Dynamic Step Size Research Articles

Research on the Sustainability of Traditional Music and the Adaptability of Ecological Environment.

Under the new situation, the world is showing a trend of becoming more and more open, and the frequency of cultural exchanges in China is also accelerating. In this case, traditional music has been impacted to a certain extent, which also affects the sustainable use of traditional music. This is the general law of cultural development, and the impact is also accompanied by vitality. The analysis of the sustainability of traditional music and the adaptability of the ecological environment, how to accurately improve the matching results will directly affect the accuracy of the analysis of the sustainability of traditional music and the adaptability of the ecological environment. This study proposes an adaptive analysis method for the sustainability and ecological environment of traditional music. While analyzing the characteristics of traditional music in detail, the important characteristics of other types of music have also been fully considered. In this study, the adaptive evaluation method is used to obtain the optimal dynamic evaluation parameters. At the same time, the dynamic amplitude difference step size is set, and the adaptability analysis is performed according to the change threshold to improve the accuracy of the adaptability analysis of sustainability and the ecological environment. The adaptive evaluation method is used to limit the weight of lyrics and the semantic information of music. Finally, the experimental analysis shows that the traditional adaptive evaluation method and the algorithm of this study are experimentally tested. Adding adaptive evaluation methods further improves the accuracy of evaluation.

Incremental Quasi-Subgradient Method for Minimizing Sum of Geodesic Quasi-Convex Functions on Riemannian Manifolds with Applications

To find the optimal solution of a sum of geodesic quasi-convex functions, we introduce a new path incremental quasi-subgradient method in the setting of a Riemannian manifold whose sectional curvature is nonnegative. To study the convergence analysis of the proposed algorithm, some auxiliary results related to geodesic quasi-convex functions and an existence result for a Greenberg-Pierskalla quasi-subgradient of the geodesic quasi-convex function in the setting of Riemannian manifolds are established. The convergence result of the proposed algorithm with the dynamic step size is presented in the case when the optimal solution is unknown. To demonstrate practical applicability, we show that the proposed method can be used to find a solution of the (geodesic) quasi-convex feasibility problems and the sum of ratio problems in the setting of Riemannian manifolds.

Efficient and High Path Quality Autonomous Exploration and Trajectory Planning of UAV in an Unknown Environment

The ability of an autonomous Unmanned Aerial Vehicle (UAV) in an unknown environment is a prerequisite for its execution of complex tasks and is the main research direction in related fields. The autonomous navigation of UAVs in unknown environments requires solving the problem of autonomous exploration of the surrounding environment and path planning, which determines whether the drones can complete mission-based flights safely and efficiently. Existing UAV autonomous flight systems hardly perform well in terms of efficient exploration and flight trajectory quality. This paper establishes an integrated solution for autonomous exploration and path planning. In terms of autonomous exploration, frontier-based and sampling-based exploration strategies are integrated to achieve fast and effective exploration performance. In the study of path planning in complex environments, an advanced Rapidly Exploring Random Tree (RRT) algorithm combining the adaptive weights and dynamic step size is proposed, which effectively solves the problem of balancing flight time and trajectory quality. Then, this paper uses the Hermite difference polynomial to optimization the trajectory generated by the RRT algorithm. We named proposed UAV autonomous flight system as Frontier and Sampling-based Exploration and Advanced RRT Planner system (FSEPlanner). Simulation performs in both apartment and maze environment, and results show that the proposed FSEPlanner algorithm achieves greatly improved time consumption and path distances, and the smoothed path is more in line with the actual flight needs of a UAV.

Optimization and buckling of rupture building beams reinforced by steel fibers on the basis of adaptive improved harmony search-harmonic differential quadrature methods

The optimization and buckling analysis of the rupture beams reinforced by steel fibers are presented in this article. The structure is simulated by hyperbolic shear deformation theory (HSDT). Utilizing Harmonic differential quadrature (HDQ) method, the buckling load is calculated. Utilizing adaptive improved harmony search (AIHS) algorithm, the optimization process of the rupture beam is examined. The mentioned algorithm is improved adaptively utilizing dynamic variables. The harmony memory is corrected at first and second adjustments, respectively on the basis of dynamic bandwidth and step size randomly. The optimum conditions of the rupture beam are evaluated with various harmony existing search methods. The influences of various constants including beam length to thickness ratio, rupture parameter, steel fibers volume fraction, steel fibers orientation angle and boundary supports are shown on the buckling load. The outcomes indicate that the proposed AIHS is performed better with respect to other harmony search methods. Meanwhile, the buckling load is reduced by enhancing the rupture parameter. Furthermore, the buckling load of the rupture beam with horizontal steel fibers is better with respect to other orientation angles.

