch model for power-gas integrated systems(PGISs), where power-to-gas(P2G) devices and traditional automatic generation control units are cooperated to manage wind power uncertainty. To improve the economics of the real-time dispatch in regard to the current high operation cost of P2Gs, the wind power uncertainty set is divided into several segments, and a segmented linear decision rule is developed, which assigns adjustment tasks differently when wind power uncertainty falls into different segments. Thus, the P2G operation with high costs can be reduced in real-time adjustment. Besides, a novel segmented stochastic robust optimization is proposed to improve the efficiency and robustness of PGIS dispatch under wind power uncertainty, which minimizes the expected cost under the empirical wind power distribution and builds up the security constraints based on the robust optimization. The expected cost is formulated using a Nataf conversion-based multi-point estimate method, and the optimal number of estimate points is determined through sensitivity analysis. Furthermore, a difference-ofconvex optimization with a partial relaxation rule is developed to solve the non-convex dispatch problem in a sequential optimization framework. Numerical simulations in two testing cases validate the effectiveness of the proposed model and solving method.

深部地层智能压井多解性分析与优化策略

开发深部地层油气资源时普遍存在地质条件复杂、钻井周期长和井筒压力控制困难等问题,采用智能压井方法结合多源实时信息反馈,可实现井筒内气液分布状态和压力变化规律的实时预测与更新,但不同修正系数组合可能得到相同的压力计算结果,导致模型存在多解性难题。为此,分析了不同历史时间节点解空间形态的演变规律,揭示了模型多解性的本质源于少量数据约束下模型训练的不完善性;并对应建立了基于实时信息序列的模型全局训练优化方法及动态随机种群训练优化方法,测试了其对于模型全局最优解的搜索能力及适用条件。测试结果表明,全局训练优化方法在压井初期能够实现精准调控,但计算耗时较长;而动态随机种群训练优化方法在压井初期与预期值略有差异,但计算耗时较少。根据可用计算资源情况选择合适的训练优化方法,可实现多源实时数据约束下模型关于井筒气液流动规律的深度学习。

基于STM32的智能航行系统设计

为解决航行器智能确定航向并航行的问题,本文采取模块化的思想设计了一款基于STM32的红外导引的智能航行控制系统。该智能航行系统利用红外接收模块进行红外导引信号的接收,采取线性加权法进行航向与导引信号之间的偏离角确定,利用比例-积分-微分(proportion-integration-differentiation,PID)控制算法对控制舵机转角的脉冲宽度调制(pulse width modulation,PWM)信号的占空比进行确定,从而对螺旋桨的转速与舵桨的转角进行可靠控制。根据水池测试结果,该智能航行控制系统能够对红外接收信号进行快速处理,在保持较高航速的条件下进行可靠的航行控制。

Gas–solid reactor optimization based on EMMS-DPM simulation and machine learning

Design,scaling-up,and optimization of industrial reactors mainly depend on step-by-step experiments and engineering experience,which is usually time-consuming,high cost,and high risk.Although numerical simulation can reproduce high resolution details of hydrodynamics,thermal transfer,and reaction process in reactors,it is still challenging for industrial reactors due to huge computational cost.In this study,by combining the numerical simulation and artificial intelligence(AI)technology of machine learning(ML),a method is proposed to efficiently predict and optimize the performance of industrial reactors.A gas–solid fluidization reactor for the methanol to olefins process is taken as an example.1500 cases under different conditions are simulated by the coarse-grain discrete particle method based on the Energy-Minimization Multi-Scale model,and thus,the reactor performance data set is constructed.To develop an efficient reactor performance prediction model influenced by multiple factors,the ML method is established including the ensemble learning strategy and automatic hyperparameter optimization technique,which has better performance than the methods based on the artificial neural network.Furthermore,the operating conditions for highest yield of ethylene and propylene or lowest pressure drop are searched with the particle swarm optimization algorithm due to its strength to solve non-linear optimization problems.Results show that decreasing the methanol inflow rate and increasing the catalyst inventory can maximize the yield,while decreasing methanol the inflow rate and reducing the catalyst inventory can minimize the pressure drop.The two objectives are thus conflicting,and the practical operations need to be compromised under different circumstance.

