Electrophysiological/Magnetoencephalography source imaging(E/MSI) is widely used to noninvasively reconstruct brain sources from scalp measurements with high temporal *** of the key points of E/MSI is solving ill-cond...
详细信息
Electrophysiological/Magnetoencephalography source imaging(E/MSI) is widely used to noninvasively reconstruct brain sources from scalp measurements with high temporal *** of the key points of E/MSI is solving ill-conditioned equations,which highly depends on optimization *** optimizationalgorithms are more sensitive to the initial iteration points and their results are more influenced by the initial *** Particle Swarm optimization(QPSO) is an intelligent optimization algorithm with strong global optimization ability due to the randomness of particle *** this paper,the concept of population diversity was introduced,to control the population diversity of the QPSO,and then the DQPSO(QPSO with population diversity control) algorithm was *** maintaining a high population diversity in the whole iterative process of DQPSO,the global search capability was further *** the test and application of the algorithm were carried *** of all,the superiority of DQPSO algorithm in solving ill-conditioned equations was preliminarily verified through solving Hilbert ***,the DQPSO was used for the localization of simulated epileptic foci to deal with the ill-condition of the E/*** with the head structure information of high spatial resolution provided by MRI of brain structure,the personalized source imaging and localization with high spatial and temporal resolution was realized based on DQPSO algorithm in this *** results show that the loss of population diversity in the late iteration is avoided through the *** higher population diversity endows the DQPSO with a more powerful global traversal *** improves the accuracy and stability of the source imaging and localization compared with other algorithms,providing a more suitable tool for realizing E/MSI.
In this paper,for the synchronization control of the dynamically coupled fractional-order Rossler system,by reasonably selecting the optimization objective function and using the efficient global optimization ability ...
详细信息
In this paper,for the synchronization control of the dynamically coupled fractional-order Rossler system,by reasonably selecting the optimization objective function and using the efficient global optimization ability of the intelligent optimization algorithm,the optimal values of the error system parameters are directly determined;besides,a fractional integral optimal sliding mode control laws is designed,so that the parameters are changed from unknown to known,which avoids the construction of Lyapunov function according to the Lyapunov stability principle and the large amount of calculation of trial and error method,and provides an idea for us to determine the parameters in the system or *** the help of MATLAB Simulink,the fractional differential solver is obtained and the Simulink block diagrams of the systems is *** fast synchronization of the dynamically coupled drive and response system can be achieved by substituting the optimal parameter value into the *** numerical simulation results verify the feasibility and effectiveness of the proposed method.
"The last mile" is a key problem that needs to be solved urgently by major e-commerce companies, and also an important means to improve their competitiveness in e-commerce activities. In this paper, aiming a...
详细信息
"The last mile" is a key problem that needs to be solved urgently by major e-commerce companies, and also an important means to improve their competitiveness in e-commerce activities. In this paper, aiming at the problems existing in the current community intelligent logistics, from the perspective of the entire logistics system, combined with the actual distribution of logistics sites and distribution methods, we studied the scheduling problem of UAVs using different strategies when returning to the site after distribution. First, the UAV distribution time model is established, and then the model is solved with the minimum completion time of goods distribution as the goal. Finally, the feasibility of the algorithm is verified through simulation experiments, and the optimal scheduling strategy of UAV is given in combination with the actual operation, thus guiding e-commerce to build a reasonable intelligent logistics system.
Colleges and universities not only shoulder the major task of talent training, but also are responsible for recommending the trained talents to suitable positions. In the face of fierce competition for talents, how to...
详细信息
ISBN:
(纸本)9781450384100
Colleges and universities not only shoulder the major task of talent training, but also are responsible for recommending the trained talents to suitable positions. In the face of fierce competition for talents, how to effectively provide students with suitable jobs or employment information has become an urgent problem facing the employment departments of major universities. This article focuses on the research on the design and implementation of the employment management system for college students based on intelligent optimization algorithms. This article first investigated the needs of college students for employment services provided by the employment guidance center, summarized the functional needs of the employment management system, divided the system according to user roles and designed the system with corresponding functional modules. Finally, through the system function test and performance test, the reliability of the system is verified. Experimental data shows that when the number of users is 300, the response time for logging in to the system, information entry, and information query is 1.60s, 2.46s, and 1.66s, respectively. It can be seen that the system responds quickly and meets the needs of users.
We propose a new wavelength selection algorithm based on combined moving window(CMW) and variable dimension particle swarm optimization (VDPSO) algorithm. CMW retains the advantages of the moving window algorithm, and...
详细信息
We propose a new wavelength selection algorithm based on combined moving window(CMW) and variable dimension particle swarm optimization (VDPSO) algorithm. CMW retains the advantages of the moving window algorithm, and different windows can overlap each other to realize automatic optimization of spectral interval width and number. VDPSO algorithms improve the PSO algorithm. They can search the data space in different dimensions, and reduce the risk of limited local extrema and over fitting. Four different high-performance variable selection algorithms-BOSS, VCPA, iVISSA and IRF-are compared in three NIR data sets (corn, beer and fuel). The results show that VDPSO-CMW has better performance. The Matlab codes for implementing PSO-CWM and VDPSO-CMW are freely available on the website: https://***/matlabcentral/fileexchange/ 75828-a-variable-selection-method. (C) 2020 Elsevier B.V. All rights reserved.