Robust Adaptive HCS MPPT Algorithm-Based Wind Generation System Using Model Reference Adaptive Control.

To treat the stochastic wind nature, it is required to attain all available power from the wind energy conversion system (WECS). Therefore, several maximum power point tracking (MPPT) techniques are utilized. Among them, hill-climbing search (HCS) techniques are widely implemented owing to their various features. Regarding current HCS techniques, the rotor speed is mainly perturbed using predefined constants or objective functions, which makes the selection of step sizes a multifaceted task. These limitations are directly reflected in the overall dynamic WECS performance such as tracking speed, power fluctuations, and system efficiency. To deal with the challenges of the existing HCS techniques, this paper proposes a new adaptive HCS (AD-HCS) technique with self-adjustable step size using model reference adaptive control (MRAC) based on the PID controller. Firstly, the mechanical power fluctuations are detected, then the MRAC continuously optimizes the PID gains so as to generate an appropriate dynamic step size until harvesting the maximum power point (MPP) under the optimal tracking conditions. Looking specifically at the simulation results, the proposed AD-HCS technique exhibits low oscillations around the MPP and a small settling time. Moreover, WECS efficiency is increased by 5% and 2% compared to the conventional and recent HCS techniques, respectively. Finally, the studied system is confirmed over a 1.5 MW, gird-tied, double-fed induction generator (DFIG) WECS using MATLAB/Simulink.

Decentralized optimal scheduling for integrated community energy system via consensus-based alternating direction method of multipliers

To transform the original centralized scheduling framework of integrated community energy systems (ICES), a decentralized scheduling method is proposed for the optimal operation of ICES. Firstly, linearized mathematical models of the electric distribution system and natural gas distribution system are established to eliminate the nonconvexity and nonlinearity of the multi-energy systems. Meanwhile, an improved energy hub model is proposed by introducing the state variable, which is able to obtain the optimal solution with better accuracy and higher efficiency. Secondly, a decentralized scheduling framework is developed via an improved consensus-based alternating direction method of multipliers (ADMM). By introducing several coupling links as consensus variables, the optimal operation of the ICES is decoupled according to the physical interactions among each sub-system, in which the original model can be solved parallelly in a decentralized manner. Only the consensus variables are exchanged among sub-systems so that the privacy and confidential items of each sub-system operated by different entities can be ensured. Finally, two different ICES test cases with different system scales are utilized to demonstrate the effectiveness of the proposed method, which is able to provide scheduling scheme same as the centralized method. Dynamic step size modification is further considered in the consensus-based ADMM. By updating the step size during each iteration, the computation expense of optimization process is reduced with less consuming time, iteration and apparent oscillation, by which the convergence performance of ADMM is improved with higher computation efficiency.

Path planning for UAV based on improved dynamic step RRT algorithm

Path planning is an important problem in the field of unmanned aerial vehicles (UAVs). However, it is inefficient for many rapidly-exploring random tree (RRT) based methods to rapidly find a feasible solution in a complex environment. To solve the path planning problem for the UVA in a complex environment, an improved dynamic step size RRT algorithm combined with a new path length control strategy is proposed. Firstly, the algorithm adopts a biased-goal sampling strategy to guide the growth of the tree, and an expansion direction constraint strategy is adopted to limit the expansion direction of the tree. Secondly, an improved dynamic step size strategy is proposed to speed up the path searching process. Thirdly, a path length constraint strategy is designed, which is used to constrain the path length during the searching process for a solution. Finally, simulation results demonstrate that the proposed method achieves improvement in both computational time and the path quality.

