Hierarchical Optimization Research Articles

Performance-Driven Safe Bayesian Optimization for Intelligent Tuning of High-Order Cascade Systems

Automatic parameter tuning of high-order cascade controllers suffers from sampling inefficiency and strong couplings. This work presents a performance-driven, systematic, and safe intelligent parameter-tuning framework for high-order cascade systems. To achieve data-efficient and noise-robust hyper-parameter calibration, an intelligent tuning framework based on Bayesian optimization is proposed to calibrate the control parameters from the buffer of performance-metric measurements. Furthermore, we improve the Bayesian-optimization-based framework in three aspects, involving effectiveness, security, and sampling efficiency. Firstly, a comprehensive control performance assessment combining the error-integral and statistical performance criteria is designed to evaluate the cost of a sampling point in terms of final precision, response rapidity, and vibration. Meanwhile, the security of sampling exploration is heightened by imposing composite hard parametric and soft performance-metric (maximum input, overshoot, etc.) constraints on the acquisition points. Additionally, a hierarchical optimization strategy is proposed to further boost sampling exploitation by alternatively tuning the control parameters of subsystems and refining the constraint on the sampling space. The proposed framework is applied to automatic parameter tuning of high-order dynamic-surface-based backstepping control combined with anti-disturbance rejection control. The comparative simulation results demonstrate that the proposed intelligent tuning possesses superiorities in sampling efficiency and security for parameter calibration problems.

Evaluation and optimization of ecological spatial resilience of Yanhe River Basin based on complex network theory

The loess hilly and gully areas have broken terrain, vertical and horizontal ravines and fragile ecological environments. Improving the resilience of the regional ecological space is conducive to improving the quality of the local ecological environment. With the ecological space of the Yanhe River Basin selected as the research object, this paper constructs a research framework of "network identification topology-resilience evaluation-spatial optimization" and uses morphological spatial pattern analysis (MSPA) and the minimum cumulative resistance model (MCR) to identify ecological spatial networks. Based on circuit theory, the ecological pinch point is identified, the ecological spatial network is optimized, and scenario simulation is performed. Through complex network theory and related indicators, the ecological spatial resilience of the basin is evaluated, and the hierarchical optimization strategy of the ecological space is confirmed. According to the ecological function of the source area and the results of the resilience evaluation, the boundaries of the protected control area, guidance development area, remediation area, and maintenance and improvement area of the basin are delineated. The importance of ecological source and corridor protection is classified, and corresponding protection strategies are proposed. The research results can provide theoretical support and practical guidance for the territorial spatial planning and ecological space construction of the Yanhe River Basin and provide a reference for the ecological restoration, resource development and environmental governance of the Yanhe River Basin.

Hierarchical multi-objective optimization of proton exchange membrane fuel cell with parameter uncertainty

It is important to improve the performance indexes including the net power and system efficiency for polymer electrolyte membrane fuel cell (PEMFC), especially considering the parameter uncertainty. To this end, this paper proposes a hierarchical multi-objective optimization framework including the off-line and on-line optimization. Firstly, a steadystate non-linear PEMFC system model is established as the base model. Secondly, an adaptive multi-objective particle swarm optimization (AMPSO) algorithm scheme is proposed in the off-line optimization procedure for the exact optimization of operation parameters. Meanwhile, an adaptive flight parameter strategy based on the particle dispersity (PD) information is proposed to balance the convergence and diversity of Pareto solutions. Finally, to decrease the influence caused by parameter uncertainty, the interval optimization method is proposed in the on-line optimization layer based on the results of AMPSO. The convergence condition of the proposed optimization scheme is verified by theory analysis. The proposed AMPSO is compared with different algorithms in the numerical simulation and hardware-in-loop (HIL) experiments. Meanwhile, the performance of the proposed method is tested on four representative benchmark problems. These results demonstrate that the PEMFC system with the proposed optimization scheme performs better than the base model and classical optimization algorithms in terms of the net power and system efficiency indexes, revealing the success of this hierarchical optimization approach in solving the accurate optimization and parameter uncertainty problems. By using statistical test methods, the proposed algorithm performs better hypervolume (HV) metric from the Pareto solution distribution.

