Fast Optimization Algorithm Research Articles

Fault-Tolerant Guidance of Rocket Vertical Landing Phase Based on MPC Framework

For the vertical landing process of reusable rockets, the landing accuracy is likely to be affected by disturbances and faults during flight. In this paper, a fault-tolerant guidance method based on the MPC framework is put forward. First, we propose a piecewise guidance algorithm that combines a trajectory optimization algorithm based on convex optimization with the MPC framework. With the fast trajectory optimization algorithm and the MPC framework that recursively introduces the real-time state, this algorithm forms a robust closed loop. Then, we design an integrated guidance, navigation, and control (GNC) system to enhance the fault tolerance and robustness of the guidance method. Simulation experiments verify that this method is fault-tolerant to various fault conditions including navigation system failures, control system failures, drag coefficient deviations, and atmospheric density deviations. This guidance method is robust enough to overcome disturbances and faults, and it has great potential for online use.

Guest Editorial Artificial Intelligence: New Frontiers in Real-Time Inverse Scattering and Electromagnetic Imaging

Understanding and solving complex problems in the physical world has been an intelligent endeavor of humankind. Moreover, the study of artificial intelligence (AI) embodies the dream of designing machines like humans. Research in deep-learning (DL) techniques has attracted much attention in many application areas. With the help of big data technology, massive parallel computing, and fast optimization algorithms, DL has greatly improved the performance of many problems in speech and image processing, power transportation networks, and bio-electromagnetics, among others.

Fast and on-line link optimization for the long-distance two-way fiber-optic time and frequency transfer.

The performance of long-distance two-way fiber-optic time and frequency (T/F) transfer is directly affected by link parameters. In this paper, a fast on-line scheme to maximize the signal-to-noise ratio at the T/F transfer link ends (denoted by SNRW and SNRE) is proposed by optimizing the gains of bidirectional Erbium-doped fiber amplifiers (Bi-EDFAs). The dedicatedly defined objective optimization functions, ln(1/SNRW + 1/SNRE) for microwave and time transfer, and ln(1/SNRW) for optical frequency transfer, are proven to be a convex function and enable the computation time to be reduced significantly by employing the fast and mature convex optimization algorithm. The linear equations about the objective optimization function and the fiber parameters required by optimization are constructed, and then the fiber parameters are proposed to be on-line obtained accordingly by measuring the concerned SNR at the link end under different EDFA gain configurations and solving the linear equations. The transmission matrixes for the proposed optimization scheme are further derived, which can simplify the implementation in the long-distance multiple EDFAs links. The simulation shows that the proposed algorithm can obtain the optimal parameters consistent with the reported experimental results, and can reduce the computation time by more than four orders compared with the scanning optimization method for a link with more than 6 Bi-EDFAs.

Fast atomic structure optimization with on-the-fly sparse Gaussian process potentials *

We apply on-the-fly machine learning potentials (MLPs) using the sparse Gaussian process regression (SGPR) algorithm for fast optimization of atomic structures. Great acceleration is achieved even in the context of a single local optimization. Although for finding the exact local minimum, due to limited accuracy of MLPs, switching to another algorithm may be needed. For random gold clusters, the forces are reduced to ∼0.1 eV Å−1 within less than ten first-principles (FP) calculations. Because of highly transferable MLPs, this algorithm is specially suitable for global optimization methods such as random or evolutionary structure searching or basin hopping. This is demonstrated by sequential optimization of random gold clusters for which, after only a few optimizations, FP calculations were rarely needed.

Dynamic Antenna Selection for Colocated MIMO Radar

Antenna distribution plays an important role for the performance gain in multiple-input–multiple-output (MIMO) radar target tracking. Since to meet the requirements of the low probability of interception, especially in a hostile environment, only a finite number of antennas can be activated at each step. This naturally leads to a performance-driven resource management problem. In this paper, a dynamic antenna selection strategy is proposed for tracking targets in colocated MIMO radar. The derived posterior Cramér–Rao lower bound (PCRLB) of joint direction-of-arrival (DOA) and Doppler estimate were chosen as the optimization criteria. Furthermore, in the deviation, the target radar cross-section (RCS) as the determining variable and the random variable are both discussed. The objective function is related to the antenna allocation and non-convex, and an efficient fast discrete particle swarm optimization (FDPSO) algorithm is proposed for the solution exploration. Additionally, a closed-loop feedback system is established, where the main idea is that the tracking information from the current time epoch is utilized to predict the PCRLB and to guide the antenna adjustment for the next time epoch. The simulation results demonstrate the performance improvement compared with the three fixed-antenna configurations. Moreover, the FDPSO can provide close-to-optimal solutions while satisfying the real-time demand.