水库群调度高维优化问题约束处理方法研究

随着水库群优化调度的调度规模的增加和调度时间步长的精细化,水库群调度高维优化问题的决策变量维度逐渐增加到数百数千维。在具有高维决策变量的梯级水库优化调度中,往往需要考虑多重复杂约束。现有传统优化方法在处理此类问题时难以找到有效可行解;而智能优化算法的多维度联动随机搜索,寻优空间大但寻优效率低。为此,本文提出了一种结合罚函数的嵌套DPSA–POA和智能算法的约束处理方法,将罚函数与DPSA–POA和智能算法嵌套,一方面可克服DPSA–POA收敛结果容易受初值影响和寻优空间狭窄的缺陷,另一方面可提升智能算法随机搜索策略的寻优效率。随后,本文以决策变量高达2 196维的赣江中游梯级水库群防洪优化调度问题为例开展分析,相关分析结果表明:1)结合罚函数嵌套DPSA–POA智能算法的3种约束处理方式,在不同来水情形下均能得到高维优化问题可行解;2)3种约束处理方式中,嵌套优化得到可行解后只进行DE优化的方式2收敛精度最高,计算时间约10 h;嵌套优化得到可行解后只进行DPSA–POA优化的方式3收敛精度次之,计算时间约1~3 h;3)现有可行解优先策略(SF)、随机排序策略(SR)、罚函数策略(PF)和ε–松弛约束策略(EC)配合现代智能算法,无法在不同来水情形下稳定收敛到可行解,且可行解的收敛精度相比本文提出的方法有明显差距。综上,本文提出的高维优化问题约束处理方法可有效解决水库群调度高维优化问题。

考虑资源弹性配置的配电网保护控制终端协同任务分配方法

随着新型电力系统的发展,利用智能终端处理愈发复杂的配电网保护控制任务时,对资源供给与需求的平衡要求越来越高。因此,文中提出一种考虑资源弹性配置的配电网保护控制终端(protect and control intelligent terminal,PCIT)协同任务优化分配方法。首先,阐述多终端协同的技术架构,并建立基于容器的PCIT的弹性资源模型、任务处理模型。其次,提出双层模型用于优化保护控制任务在终端间的协同分配、资源的弹性调度,并利用隐枚举法对该模型进行求解,从而充分发挥任务处理时资源的灵活性,提升任务处理性能。最后,算例验证了文中所提方法的可行性与先进性,各智能终端计算资源的占用率降低约28.85%,任务平均处理延时减少约4.12%。

基于分区选择的主动配电网分布式最优潮流分析

基于交替方向乘子法(ADMM)在分布式形式下解决主动配电系统最优潮流问题(OPF),针对分布式算法性能受到配电系统区域划分影响的问题,提出了一种基于量测数据驱动的电网分区方法,以加速优化算法的收敛速度。与传统的ADMM算法依赖于全局信息不同,本文引入了一致性方法来协调区域交界的平衡问题,从而实现最优潮流问题的完全分布式求解。此外,本文采用LinDistFlow(Linearized Distribution Flow)交流近似模型来应对配电网最优潮流问题的非凸性挑战。通过在不同规模的IEEE配电网案例上进行测试,验证了所提方法的有效性,且其在优化算法的迭代次数、计算时间和误差精度等性能上均优于其他分区方法。

GPU加速的演化算法求解多目标流水车间调度问题

智能制造和环境可持续性研究中,多目标调度问题对于协调生产效率、成本管理与环境保护之间的平衡具有至关重要的意义,但现有基于CPU的调度解决方案在处理大规模生产任务时仍面临效率和时效性的限制,而GPU的并行计算能力可为优化大规模流水车间调度问题提供新的解决途径。针对多目标零等待流水车间调度问题(NWFSP),以同时最小化最大完成时间和总能耗(TEC)为优化目标,构建了混合整数线性规划模型(MILP)表征该调度问题,并提出一种基于GPU加速的张量化演化算法(Tensor-GPU-NSGA-Ⅱ)求解该问题。Tensor-GPU-NSGA-Ⅱ的主要创新在于对NWFSP关于最小化最大完成时间和TEC的计算过程的张量化处理,并提出了一种基于GPU的并行种群更新方法。实验结果表明,在500工件和20机器的问题规模下,Tensor-GPU-NSGA-Ⅱ在计算效率上相较于传统NSGA-Ⅱ算法取得了9761.75的加速比;且随着种群规模的增加,它的加速性能有显著提升。

Real-time control for fuel-optimal Moon landing based on an interactive deep reinforcement learning algorithm

In this study,a real-time optimal control approach is proposed using an interactive deep reinforcement learning algorithm for the Moon fuel-optimal landing problem.Considering the remote communication restrictions and environmental uncertainties,advanced landing control techniques are demanded to meet the high requirements of real-time performance and autonomy in the Moon landing missions.Deep reinforcement learning(DRL)algorithms have been recently developed for real-time optimal control but suffer the obstacles of slow convergence and difficult reward function design.To address these problems,a DRL algorithm is developed using an actor-indirect method architecture to achieve the optimal control of the Moon landing mission.In this DRL algorithm,an indirect method is employed to generate the optimal control actions for the deep neural network(DNN)learning,while the trained DNNs provide good initial guesses for the indirect method to promote the efficiency of training data generation.Through sufficient learning of the state-action relationship,the trained DNNs can approximate the optimal actions and steer the spacecraft to the target in real time.Additionally,a nonlinear feedback controller is developed to improve the terminal landing accuracy.Numerical simulations are given to verify the effectiveness of the proposed DRL algorithm and demonstrate the performance of the developed optimal landing controller.