WHRIME: A weight-based recursive hierarchical RIME optimizer for breast cancer histopathology image segmentation

In medical image processing, multi-threshold image segmentation has been challenging, as selecting appropriate thresholds is crucial for distinguishing different structures within an image, especially when dealing with breast cancer images. Breast cancer images are complex with multiple tissue types, which pose challenges to precise diagnosis. A weight-based recursive hierarchical bootstrapping rime algorithm (WHRIME) is proposed to tackle this case effectively. The proposed WHRIME segregates the population into elite and non-elite individuals. The hierarchical bootstrapping strategy ensures comprehensive exploration of the solution space, with elite individuals guiding the position updates of non-elite individuals. A weighted method is introduced based on solution quality differences to maintain population diversity and enhance convergence accuracy. We apply WHRIME in multi-threshold image segmentation on breast cancer histopathology images. Experimental results on both the IEEE CEC 2017 benchmark suit and breast cancer histopathology images from the Databiox dataset validate the superiority of WHRIME over competing algorithms, affirming WHRIME’s capability in addressing complex breast cancer histopathology image segmentation problems.

Hierarchical Optimization for Personalized Hand and Wrist Musculoskeletal Modeling and Motion Estimation.

Surface electromyography (sEMG) driven musculoskeletal models are promising to be applied in the field of human-computer interaction. However, due to the individual-specific physiological characteristics, generic models often fail to provide accurate motion estimation. This study optimized the general model to build a personalized model and improve the accuracy of motion estimation. Inspired by the coupling effect of wrist/hand movement, a hierarchical optimization approach for personalizing musculoskeletal models (HOPE-MM) is proposed, which aligns with the physiological characteristics of the human wrist and hand. To verify the effectiveness of personalized musculoskeletal model, single joint motions and simultaneous joint motions are estimated. In addition, Sobol sensitivity analysis is conducted to identify the key parameters of musculoskeletal model, providing guidance for model simplification. The mean pearson correlation coefficient between the predicted joint angles and the measured joint angles are 0.95 ± 0.03 and 0.93 ± 0.01 for simultaneous wrist and metacarpophalangeal (MCP) joint movements, respectively, which have a significant improvement compared with the stateof- the-art works. By optimizing only the key parameters including tendon slack length, maximal isometric force and optimal fiber length, the performances of simplified model are comparable to the full-parameter model. These results provide insights into the effects of muscletendon parameters on musculoskeletal model, and musculoskeletal models personalized using hierarchical optimization methods can improve the accuracy of motion estimates. These findings facilitate the clinical application of musculoskeletal models in rehabilitation and robotic control.

Hierarchical Multi-Objective Optimization of a Multi-Sided Permanent Magnet Linear Synchronous Motor With Ring Structure Winding for Conveyor Systems

Hierarchical Multi-Objective Optimization of a Multi-Sided Permanent Magnet Linear Synchronous Motor With Ring Structure Winding for Conveyor Systems

Hybrid Hierarchical Particle Swarm Optimization with Evolutionary Artificial Bee Colony Algorithm for Task Scheduling in Cloud Computing

Hybrid Hierarchical Particle Swarm Optimization with Evolutionary Artificial Bee Colony Algorithm for Task Scheduling in Cloud Computing

Optimum dynamic dispatch of clean water, methane, electricity and heat considering carbon capture

This paper suggests a novel clean water, natural gas, heat and power generation system using wind turbine generators, PVT panels, bioenergy power plants, micro-hydro power plants, micro-cogeneration units (MCUs), battery energy storage system, thermal energy storage system and plug-in electric vehicles (PEVs). In this system, a power to gas technology consisting of a carbon capture unit, an electrolyzer, a hydrogen storage unit, and a methanation process, is used to supply the natural gas demand using CO2 captured from the MCUs and the hydrogen generated by the clean water electrolyzer. Furthermore, the electricity is consumed by a desalination unit to procure the clean water demand, carbon captured unit (CCU), electrolyzer water inflow, and fill up the water storage tank. At peak clean water demand periods, the excess water is pumped from the storage tank to the consumption side. Demand response program (DRP) is used for levelling the demand. Here, total cost for water-gas-heat-power nexus model throughout the day with and without DRP for three different scenarios is minimized by using quasi-oppositional fast convergence evolutionary programming, fast convergence evolutionary programming, self-organizing hierarchical particle swarm optimizer with time-varying acceleration coefficients and differential evolution. It is seen from numerical results that cost attained with DRP is less than the cost attained without DRP.