Improving the solidification performance of a latent heat thermal energy storage unit using arrow-shaped fins obtained by an innovative fast optimization algorithm

Aiming to improve the solidification performance of the shell-and-tube latent heat thermal energy storage unit (LHTESU), a fin structure based on an innovative fast optimization algorithm is proposed. The fast optimization algorithm takes the distance from PCM to fin or cold source as the optimization objective. Genetic algorithm (GA) is used to find the minimum distance between PCM and fins or cold source. In order to deal with different types of fins, a unified method to calculate the distance between PCM and fins is designed. That is, the PCM region and the fin region are divided into smaller grids. The cold source/sink surface is also divided into smaller parts. The distance between PCM and fins or cold source is represented by the sum of the distances between each PCM grid and its nearest fin gird or cold source/sink surface. The cross-section structure of one-level Y-shaped fin in shell-and-tube LHTESU is optimized based on the proposed fast optimization algorithm. The length of each branch, the width ratio of sub-branch to main-branch, and the angle between the sub-branch are selected as the variables. The fin mass expressed by the fin cross area is selected as the constraint condition. Monte Carlo method is used to calculate the fin area due to the complex structure of Y-shaped fin. Meanwhile the effects of cold source and fin number on the final optimization results are also studied. The results show that the optimized fin has an arrow shape. When the number of fins is 2, compared with the traditional Y-shaped fin, the arrow-shaped fin can shorten the complete solidification time by more than 52.8% and increase the heat release rate by more than 110%.

Efficient independent vector extraction of dominant source (L).

The complete decomposition performed by blind source separation is computationally demanding and superfluous when only the speech of one specific target speaker is desired. This letter proposes a computationally efficient blind source extraction method based on the fast fixed-point optimization algorithm under the mild assumption that the average power of the source of interest outweighs the interfering sources. Moreover, a one-unit scaling operation is designed to solve the scaling ambiguity for source extraction. Experiments validate the efficacy of the proposed method in extracting the dominant source.

Functional group bridge for simultaneous regression and support estimation.

This paper is motivated by studying differential brain activities to multiple experimental condition presentations in intracranial electroencephalography (iEEG) experiments. Contrasting effects of experimental conditions are often zero in most regions and nonzero in some local regions, yielding locally sparse functions. Such studies are essentially a function-on-scalar regression problem, with interest being focused not only on estimating nonparametric functions but also on recovering the function supports. We propose a weighted group bridge approach for simultaneous function estimation and support recovery in function-on-scalar mixed effect models, while accounting for heterogeneity present in functional data. We use B-splines to transform sparsity of functions to its sparse vector counterpart of increasing dimension, and propose a fast nonconvex optimization algorithm using nested alternative direction method of multipliers (ADMM) for estimation. Large sample properties are established. In particular, we show that the estimated coefficient functions are rate optimal in the minimax sense under the L2 norm and resemble a phase transition phenomenon. For support estimation, we derive a convergence rate under the norm that leads to a selection consistency property under δ-sparsity, and obtain a result under strict sparsity using a simple sufficient regularity condition. An adjusted extended Bayesian information criterion is proposed for parameter tuning. The developed method is illustrated through simulations and an application to a novel iEEG data set to study multisensory integration.

Development of a back-propagation neural network combined with an adaptive multi-objective particle swarm optimizer algorithm for predicting and optimizing indoor CO2 and PM2.5 concentrations

