Fast and accurate suppression of ferroresonance fault can not only effectively protect electromagnetic voltage transformers, but also avoid potential risks of power system security and stability. Therefore, an improve...
详细信息
Aiming to solve the issues of large time cost, low efficiency, and high computational complexity in the design of cavity filters using traditional approaches, a multi-objective design approach for cavity filters based...
Aiming to solve the issues of large time cost, low efficiency, and high computational complexity in the design of cavity filters using traditional approaches, a multi-objective design approach for cavity filters based on the Slime Mould Algorithm (SMA) optimized Sparse Regularized BP neural network (SR-BP) surrogate model is proposed. This method first utilizes BP neural network to construct a surrogate model for cavity filter. Subsequently, a regularization term is added to its cost function, enabling further sparsity in the standard fully connected neural network. This achieves the reduction of model computational complexity and mitigates the risk of overfitting. To tackle the challenges of global optimization and control of sparse regularization intensity, the SMA is draw into the SR-BP neural network to optimize initial connection parameters and the sparse regularization factor. Finally, the optimized SMA-SR-BP surrogate model for the cavity filter is integrated with the multi-objective bionic intelligent optimization algorithm to solve the optimization design problem of multi-objective cavity filters.
This paper proposes an action space preprocessing method adapted to the distribution network self-healing decision-making problem for the action space of the distribution network power load adaptive decision-making pr...
This paper proposes an action space preprocessing method adapted to the distribution network self-healing decision-making problem for the action space of the distribution network power load adaptive decision-making problem. On this basis, the traditional hybrid action space solving methods are analyzed, their limitations are pointed out, and reinforcement learning hybrid action space algorithms based on the state action value function, based on the state value function and based on the representation learning are proposed respectively. Based on the hybrid action characterization method, a high-quality implicit action space is constructed by mapping the discrete and continuous variables of the self-healing decision-making problem of the distribution network to the hidden variable space through embedded tables and conditional variational self-encoders. Finally, the action representations are decoded back to the original action space by the decoder of the action representations, and a hybrid encoding-decoding power load adaptive optimization algorithm for distributed distribution networks with discrete and continuous variables is established.
In this paper, an optimal state feedback controller is designed for a nonaffine nonlinear system. The nonaffine system is converted into affine system using mean value theorem, since the analysis of affine nonlinear s...
详细信息
ISBN:
(纸本)9781665425360
In this paper, an optimal state feedback controller is designed for a nonaffine nonlinear system. The nonaffine system is converted into affine system using mean value theorem, since the analysis of affine nonlinear system is easy compared to nonaffine systems. Feedback linearization is used for linearizing the affine system. Then a gradient descent algorithm-based optimization approach is designed for feedback linearized system for optimal performance. The performance of the proposed method is presented by conducting simulation study on magnetic levitation system. A significant reduction in the minimum value of cost function is obtained using proposed method compared with the conventional LQR technique and PSO method.
Based on the computer automatic control technology, this paper designs the energy efficiency optimization method for the traditional water-cooled central air conditioning host system. The chilled water control based o...
详细信息
Based on the computer automatic control technology, this paper designs the energy efficiency optimization method for the traditional water-cooled central air conditioning host system. The chilled water control based on load prediction fuzzy control and the cooling water control based on adaptive fuzzy algorithm are used in the system, and the cooling water supply of the chiller is matched with the end load demand, and the cooling water system is running at maximum efficiency. Then on the basis of the single point fault tolerance design and the key points of energy-saving automatic control design, the design effect of HVAC system in data room is analyzed. Compared with PID control and single neural network reinforcement learning control, the total energy consumption of the system is reduced by 5.36% and 1.64%, and the total proportion of non-comfort time is reduced by 2.32% and 1.37%, respectively.
With the development of technology, there are more and more robots around us. These robots need to recognize human pose to monitor people or interact with people, so the human pose estimation model is required to be d...