An Efficient Maximum Power Point Tracking Algorithm for DFIG based Wind Energy Conversion System

This paper proposes a new adaptive perturb and observe (AD-PO) algorithm for maximum power point tracking (MPPT) of DFIG based Wind Energy Conversion Systems (WECSs).The proposed algorithm is able to solve the limitations of conventional fixed step-size perturb and observe (PO) algorithm. It is able to achieve the optimum power with small time and oscillations compared with the conventional P&O techniques. This algorithm is mainly based on dividing the power-speed curve into four sectors by comparing the power-speed curve and a special synthesize curve. Hence, the selection of appropriate dynamic perturbation step-size (DPSS) of rotor speed is changed related to the operating sector. For the two sectors far from the maximum power point (MPP), a large DPSS is applied. Otherwise, the controller utilizes a small DPSS. Theproposed algorithm is validated using a large-scale 1.5 MW double-fed induction generator (DFIG); where the stator terminals of DFIG are connected directly to the grid and the rotor terminals of DFIG are connected to the electric gridvia a back-to-back converter (BTBC). The proposed algorithm has been implemented using MATLAB/SIMULINK. The results prove the efficient operation of the proposed algorithm in term of the fast system response compared with traditional P&O techniques.

The optimized GRNN based on the FDS-FOA under the hesitant fuzzy environment and its application in air quality index prediction

The generalized regression neural network (GRNN) is one of the most effective neural models in the field of information prediction. In this paper, we take advantage of the hesitant fuzzy set (HFS) and develop the GRNN under the hesitant fuzzy environment. However, the performance of the GRNN only depends on the smooth factor after determining the input and output variables. To determine the most appropriate smooth factor of network, we propose an improved fruit fly optimization algorithm with fast decreasing step (FDS-FOA) based on a dynamic step size function. A specific implementation process of the optimized GRNN based on FDS-FOA is also presented. Moreover, we apply the proposed model to the prediction of air quality index (AQI) in Beijing. Comparative analysis and sensitivity analysis are further conducted to illustrate the advantages of the optimized GRNN based on FDS-FOA.

Subgradient method with feasible inexact projections for constrained convex optimization problems

In this paper, we propose a new inexact version of the projected subgradient method to solve nondifferentiable constrained convex optimization problems. The method combine ϵ-subgradient method with a procedure to obtain a feasible inexact projection onto the constraint set. Asymptotic convergence results and iteration-complexity bounds for the sequence generated by the method employing the well-known exogenous stepsizes, Polyak's stepsizes, and dynamic stepsizes are established.

Noise segmentation for step detection and distance estimation using smartphone sensor data

As for the low-cost and accuracy benefits of integrating Artificial Intelligence (AI) technologies in smartphones, the embedded smart sensors have been utilized in indoor localization and distance estimation applications rather than the requirement of wearable sensors. However, the effect of a potential noise during the user’s movement is still an issue, i.e. irrelevant movements. Knowing that the noise can severely impair the accuracy of distance estimation by increasing or decreasing the number of estimated steps. Therefore, this paper proposes a distance estimation method based on dynamic step size and shape. It aims at removing the noise from the bit signal and detecting the number of steps regardless of the noise using the magnitude value for the three-axis accelerometer data. Furthermore, it involves the variables of the user's height with relation to the step size to estimate the walking distance. The experiment sets allow four phases including data collection and filtering, peak detection, dynamic step size detection, and distance estimation. The low-pass filter is applied to remove the resultant noise of irrelevant movements such as handshaking, texting, swinging, etc. The experimental results showed that the proposed method of step detection, step size, and walking distance estimations achieved significant accuracy improvement over existing current approaches reaches (98%) when a pedestrian was walking in multiple mode changes.

Distributed Subgradient Algorithm for Multi-Agent Optimization With Dynamic Stepsize

In this paper, we consider distributed convex optimization problems on multi-agent networks. We develop and analyze the distributed gradient method which allows each agent to compute its dynamic stepsize by utilizing the time-varying estimate of the local function value at the global optimal solution. Our approach can be applied to both synchronous and asynchronous communication protocols. Specifically, we propose the distributed subgradient with uncoordinated dynamic stepsizes (DS-UD) algorithm for synchronous protocol and the AsynDGD algorithm for asynchronous protocol. Theoretical analysis shows that the proposed algorithms guarantee that all agents reach a consensus on the solution to the multi-agent optimization problem. Moreover, the proposed approach with dynamic stepsizes eliminates the requirement of diminishing stepsize in existing works. Numerical examples of distributed estimation in sensor networks are provided to illustrate the effectiveness of the proposed approach.