In this era of rapid development, the development of a multi-level capital market is an urgent social concern. In the new era, new requirements are put forward for the study of portfolio optimization model in China. U...
详细信息
ISBN:
(纸本)9781728196190
In this era of rapid development, the development of a multi-level capital market is an urgent social concern. In the new era, new requirements are put forward for the study of portfolio optimization model in China. Under the background of big data era, if we want to further reduce and disperse the risk of financial investment market and increase the universality of portfolio investment optimization field in China, we must correctly deal with the impact and challenge brought by the era, combine with the new requirements of the new era, carry out reform and innovation on portfolio optimization, so as to promote its development. In order to better analyze the development progress of portfolio optimization problem;this paper puts forward a research method of intelligent optimization algorithm which applies big data technology to portfolio optimization problem, so as to put forward a set of new scheme of portfolio optimization problem which fully meets the development requirements of the new era. Through long-term research and analysis, it is found that the research scheme proposed in this paper successfully provides a new idea for the research method of intelligent optimization algorithm based on big data for combinatorial optimization problems.
With the development of the times, people's demand for oil and gas has been increasing;this trend has also led to the progress and development of oil and gas exploration and testing technology. As a new and scient...
详细信息
With the development of the times, people's demand for oil and gas has been increasing;this trend has also led to the progress and development of oil and gas exploration and testing technology. As a new and scientific theory, pre-stack AVO inversion technique is of great significance for oil and gas exploration, and the intelligent optimization algorithm also provides some opportunities for the development and progress of the technology. In view of the current situation of the increasing use of oil and gas in China and the not mature application of pre-stack AVO inversion technology, the technology was analyzed in this paper. The results show that the pre-stack AVO inversion technique is of positive significance to oil and gas exploration.
Big data is the inevitable outcome of the rapid development of modern information technology, effective analysis and processing of big data will not only bring great economic value, but will also promote social develo...
详细信息
Big data is the inevitable outcome of the rapid development of modern information technology, effective analysis and processing of big data will not only bring great economic value, but will also promote social development. Under the big data environment, the data scale, speed of emergence and its difficulty make optimizing issues very complex. In recent decades, genetic algorithm(GA), particle swarm optimization(PSO), ant colony optimization(ACO), Artificial Fish School algorithm, Bacteria Foraging optimizationalgorithm(BFOA), artificial neural networks(ANNs) and other multi-population intelligentalgorithms appeared. In this paper, several typical intelligent optimization algorithms are introduced, including genetic algorithm, particle swarm optimizationalgorithm, ant colony algorithm, artificial fish swarm algorithm and bacterial foraging algorithm. The basic principles of five algorithms are described respectively, along with the direction of improvement and feasible applications.
Given a set of potential sites for locating facilities, the disruptive anti-covering location problem (DACLP) seeks to find the minimum number of facilities that can be located on these sites in such a way that each p...
详细信息
ISBN:
(纸本)9783031248474;9783031248481
Given a set of potential sites for locating facilities, the disruptive anti-covering location problem (DACLP) seeks to find the minimum number of facilities that can be located on these sites in such a way that each pair of facilities are separated by a distance which is more than R from one another and no more facilities can be added. DACLP is closely related with anti-covering location problem (ACLP), which is concerned with finding the maximum number of facilities that can be located such that all the facilities are separated by a distance which is more than R from each other. The disruptive anti-covering location problem is so named because it prevents the "best or maximal" packing solution of the anti-covering location problem from occurring. DACLP is an NP-hard problem and plays an important role in solving many real world problems including but not limited to forest management, locating bank branches, nuclear power plants, franchise stores and military defence units. In contrast to ACLP, DACLP is introduced only recently and is a relatively under-studied problem. In this paper, two intelligentoptimization approaches namely genetic algorithm (GA) and discrete differential evolution (DDE) are proposed to solve the DACLP. These approaches are the first heuristic approaches for this problem. We have tested the proposed approaches on a total of 80 DACLP instances containing a maximum of 1577 potential sites. The effectiveness of the proposed approaches can be observed from the results on these instances.
Magnetic localization techniques are used in in vivo therapy, drug delivery, and surgical localization widely due to their ability to locate in an unobstructed environment precisely. The Levenberg-Marquardt (L-M) algo...
详细信息
Magnetic localization techniques are used in in vivo therapy, drug delivery, and surgical localization widely due to their ability to locate in an unobstructed environment precisely. The Levenberg-Marquardt (L-M) algorithm is one of the most common magnetic localization algorithms due to its fast optimization speed and high precision, but it requires high initial values such that improper initial values can easily lead to local optimization. In this paper, the performance of ten commonly used intelligent optimization algorithms is compared based on the mathematical model of the permanent magnet. The final position of the magnet is obtained by a fusion algorithm that uses the result of the intelligent optimization algorithm as the initial value of L-M. The real-time localization accuracy of this fusion algorithm is verified on a hardware platform and experiments demonstrate that the integration of the TSA algorithm and L-M algorithm fusion yields higher performance of the magnetic positioning system.
暂无评论