Co-optimization of Energy Management and Eco-Driving Considering Fuel Cell Degradation Via Improved Hierarchical Model Predictive Control

An advanced eco-driving technology is widely recognized as having enormous potential to reduce the vehicle fuel consumption. However, most research on eco-driving focuses on the stability and safety for vehicle operating while disregarding its comfort and economy. To meet the requirements for safety and comfort, at the same time, enhance the economic performance of the vehicles, an improved hierarchical model predictive control cooperative optimization strategy is proposed for fuel cell hybrid electric vehicle with car-following scenario. Specifically, the upper-level model predictive controller controls the velocity, inter-vehicle distance and acceleration to guarantee safety and comfort for driving. According to the velocity information obtained from the upper model predictive controller, the lower-level improved model predictive controller considers the impact of disturbance changes on vehicle economy and aims to minimize the vehicle operating cost considering fuel cell degradation, so as to allocate energy rationally. Finally, the enhancement of economic performance of proposed strategy is verified with the results of comparative study that 3.09% economic improvement on the premise of assuring safety and comfort of driving.

Multi-UAV Hierarchical Intelligent Traffic Offloading Network Optimization Based on Deep Federated Learning

Multi-UAV Hierarchical Intelligent Traffic Offloading Network Optimization Based on Deep Federated Learning

Hierarchical Collaborative Optimization of Shared Energy Storage With Co-Generation Based on Deep Reinforcement Learning and P2P Network Game Theory

Hierarchical Collaborative Optimization of Shared Energy Storage With Co-Generation Based on Deep Reinforcement Learning and P2P Network Game Theory

Multi-objective Hierarchical Optimization Method with Decoupling Capability for Hybrid-Magnetic-Circuit Variable Flux Memory Machine

Multi-objective Hierarchical Optimization Method with Decoupling Capability for Hybrid-Magnetic-Circuit Variable Flux Memory Machine

A hierarchical surrogate assisted optimization algorithm using teaching-learning-based optimization and differential evolution for high-dimensional expensive problems

Surrogate-assisted evolutionary algorithms (SAEAs) are increasingly used in solving computationally expensive optimization problems. However, when tackling high-dimensional expensive problems, a large number of exact function evaluations (FEs) need to be consumed for existing SAEAs to achieve an acceptable solution. In this paper, a hierarchical surrogate assisted optimization algorithm (HSAOA) using teaching-learning-based optimization and differential evolution is proposed for solving high-dimensional expensive problems with a relatively small number of exact FEs. To keep a balance between global exploration and local exploitation, a hierarchical surrogate framework with hybrid evolutionary algorithms is devised. In the global search phase, a radial basis function surrogate is utilized to assist the teaching-learning-based optimization in locating the promising sub-regions. In the local search phase, a novel dynamic ensemble of surrogates is proposed to assist the differential evolution in speeding up the convergence process. Eight test functions with 10 to 100 dimensions and a spatial truss design problem are employed to compare the proposed method with several state-of-the-art SAEAs. The results show that the proposed HSAOA is superior to the comparison algorithms for solving expensive optimization problems, and needs a much smaller number of exact FEs than other competing SAEAs to produce competitive or even better results for high-dimensional expensive problems.

MODEL-BASED OPTIMIZATION OF ICVs’ CONTROL FOR PRODUCERS, INJECTORS, AND WAG-CO2 CYCLES SIZE IN A CARBONATE RESERVOIR UNDER UNCERTAINTY