With the development of technology, there are more and more robots around us. These robots need to recognize human pose to monitor people or interact with people, so the human pose estimation model is required to be deployed on robots to achieve these functions. Human pose estimation involves the processing of more than a dozen joints, so the human pose estimation model usually has a relatively large amount of calculation resulting in a relatively slow detection speed. Transferring these large amounts of image data to cloud servers for human pose estimation detection will consume a lot of network resources and detection time will be delayed. And the robot’s local computing power is limited. So it is difficult to deploy both human pose estimation algorithm models and other complex algorithm models locally on robots. This requires various algorithms running locally on the robot to consume as little computing resources and real-time as possible. This paper studies the optimization methods of real-time 2D human pose estimation models studies in recent years, which are divided into three aspects: the optimization of the feature extraction network, the optimization of the joint point detection and the optimization of the joint point connection. And summarized the advantages and disadvantages of these optimization methods, attempting to apply them to robots.
The proceedings contain 121 papers. The topics discussed include: mathematical modelling and analysis of linear induction motor;switched capacitor mli design ideology with less number of switches using hysteresis band...
ISBN:
(纸本)9781665491754
The proceedings contain 121 papers. The topics discussed include: mathematical modelling and analysis of linear induction motor;switched capacitor mli design ideology with less number of switches using hysteresis bandwidth control;a novel strategy for ramp rate control of wind farms;structural analysis of air transport network using network indicators;miniaturized micro-strip band-pass filters;advanced warning mechanism to reduce man animal conflict on roads using renewable energy;secure and light weight data sharing scheme in cyber physical cloud environment;internal ventilation and thermal characteristics of a new ventilation system for a large offshore permanent magnet wind turbine generator;performing pulse-width modulation using two different methods in a modular topology;economic analysis of grid level battery energy storage system using repurposed EV batteries;and de-orbit low thrust optimization to compute rapid orbit lowering trajectories.
This paper studies worst-case robust optimal tracking using noisy input-output data. We utilize behavioral system theory to represent system trajectories, while avoiding explicit system identification. We assume that ...
详细信息
This paper studies worst-case robust optimal tracking using noisy input-output data. We utilize behavioral system theory to represent system trajectories, while avoiding explicit system identification. We assume that the recent output data used to implicitly specify the initial condition are noisy and we provide a non-conservative design procedure for robust control based on optimization with a linear cost and linear matrix inequality (LMI) constraints. Our methods rely on the parameterization of noise sequences compatible with the data-dependent system representation and on a suitable reformulation of the performance specification, which further enable the application of the S-lemma to derive an LMI optimization problem. The performance of the new controller is discussed through simulations.
This paper presents the structure of a Permanent Magnet Synchronous Motor (PMSM) multi-motor sensorless automation control system based on the Field Oriented control (FOC) type method. It presents the description of t...
详细信息
ISBN:
(纸本)9781665435123
This paper presents the structure of a Permanent Magnet Synchronous Motor (PMSM) multi-motor sensorless automation control system based on the Field Oriented control (FOC) type method. It presents the description of the component blocks of each Master/Slave subsystem, the optimal tuning of the Master and Slave controllers by using a Particle Swarm optimization (PSO) algorithm and estimations of the PMSM rotor position and velocity by using a Model Reference Adaptive System (MRAS) observer. In Simulink development software are achieved the numerical simulations to prove the performance of optimized controllers in case of uneven distribution of the load torque between Master and Slave. It also presents the implementation of a Supervisory control and Data Acquisition (SCADA) system which is characterized both by the implementation of PMSM Master/Slave control algorithms in dedicated controllers and the communication between Simulink and LabVIEW development software based on the Open Platform Communications (OPC) technology.
This paper presents an improved ant colony optimization for vehicular ad-hoc network routing. The algorithm can quickly find the route with optimal network connectivity. Assuming that each vehicle has a digital map co...
详细信息
暂无评论