Heuristic Particle Swarm Optimization Algorithm Research Articles

A cooperative image object recognition framework and task offloading optimization in edge computing

The development of AIoT (Artificial Intelligence of Things) and communication technologies has recently brought with it many promising applications. Intelligent surveillance is one of the new models of applications that synthesizes both technologies of IoTs and the object recognition method of artificial intelligence. However, this system suffers from the heavy computation load problem, but the processing devices are generally lightweight with only limited memory and computation power. In this paper, we propose a cooperative image object recognition framework to integrate the process of AIoT devices into the edge computing environment to overcome this problem. The proposed scheme offloads some captured surveillance video frames to nearby GPU equipped edge servers and then integrates the returned results to enhance the overall recognition accuracy of the surveillance system. We also formulate the AIoT Task Offloading scheduling to a mathematical programming Problem (named ATOP) and prove it to be an NP-complete problem. A heuristic Particle Swarm Optimization algorithm is also proposed to determine near optimal offloading scheduling for the ATOP such that the overall object recognition precision is maximized. Although the object recognition precision is enhanced, there still remains a gap to the ground truth, and the cumulative gap might cause significant errors as times goes by. In this research, we also develop a mechanism to help the designer to automatically diagnose when the cumulative errors become significant. Simulation results demonstrate that the proposed framework is feasible and the proposed Particle Swarm Optimization algorithm for the ATOP outperforms the greedy approach and the genetic algorithm. The simulation results also show that the abnormal data diagnosis mechanism can figure out when the cumulative errors happen using a real-world human-flow counting data set.

Particle Swarm Optimization-Based Fuzzy PID Controller for Stable Control of Active Magnetic Bearing System

In this paper, the application of particle swarm optimization (PSO) algorithm based fuzzy PID control for the stable control of an active magnetic bearing (AMB) system is studied. Active magnetic bearings are known to be highly nonlinear multivariable systems. An AMB system is used in motors, turbines and various other machineries in different industries to provide an active suspension to the rotor shafts. The heuristic PSO algorithm is applied to optimize the parameters of the PID controller offline. The fuzzy PID controller based on PSO algorithm is designed to adjust the control parameters of AMB system online. The comparison of controlled responses of closed-loop systems resulting from the use of conventional PID, Incomplete differential PID and particle swarm optimization-based fuzzy logic PID control strategies is discussed. The proposed controller performance is superior as compared to PID and incomplete differential PID controllers in improving the system response of an AMB under normal conditions as well as under external disturbances.

Estimation of Structural Parameters Using Transfer Matrices and State Vectors

A new system identification (SI) method based on Transfer Matrix (TM) concept is proposed here to identify structural stiffness. Transfer matrices based on lumped mass and the more accurate consistent mass models are used here, based on displacement measurements in the time domain. The consistent mass based TM is derived from the dynamic stiffness matrix of a beam element. The state vector at a location is the sum of the internal and external contributions of displacements, forces and moments at that point - when multiplied with the (TM), we obtain the adjacent state vectors. The method of identification proposed here involves predicting the displacements at certain locations using the TM, and comparing them with the measured displacements at those locations. The mean square deviations between the measured and predicted responses at all locations are minimized using an optimization algorithm, and the optimization variables are the unknown stiffness parameters in the TM. A non-classical heuristic Particle Swarm Optimization algorithm (PSO) is used, since it is especially suited for global search. Different strategies for calculating the initial state vector, as well as two identification processes viz., simultaneous and successive methods, are discussed. Numerical simulations are carried out on four examples ranging from a simple spring mass system to a nine member framed structure. The speed and accuracy of identification using this method are good. One main advantage of this method is that it can be applied at any portion of the structure to identify the local parameters in that zone without the need to model the entire global structure.

Performance Assessment of a Coupled Particle Swarm Optimization and Network Flow Programming Model for Optimum Water Allocation

Water resources allocation problems are mainly categorized in two classes of simulation and optimization. In most cases, optimization problems due to the number of variables, constraints and nonlinear feasible search space are known as a challenging subject in the literature. In this research, by coupling particle swarm optimization (PSO) algorithm and a network flow programming (NFP) based river basin simulation model, a PSO-NFP hybrid structure is constructed for optimum water allocation planning. In the PSO-NFP model, the NFP core roles as the fast inner simulation engine for finding optimum values for a large number of water discharges in the network links (rivers and canals) and nodes (reservoirs and demands) while the heuristic PSO algorithm forms the outer optimization cover to search for the optimum values of reservoirs capacities and their storage priorities. In order to assess the performance of the PSO-NFP model, three hypothetical test problems are defined, and their equivalent nonlinear mathematical programs are developed in LINGO and the results are compared. Finally, the PSO-NFP model is applied in solving a real river basin water allocation problem. Results indicate that the applied method of coupling PSO and NFP has an efficient ability for handling river basin-scale water resources optimization problems.

A novel nonlinear model predictive control design based on a hybrid particle swarm optimization-sequential quadratic programming algorithm: application to an evaporator system

This paper proposes a hybrid algorithm by combining particle swarm optimization (PSO) with sequential quadratic programming (SQP) to handle the optimization part of the nonlinear model predictive control (NMPC). In the proposed method, a nonlinear model of the plant is directly applied to predict the future behaviour of the system for a certain horizon. In each sampling time a constrained nonlinear optimization (CNO) problem has to be solved in order to find the optimizing input sequence with predefined length. These solutions minimize the errors between the predicted states and the desired ones of the system. Only the first optimizing input is applied to the system and the rest of the sequence is discarded. The whole prediction and optimization must be repeated for the next upcoming steps. The evolutionary and heuristic PSO algorithm in cooperation with the powerful local minimizing SQP algorithm can quickly find these optimizing solutions and also can satisfy the constrains. The evaporation process due to its nonlinear dynamics and the existence of disturbances affecting the evaporator is considered to evaluate the performance of the proposed method. The simulation results show that the controller designed via the proposed approach can make the system track the reference commands and satisfy the restrictions properly. Moreover, based on the simulation results it can be seen that this approach is more efficient in comparison with the NMPC method, in which only the SQP algorithm has been utilized for optimization, and also the linear model predictive control method.