Life cycle management of oil fields involves complex decisions, with many variables to be considered. In some fields of the Brazilian pre-salt, this task is even more challenging given the ultra-deep water location, with an expressive amount of gas (including CO2) production. This feature becomes a restricting factor for oil production in these types of fields, and a proper management is essential to overcome this issue, improving both production and economic returns. Therefore, Water-Alternate-Gas (WAG) is often used as an Enhanced Oil Recovery- EOR mechanism, providing a better mobility control, improving reservoir sweep and oil recovery. Moreover, WAG can be an alternative for recycling the produced gas, providing an adequate destination (mainly for CO2) and avoiding environmental problems. In this case, we name the method as WAG-CO2. Moreover, Interval Control Valves (ICVs) are also a good alternative to deal with these fields, since they work as flow control devices and allow for a better balance of the fluid flow in the reservoir. In this work, we evaluate the impact of the use of ICVs in producers, injectors, along with WAG cycles size to improve the performance of a developed field. The work consists of two parts. In Part I, using a single simulation model, we performed (1) six hierarchical optimization approaches to verify if the order in which the optimization occurs influences the final results, and (2) a joint optimization of all parameters for comparison purposes. The final strategies are simulated in a reference case (representing the “true response” of the field) to evaluate how the achieved control rules would behave in a real field. In Part II, we considered a probabilistic procedure, applying one of the hierarchical approaches in several representative models (RMs) expressing uncertain scenarios. In this case, we compared nominal optimization of individual RMs against a robust optimization (in which all RMs are optimized simultaneously). The study is applied to the benchmark case SEC1_2022, a synthetic model analogous to a pre-salt field. The results show an expressive increase in the economic return of the field with a proper operation of the ICVs and size of the WAG-CO2 cycles, despite the platform being restricted by gas production capacity. This result occurred because of a better balance in fluid flow through the reservoir and among wells. Moreover, in the first part of the work, all hierarchical approaches presented very similar results in terms of NPV, so the order of the optimization steps did not significantly affect the results. The non-hierarchical approach presented a slightly better result, but still close to those obtained in the hierarchical approaches. Furthermore, the most impacting variable was the ICV for producers, followed by ICV for injectors and, then, WAG-CO2 cycles. Similar tendencies were observed in the second part of the work, with higher expected economic return for the robust optimization.

Optimizing near-adiabatic compressed air energy storage (NA-CAES) systems: Sizing and design considerations

This paper studies the challenges of designing and operating adiabatic compressed air energy storage (A-CAES) systems, identifies core causes for the reported discrepancies between round-trip efficiencies from current literature models versus experiments, and presents a near-adiabatic CAES (NA-CAES) system design that addresses these issues. The core causes are overly simplified heat management modeling, neglect of heat losses in the thermal energy reservoir, single-cycle modeling, replacing a diabatic CAES combustor with a TES without other major design changes, using a mirrored compression-expansion design, and/or efficiency definition limitations.To demonstrate an approach to designing an NA-CAES system that addresses the core causes of modeling vs. experiment efficiency discrepancies, a case study is presented for Ontario, Canada. This full system view approach yielded NA-CAES systems with maximum round-trip efficiencies of 60.5, 61.1, and 61.5% for a constant reservoir volume throttled system, a constant reservoir volume variable pressure system, and a constant pressure reservoir system, respectively.A sensitivity analysis was conducted to identify primary with a strong influence on efficiency (such as reservoir pressure limits and the expanders’ pressure ratio), and secondary parameters that more strongly affect TES size (including TES charging and discharging mass flow rates, inlet temperatures to the expanders, and TES tank temperatures). Multi-objective hierarchical optimization was employed to improve efficiency while reducing TES capacity. In brief, the results reveal that through design configuration selection it should still be possible to achieve in practice high NA-CAES efficiencies in excess of 60%.

Cross-subject emotion recognition using hierarchical feature optimization and support vector machine with multi-kernel collaboration

Objective. Due to individual differences, it is greatly challenging to realize the multiple types of emotion identification across subjects. Approach. In this research, a hierarchical feature optimization method is proposed in order to represent emotional states effectively based on peripheral physiological signals. Firstly, sparse learning combined with binary search is employed to achieve feature selection of single signals. Then an improved fast correlation-based filter is proposed to implement fusion optimization of multi-channel signal features. Aiming at overcoming the limitations of the support vector machine (SVM), which uses a single kernel function to make decisions, the multi-kernel function collaboration strategy is proposed to improve the classification performance of SVM. Main results. The effectiveness of the proposed method is verified on the DEAP dataset. Experimental results show that the proposed method presents a competitive performance for four cross-subject types of emotion identification with an accuracy of 84% (group 1) and 85.07% (group 2). Significance. The proposed model with hierarchical feature optimization and SVM with multi-kernel function collaboration demonstrates superior emotion recognition accuracy compared to state-of-the-art techniques. In addition, the analysis based on DEAP dataset composition characteristics presents a novel perspective to explore the emotion recognition issue more objectively and comprehensively.