People now spend between 80% and 90% of their lifetimes in indoor locations such as offices and residential buildings. These extended hours indoors have substantial health impacts, making it vital that people have adequate indoor air quality (IAQ). The management of IAQ presents two issues: (1) rapidly predicting and controlling IAQ using current intelligent ventilation systems is difficult, and (2) simultaneously controlling both gas pollutants (of which CO2 is a representative example) and particulate matter concentrations (for example, PM2.5) while achieving optimal outcomes is challenging. Therefore, this study aims to develop a fast and accurate optimization algorithm to simultaneously predict and control indoor CO2 and PM2.5 concentrations. To this end, a back-propagation neural network (BPNN) combined with an adaptive multi-objective particle swarm optimizer (AMOPSO) algorithm based on computational fluid dynamics (CFD) is proposed. The CFD model first creates a database of indoor CO2 and PM2.5 concentrations. Then, based on the CFD database, the BPNN model is used to predict indoor air pollutant concentrations. If the predicted concentrations do not meet predetermined limits, the AMOPSO algorithm is initialized to optimize the concentrations of the indoor air pollutants. In test examples, the proposed optimization algorithm reduces CO2 concentrations by up to 30.5%, while also reducing PM2.5 concentrations by as much as 77.1%.

Time-varying dual accelerated gradient ascent: A fast network optimization algorithm

We propose a time-varying dual accelerated gradient method for minimizing the average of n strongly convex and smooth functions over a time-varying network with n nodes. We prove that the time-varying dual accelerated gradient ascent method converges at an R-linear rate with the time to reach an ϵ-neighborhood of the solution being of O(1ln⁡(1/c)ln⁡Mϵ), where c is a constant depending on the graph and objective function parameters and M is a constant depending on the initial values. We test the proposed method on two classes of problems: L2-regularized least squares and logistic classification problems. For each class, we generate 1000 problems and use the Dolan-Moré performance profiles to compare our obtained results with the ones obtained by several state-of-the-art algorithms to illustrate the efficiency of our method.

An Efficient Riemannian Gradient Based Algorithm for Max-Cut Problems

The max-cut problem addresses the problem of finding a cut for a graph that splits the graph into two subsets of vertices so that the number of edges between these two subsets is as large as possible. However, this problem is NP-Hard, which may be solved by suboptimal algorithms. In this paper, we propose a fast and accurate Riemannian optimization algorithm for solving the max-cut problem. To do so, we develop a gradient descent algorithm and prove its convergence. Our simulation results show that the proposed method is extremely efficient on some already-investigated graphs. Specifically, our method is on average 50 times faster than the best well-known techniques with slightly losing the performance, which is on average 0.9729 of the max-cut value of the others.

CP-Squared: A method for change point detection in core–periphery networks

Time series of networks are increasingly prevalent in modern data and pose unique challenges to pattern extraction and change detection. In this paper we develop and present a novel methodology to detect regime changes within a sequence of networks that have overlapping and evolving community structure. The core of the methodology is a non-negative matrix factorization that maximizes a Poisson likelihood subject to a penalty that accounts for sparsity in the network. By fitting the factorization model over a rolling window with a fast numerical optimization algorithm, change detection is accomplished by statistical monitoring of the matrix factors’ evolution. A novel statistic is used to characterize the overall network evolution as well as the contribution of each node to the change. We demonstrate that the proposed methodology compares favorably with alternative techniques for on-the-go network change detection using synthetic and real data. A detailed case study on the 2007–2009 financial crisis and the European sovereign debt crisis shows the promise of the methodology for regulators as it identifies particular banks that contributed to each crisis in addition to identifying changing market conditions.

Multilevel Survival Modeling With Structured Penalties for Disease Prediction From Imaging Genetics Data.

This paper introduces a framework for disease prediction from multimodal genetic and imaging data. We propose a multilevel survival model which allows predicting the time of occurrence of a future disease state in patients initially exhibiting mild symptoms. This new multilevel setting allows modeling the interactions between genetic and imaging variables. This is in contrast with classical additive models which treat all modalities in the same manner and can result in undesirable elimination of specific modalities when their contributions are unbalanced. Moreover, the use of a survival model allows overcoming the limitations of previous approaches based on classification which consider a fixed time frame. Furthermore, we introduce specific penalties taking into account the structure of the different types of data, such as a group lasso penalty over the genetic modality and a l2-penalty over the imaging modality. Finally, we propose a fast optimization algorithm, based on a proximal gradient method. The approach was applied to the prediction of Alzheimer's disease (AD) among patients with mild cognitive impairment (MCI) based on genetic (single nucleotide polymorphisms - SNP) and imaging (anatomical MRI measures) data from the ADNI database. The experiments demonstrate the effectiveness of the method for predicting the time of conversion to AD. It revealed how genetic variants and brain imaging alterations interact in the prediction of future disease status. The approach is generic and could potentially be useful for the prediction of other diseases.