Heuristic particle swarm optimization approach for test point selection with imperfect test

The problem of near-optimal test point set selection with imperfect test is solved by using the heuristic particle swarm optimization (HPSO) algorithm. First, to describe the uncertainty of each test, the testability analysis model and such indexes as fault detection rate, fault isolation rate, and false alarm rate are redefined. A heuristic function is then established to evaluate the detection isolation capability and uncertainty of the test point, which can provide heuristic information to improve the searching efficiency of particle swarm optimization (PSO). The heuristic function and least test cost principle are used as bases to design a fitness function of PSO algorithm for test point selection. Finally, the HPSO algorithm is proposed to select the optimal test point set for two practical systems. Simulation and experiment results show that the method can determine the global optimal test point accurately and effectively while meeting the requirements of testability indexes with least cost.

A linked simulation–optimization model for simultaneously estimating the Manning’s surface roughness values and their parameter structures in shallow water flows

• Parameter structures of the Manning’s surface roughness is identified. • The first application that integrates the HEC-RAS, Voronoi Diagram and PSO. • The model allows of reversing the flow parameters from the estimates of field data. • A sensitivity analysis is conducted for different PSO solution parameters. • Model is evaluated on a real river reach and results are compared with literature. A linked simulation–optimization model is proposed for simultaneously estimating the Manning’s surface roughness parameters and their associated parameter structures for one-dimensional shallow water flows. In the simulation part of the model, hydrodynamic flow process is simulated by modeling the given flow reach on HEC-RAS. The association of unknown parameter structure with the roughness values is accomplished by partitioning the given flow reach into sub-regions using the one dimensional Voronoi Diagram (VD). The developed simulation model is then linked to an optimization model where heuristic Particle Swarm Optimization algorithm is used. The main objective of the PSO based optimization model is to determine the roughness parameters and their associated parameter structures along the flow reach by minimizing the discrepancies between simulated and measured water elevations at several observation locations. The applicability of the model is evaluated on a flow reach of East Fork River, WY, USA under three scenarios by considering the known/unknown roughness parameter structures and unsteady flow conditions. Furthermore, a sensitivity analysis is conducted to determine the influence of different PSO parameters on solution accuracy. Identified results indicated that the model provides better results than those obtained by different optimization methods in literature and significantly improves the calibration statistics by considering the variability of roughness parameters along the flow reach.

STRUCTURAL PARAMETER IDENTIFICATION USING DAMPED TRANSFER MATRIX AND STATE VECTORS

A new method for identification of structural parameters is proposed using Damped Transfer Matrices (DTM) and state vectors. A new transfer matrix is derived for continuous mass systems including the damping parameters. The state vector at a location is the sum of the internal and external contributions of displacements, forces, and moments at that point, when it is multiplied with the transfer matrix, state vector at the adjacent location is obtained. The structural identification algorithm proposed here involves prediction of displacement responses at selected locations of the structure using DTM and compares them with the measured responses at the respective locations. The mean square deviations between the measured and predicted responses at all locations are minimized using a nonclassical optimization algorithm, and the optimization variables are the unknown stiffness and damping parameters in the DTM. A nonclassical heuristic Particle Swarm Optimization algorithm (PSO) is used, since it is especially suited for global search. This DTM algorithm with successive identification strategy is applied on one element or sub-structure of a structure at a time and identifies all the parameters of adjacent elements successively. The algorithm is applied on numerically simulated experiments of two structures such as ten degrees of freedom lumped mass system and a cantilever with seven finite elements and one sub-structure of a nine member frame structure. Also this algorithm is verified experimentally on a sub-structure of a fixed beam. The main advantage of this algorithm is that it can be used for the local identification in a zone in a structure without modeling the entire global structure.

Identification of structural parameters using consistent mass transfer matrix

The stiffness parameters of structural members are identified by a novel inverse scheme based on consistent mass transfer matrix (TM), and the results are experimentally verified. The parameters are identified from the measurement of the state vector at a starting point, and displacements at few other locations. A new consistent mass-based TM is derived from the dynamic stiffness matrix of a beam element. The state vector at a location is the sum of the internal and external contributions of displacements, forces and moments at that point. When these are multiplied with the TM we obtain the state vector at the adjacent node. The identification algorithm proposed in this inverse problem involves predicting the displacements at certain locations using the TM, and comparing them with the measured displacements at those locations. The mean square deviations between the measured and predicted responses at all locations are minimized using an optimization algorithm, and the optimization variables are the unknown stiffness parameters in the TM. A non-classical heuristic particle swarm optimization algorithm (PSO) is used, since it is especially suited for global search. This TM method is particularly suited for local identification in large structures. Numerical simulations are carried out on two examples such as a cantilever, and a substructure of a nine-member frame structure. The method is also experimentally validated on a cantilever and substructure of a fixed beam. The speed and accuracy of the method are compared with existing structural identification methods, and advantages on the above are observed.