Reinforcement Learning-Based Particle Swarm Optimization for End-to-End Traffic Scheduling in TSN-5G Networks

With the rapid development of the Industrial Internet of Things (IIoT), massive IIoT devices connect to industrial networks via wired and wireless. Furthermore, industrial networks pose new requirements on communications, such as strict latency boundaries, ultra-reliable transmission, and so on. To this end, time-sensitive networking (TSN) embedded fifth-generation (5G) wireless communication technology (i.e., TSN-5G networks), is considered the most promising solution to address these challenges. TSN can provide deterministic end-to-end latency and reliability for real-time applications in wired networks. 5G supports ultra-reliable and low-latency communications (uRLLC), providing increased flexibility and inherent mobility support in the wireless network. Thus, the integration of TSN and 5G provides numerous benefits, including increased flexibility, lower commissioning costs, and seamless interoperability of various devices, regardless of whether they use a wired or wireless interface. Nonetheless, the potential barriers between the TSN and 5G systems, such as clock synchronization and end-to-end traffic scheduling, are inevitable. Time synchronization has been studied in many works, so this paper focuses on the end-to-end traffic scheduling problem in TSN-5G networks. We propose a novel integrated TSN and 5G industrial network architecture, where the 5G system acts as a logical TSN-capable bridge. Based on this network architecture, we design a Double Q-learning based hierarchical particle swarm optimization algorithm (DQHPSO) to search for the optimal scheduling solution. The DQHPSO algorithm adopts a level-based population structure and introduces Double Q-learning to adjust the number of levels in the population, which evades the local optimum to further improve the search efficiency. Extensive simulations demonstrate that the DQHPSO algorithm can increase the scheduling success ratio of time-triggered flows compared to other algorithms.

EV Hosting Capacity Enhancement in a Community Microgrid Through Dynamic Price Optimization-Based Demand Response.

Community microgrids, as an emerging technology, offer resiliency in operation for smart grids. Microgrids are seeing an increased penetration of eco-friendly electric vehicles (EVs) in recent years. However, the uncontrolled charging of EVs can easily overwhelm such electric networks. In this work, we propose an efficient demand response (DR) scheme based on dynamic pricing to enhance the capacity of the microgrid to securely host a large number of EVs. A hierarchical two-level optimization framework is introduced to realize the DR scheme. At the upper level, the dynamic prices for the participating users in DR are optimized while at the lower level, each user optimizes its energy consumption based on the price signal from the upper level. An evolutionary algorithm and a mixed-integer linear programming model is employed to solve the upper and lower level problems, respectively. Energy scheduling problems of the users are solved in a distributed manner which adds to the scalability of the approach. The proposed DR scheme is tested on a microgrid system adopted from the IEEE European low-voltage distribution network. Numerical experiments confirm the effectiveness of the proposed DR scheme compared to the benchmark pricing policies from the literature.

Dynamic Adaptive Hierarchical N-Level Hierarchy Optimization for Electric Power System Problems

Many complex problems in the real world that contain a large number of constraints, especially circuit system problems. We design the Dynamic Adaptive Hierarchical N-Level Hierarchy Optimization (DAH-NLHO) algorithm based on the hierarchy affine of populations in nature. Its hierarchy levels can change dynamically and adaptively following the dimensionality of the problem and the number of constraints. Comparative tests were performed on top of the latest CEC2020 benchmark test set. The final experimental results demonstrate the effectiveness and feasibility of dynamic adaptive hierarchy levels.

Hierarchical Optimization of Speed Planning and Energy Management for Connected Hybrid Electric Vehicles Under Multi-Lane and Signal Lights Aware Scenarios

Hierarchical Optimization of Speed Planning and Energy Management for Connected Hybrid Electric Vehicles Under Multi-Lane and Signal Lights Aware Scenarios