Label embedding semantic-guided hashing

Hashing technologies have been widely used for information retrieval tasks due to their efficient retrieval and storage capabilities. Generally, most of the current supervised learning only utilizes labels to construct a binary similarity matrix of instance pairs and ignores the rich semantic information contained in the labels. Indeed, the reason why supervised hashing is better than unsupervised hashing is that the labels itself has strong discriminative information. Therefore, how to effectively explore the label information is one of the ways to improve the performance of retrieval tasks. In addition, existing hashing methods have the problems of high time consumption and weak scalability when facing large-scale data. To remedy these problems, in this paper, we present a flexible two-step label embedding hashing method named Label Embedding Semantic-Guided Hashing (LESGH). In the first step, LESGH leverages an asymmetric discrete learning framework to learn discriminative compact hash codes only from label information, and adds the constraints of bit-balance and bit-decorrelation to boost the quality of the hash code generation. In the second step, LESGH learns the hash projection function through the generated hash codes in the first step. Moreover, an effective and fast iterative discrete optimization algorithm is presented to solve the discrete problem instead of using the relaxation-based scheme. In doing so, we can not only simplify the optimization process, but also easily scale to large-scale data. We conduct several experiments on three public datasets, i.e., WIKI, MIRFlickr and NUS-WIDE, demonstrate that LESGH can improve the retrieval performance over the compared state-of-the-art baselines.

A Grey Wolf Optimized 15-Level Inverter Design with Confined Switching Components

Multilevel inverters are a new class of dc-ac converters designed for high-power medium voltage and power applications as they work at high switching frequencies and in renewable applications by avoiding stresses like dv/dt and has low harmonic distortion in their output voltage. In variable speed drives and power generation systems, the use of multilevel inverters is obligatory. To estimate the switching positions in inverter configuration with low harmonic distortion value, a fast sequential optimization algorithm has been established. For harmonic reduction in multilevel inverter design, a hybrid optimization technique combining Firefly and the Genetic algorithm was used. In several real-time systems and for solving complex engineering problems, optimization approaches are gaining popularity. Based on Grey Wolf Optimization (GWO), this research paper proposes a novel multilevel inverter architecture. The GWO algorithm is based on the natural leadership hierarchy and hunting mechanism of grey wolves (Canis lupus). The Grey Wolf Optimizer (GWO) algorithm is used to find the best switching angles for a cascaded multilevel inverter in order to eliminate some high order harmonics while sustaining the desired fundamental voltage. The proposed inverter has 15 stages, and the circuit’s unique feature includes a limited switching device. This proposed method comprises, MOSFET-based switches well as three DC sources in the main circuit. The switching parameters of the inverter topology are tuned using GWO in this methodology. The THD value of the proposed system is reduced to 6.629% compared to that of Multilevel Inverter using Genetic Algorithm, standalone power supply, Firefly assisted Genetic Algorithm, hybrid APSO algorithm, DC voltage regulation and FPGA. A few simulation studies have been included in this paper to affirm the capability of the hybrid topologies with various condition parameters, dynamic changes, and modulation indexes.

Semiparametric maximum likelihood probability density estimation.

A comprehensive methodology for semiparametric probability density estimation is introduced and explored. The probability density is modelled by sequences of mostly regular or steep exponential families generated by flexible sets of basis functions, possibly including boundary terms. Parameters are estimated by global maximum likelihood without any roughness penalty. A statistically orthogonal formulation of the inference problem and a numerically stable and fast convex optimization algorithm for its solution are presented. Automatic model selection over the type and number of basis functions is performed with the Bayesian information criterion. The methodology can naturally be applied to densities supported on bounded, infinite or semi-infinite domains without boundary bias. Relationships to the truncated moment problem and the moment-constrained maximum entropy principle are discussed and a new theorem on the existence of solutions is contributed. The new technique compares very favourably to kernel density estimation, the diffusion estimator, finite mixture models and local likelihood density estimation across a diverse range of simulation and observation data sets. The semiparametric estimator combines a very small mean integrated squared error with a high degree of smoothness which allows for a robust and reliable detection of the modality of the probability density in terms of the number of modes and bumps.