Distributed optimization method with weighted gradients for economic dispatch problem of multi-microgrid systems

After microgrids (MGs) are built in a near region, a good idea is interconnecting those neighboring MGs to form a multi-microgrid (MMG) system and support each other. Apparently, the economic dispatch problem (EDP) of an MMG system is much more complex than that of an MG. In this paper, the EDPs of an MMG system is formulated as a two-layer interdependent network, where the bottom layer is a network that energy flows (called energy network), while the top layer is a network that information exchanges (called information network). The information network consists of two types of subnetworks, within-MG subnetworks and between-MG subnetworks, in which on within-MG subnetworks, the optimal outputs of distributed generators (DGs) are achieved and the supply-demand balance is reached in an MG, while on between-MG subnetworks, the outputs of MGs are coordinated to support one another optimally in the MMG system. Further, a distributed optimization method with weighted gradients (DOWG) is proposed to solve the EDPs of the MMG system on the information network, where equality constraints in optimization problems are dealt with by the weighted matrix and dynamic step sizes are employed to achieve faster convergence rate. Furthermore, two propositions are proved, which ensure the supply-demand balance is not broken at an arbitrary iteration. Finally, Simulations are carried out on the MMG system built in MATLAB/Simulink. The results show that the convergence rate of the proposed method (DOWG) is faster and higher accuracies are obtained, compared to two other methods. Moreover, applying the proposed method, the optimal outputs of DGs are obtained in MGs and the MMG system is coordinated optimally, when both loads and environmental conditions fluctuate largely. More importantly, the proposed method still can run the MMG system well, even if the failures of agents occur.

Structural Damage Identification Based on Improved Fruit Fly Optimization Algorithm

To locate the damage of the structure efficiently and judge the damage degree, this paper proposes an improved fruit fly optimization algorithm (IFOA). Aiming at the problem of poor convergence of the standard fruit fly optimization algorithm in the face of complex structure damage identification, the IFOA introduces the concept of collaborative search of two subpopulations. The IFOA divides the entire population into positive subgroups and negative subgroups based on the individual taste concentration results. Among them, the positive subgroup uses the improved dynamic adaptive search step size to perform a fine search locally to improve its local search ability. Negative subgroups continue to use the standard fruit fly optimization algorithm for optimization, taking advantage of the powerful global search capabilities of the standard fruit fly optimization algorithm. It enables the algorithm to balance global and local search capabilities, prevents the algorithm from falling into local optimum, and speeds up the convergence speed and accuracy of the algorithm. Simulation results show that IFOA can effectively identify the damage location and damage degree of the structure, and it still performs well when facing the complex steel truss damage identification.

PV Model Parameter Estimation Using Modified FPA With Dynamic Switch Probability and Step Size Function

The development of highly efficient models of Photovoltaic (PV) cells and modules is essential for optimized performance, evaluation and control of solar PV systems. The accurate estimation of PV cells parameters is a challenging task because of their non-linear characteristics. In this paper, an improved variant of Flower Pollination Algorithm (FPA) is proposed for accurate estimation of PV cells and modules parameters. The proposed algorithm involves double exponential based dynamic switch probability and a dynamic step size function that mitigate the limitations of conventional FPA. The dynamic switch probability improves the overall performance of algorithm by maintaining a balance between local and global search, while dynamic step function controls the search speed which avoids premature convergence and local optima stagnation. Moreover, Newton Raphson Method is utilized for accurate computation of estimated current for optimum set of estimated parameters. The proposed methodology is evaluated using seven benchmark functions and three case studies; 1- RTC France silicon PV cell, 2- Photo-watt PWP-201 PV module and 3- a practical solar PV system (EAGLE PERC 60M 310W monocrystalline PV module) under different environmental conditions by estimating parameters for single and double diode models. The analysis of results indicates that, the proposed approach improves the convergence speed, precision, avoids premature convergence and stagnation in local optima of conventional FPA. Furthermore, comparative analysis of results illustrates that, the proposed approach is more reliable and efficient than many other techniques in literature.

Fast distributed Lagrange dual method based on accelerated gradients for economic dispatch of microgrids

Distributed optimization algorithms have long been criticized due to their slow convergence rates mostly caused by fixed step-sizes. In this paper, a fast distributed Lagrange dual method (FDLDM) has been proposed, where the Nesterov accelerated gradient is integrated and dynamic step-sizes are employed, which is the key to achieve the fast convergence rate of our method. However, the economic dispatch problem (EDP) of microgrids (MGs) is a thorny optimization problem because of a lot of distributed generators (DGs) in MGs. Fortunately, it can be modeled and solved by our method quickly and effectively. The results show that the convergence rate of our method is faster than that of the Distributed Lagrange dual method (DLDM), and the economic dispatch of an MG can be achieved, even if both loads and environmental conditions fluctuate significantly.