Neural network-aided sparse convex optimization algorithm for fast DOA estimation

In this paper, a fast sparse convex optimization algorithm based on a neural network is proposed to improve the direction of arrival estimation. First, a fast [Formula: see text]-sparse representation of the array covariance vector model based on the Hermitian Toeplitz structure of array covariance is established to reduce computational complexity in data dimension and variable number. Then, the estimation error upper bound problem is investigated, and a neural network-aided coefficient selection method is developed. The direction of arrival estimation problem is solved through spectral peak search. Finally, the algorithm is extended to the case of off-grid error. The algorithm’s advantages in accuracy, calculation speed and robustness is verified by the simulations.

Whole-time scenario optimization of steam-assisted gravity drainage (SAGD) with temperature, pressure, and rate control using an efficient hybrid optimization technique

Whole-time (simultaneous) scenario optimization of oil recovery over a long time necessitates control of many parameters which is very time-consuming. Usually, to reduce the decision variables, scenarios are optimized in separate intervals. As a new work, a whole-time scenario optimization of the steam-assisted gravity drainage (SAGD) process with several decision parameters (steam injection rate, temperature, and pressure of all intervals) is performed in a 3D reservoir defining an 8-year scenario. Two relatively fast optimization algorithms, i.e. particle swarm optimization (PSO) and pattern search (PS), are utilized to maximize the net present value (NPV). First, through different optimizations, appropriate population size is selected for PSO. Next, the performance of four different scenarios is investigated, and the excellence of whole-time scenario optimization is proved. It is concluded that a higher NPV is obtained by increasing the number of time-intervals. Furthermore, conducting the whole-time optimization with 8 time-intervals can reduce the SAGD process by 1 year. Finally, a new hybrid PSO-PS algorithm is proposed that uses the advantages of both PSO and PS algorithms and reduces the number of function calls. The hybrid algorithm could result in the same results as PSO, while remarkably improved the convergence speed (about 84%).

Application of fast adaptive moth-flame optimization in flexible operation modeling for supercritical unit

As main measure to stabilize fluctuation of power grid under renewable energy utilization, the flexible operation of thermal power unit reveals remarkable significant. To serve for peak shaving and fast load varying, a novel T-S fuzzy modeling approach based on fast adaptive moth-flame optimization (FAMFO) algorithm is proposed for dynamics investigation of supercritical unit. This modeling scheme is deemed as hierarchical process with data partition stage and parameter determination stage. Firstly, an automatic clustering strategy relying on FAMFO is fulfilled for more reasonable data partition result, which devotes to accuracy modeling without subjective interference. In this stage, the FAMFO is attained with assistance of four improvements and its merits in balancing search exploration and exploitation guarantee the width of covered operation range of supercritical unit for peak shaving. Secondly, inherent advantages of least square technics in parameter identification witness the successful implementation of a well-performed exponentially-weighted least squares algorithm in later stage of fuzzy modeling. In effectiveness verification of the proposed modeling approach, the superiorities of FAMFO in promoting search precision and speed are demonstrated via benchmark function test and Friedman test. Then, the flexible operation modeling performance is proved via extensive simulation experiments using on-site data of supercritical unit.

A quantum-clustering optimization method for COVID-19 CT scan image segmentation

The World Health Organization (WHO) has declared Coronavirus Disease 2019 (COVID-19) as one of the highly contagious diseases and considered this epidemic as a global health emergency. Therefore, medical professionals urgently need an early diagnosis method for this new type of disease as soon as possible. In this research work, a new early screening method for the investigation of COVID-19 pneumonia using chest CT scan images has been introduced. For this purpose, a new image segmentation method based on K-means clustering algorithm (KMC) and novel fast forward quantum optimization algorithm (FFQOA) is proposed. The proposed method, called FFQOAK (FFQOA+KMC), initiates by clustering gray level values with the KMC algorithm and generating an optimal segmented image with the FFQOA. The main objective of the proposed FFQOAK is to segment the chest CT scan images so that infected regions can be accurately detected. The proposed method is verified and validated with different chest CT scan images of COVID-19 patients. The segmented images obtained using FFQOAK method are compared with various benchmark image segmentation methods. The proposed method achieves mean squared error, peak signal-to-noise ratio, Jaccard similarity coefficient and correlation coefficient of 712.30, 19.61, 0.90 and 0.91 in case of four experimental sets, namely Experimental_Set_1, Experimental_Set_2, Experimental_Set_3 and Experimental_Set_4, respectively. These four performance evaluation metrics show the effectiveness of FFQOAK method over these existing methods.