Moving obstacle avoidance for cable-driven parallel robots using improved RRT

Moving obstacle avoidance is one of the most challenging problems for cable-driven parallel robots (CDPRs) due to various constraints. In this work, the improved rapidly exploring random tree (RRT) method is proposed to address moving obstacle avoidance for CDPRs. Compared with the conventional RRT method which mainly focused on the static environment, the suggested method is goal-biased with dynamic step size makes it possible to implement in a dynamic environment. To deal with the particularity of CDPRs, the improved RRT method considers various constraints caused by the cable particularity, which include cable interference and the twist feasible workspace. The swept volume of CDPRs is considered to convert the spatiotemporal problem into a pure spatial problem, therefore the Gilbert–Johnson–Keerthi algorithm was applied to collision detection. Additionally, the velocity information is utilized to estimate time and distance of the closest approach, which also prevents the local minima. The simulation is conducted to illustrate the suggested method and the simulation results are compared with our previous APG-RRT method using batch evaluation. According to the simulation results, the suggested method finds an optimized collision-free path with the average path cost reduced by 32%, and the oscillation time is reduced by 36%. Finally, the suggested method is verified by the experiment.

Dynamically swarm shared mutation based bacterial foraging

In this paper, dynamically swarm shared mutation based Bacterial Foraging (DSSBFO) is proposed to optimize multidimensional, unimodal and multimodal functions. In BFO, due to fixed step size it requires more computational cost to get optimum solution with better accuracy. Chemotaxis and reproduction step of BFO are not sufficient for an effective search. So in this paper, the authors propose dynamic step size in BFO to achieve optimum solution with better accuracy with minimum cost. The dynamic step size i.e. mutation is achieved by modifying the position equation of GLBestPSO and momentum factor (mc) of SSMPSO used in modified equation to bring the bacteria in search space and not to cross the boundary of search space. The eight standard benchmark functions are used to prove the performance of DSSBFO in terms of precision and cost. DSSBFO performs well as compared to BFO and BSO (BFO hybridized with PSO) alogrithms interms of quality solution with faster convergence.

A novel local search method for LSGO with golden ratio and dynamic search step

Depending on the developing technology, large-scale problems have emerged in many areas such as business, science, and engineering. Therefore, large-scale optimization problems and solution techniques have become an important research field. One of the most effective methods used in this research field is memetic algorithm which is the combination of evolutionary algorithms and local search methods. The local search method is an important part that greatly affects the memetic algorithm’s performance. In this paper, a novel local search method which can be used in memetic algorithms is proposed. This local search method is named as golden ratio guided local search with dynamic step size (GRGLS). To evaluate the performance of proposed local search method, two different performance evaluations were performed. In the first evaluation, memetic success history-based adaptive differential evolution with linear population size reduction and semi-parameter adaptation (MLSHADE-SPA) was chosen as the main framework and comparison is made between three local search methods which are GRGLS, multiple trajectory search local search (MTS-LS1) and modified multiple trajectory search. In the second evaluation, the improved MLSHADE-SPA (IMLSHADE-SPA) framework which is a combination of MLSHADE-SPA framework and proposed local search method (GRGLS) was compared with some recently proposed nine algorithms. Both of the experiments were performed using CEC’2013 benchmark set designed for large-scale global optimization. In general terms, the proposed method achieves good results in all functions, but it performs superior on overlapping and non-separable functions.

Wavelet neural network with SOA based on dynamic adaptive search step size for network traffic prediction

The seeker optimization algorithm (SOA) is a heuristic random search algorithm that has fast convergence. SOA has good robustness and can simulate the human search behavior. In the traditional SOA, global and local searches are adjusted by linearly decreasing the inertia weights; however, this algorithm has poor local search ability and adaptive adjustment ability in the initial iteration stage. The traditional SOA also easily falls in the local optimum. To overcome the shortcomings of the SOA, we propose an SOA based on the dynamic adaptive search step (hereinafter DASSS–SOA). This algorithm dynamically adjusts the inertia weight by introducing the objective function value of the current population to enhance the local search ability of the algorithm. The effectiveness of DASSS–SOA was verified by testing and analyzing the typical functions. Finally, to optimize the weight of the wavelet neural network (WNN) and predict the network traffic, the DASSS–SOA and WNN were combined (hereinafter DASSS–SOA–WNN). The simulation results showed that DASSS–SOA–WNN had a higher prediction accuracy than the WNN and SOA–WNN prediction models.