YZYuhang Zhu

Yuhang Zhu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yuhang Zhu has a total of 14 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Multi-timescale Learn-to-Optimize Method for Unit Commitment with Renewable Power

    by Zhibo Xu, Yuhang Zhu, Qing-Shan Jia, Xiaohong Guan

    The high penetration of volatile renewable power has created an urgent need for fast methods to solve large scale security-constrained unit commitment (SCUC) problems. In this paper, we propose a multi-timescale learn-to-optimize (MT L2O) method to efficiently solve large-scale stochastic SCUC. The coarse timescale serves as a learning-based presolving stage. At this scale, we propose a group-based multi-resolution formulation where machine learning (ML) techniques are used to predict integer variables and fix those with high confidence for finer timescales. Therefore, problems at finer timescales are reduced in size and can be solved more efficiently. Compared to existing works, when learning at the coarse scale, binary status constraints spanning multiple time periods become inactive. This not only significantly reduces data dimensionality but also eliminates the need for feasibility recovery measures targeting these constraints. At finer timescales, we develop constraint reduction methods for two types of constraints based on the on/off status determined at coarser timescales. Numerical experiments on IEEE 118-bus system, IEEE 300-bus system, and a 3266-bus system based on a practical provincial grid in China demonstrate that our methods obtain near-optimal solutions with an average performance gap not exceeding 0.4% and achieve speedups of 3.03 to 34.46 times compared to the commercial solver Gurobi.

  • A Bayesian Optimization Method for Design Space Exploration of Processors With Efficient Evaluation Budget Allocation

    by Yuhang Zhu, Mingyuan Hou, Xiaoliang Lv, Qing-Shan Jia, Xiaohong Guan

    Design space exploration (DSE), which aims to optimize the parameter configurations for the processor with a limited evaluation budget, is an important but challenging problem. Processor performance for a given parameter configuration can only be obtained through expensive evaluations. A key characteristic of processor DSE is that the optimization objective is the benchmark score, which is typically computed as a weighted combination of multiple sub-benchmark scores. Each sub-benchmark score can only be obtained through a costly evaluation. Bayesian optimization (BO) is widely used for DSE through an iterative process, where a surrogate model is constructed and updated to guide the selection of parameter configurations in each iteration. However, during the optimization process, it is not always necessary to evaluate all subbenchmarks for every recommended parameter configuration. In some cases, evaluating only a subset of sub-benchmarks can still provide significant improvements to the surrogate model. This paper proposes a BO-based algorithm that effectively utilizes the evaluation budget for efficient optimization. In each iteration, the evaluation budget is divided into two parts: exploitation and exploration. For the exploitation part, we propose a diversity-aware horse racing selection rule and use ordinal optimization to identify configurations with a high likelihood of being optimal. They are evaluated on all sub-benchmarks to obtain the total scores. For the exploration part, we propose a conditional entropy-based greedy algorithm to allocate the evaluation budget across parameter configurations and their corresponding sub-benchmarks, which can effectively improve the surrogate model. We conduct a comprehensive numerical experiment to validate the effectiveness of both components of our algorithm. Furthermore, experiments on three real-world DSE problems demonstrate that our algorithm outperforms state-of-the-art Bayesian optimization methods.

  • An Efficient Parallel Single Surrogate Objective Optimization Method for Multi-Objective Black-Box Problems and Its Application in Processor Design

    by Xiaoliang Lv, Qiaozhu Zhai, Yuhang Zhu, Jianchen Hu, Yuzhou Zhou, Xiaohong Guan

    With the growing complexity of modern micro-architectures, processor design must accommodate a wide array of parameters, resulting in vast design spaces. Identifying the optimal trade-offs among various design metrics poses a significant challenge. Performance is inherently difficult to model analytically, while area can be represented by analytical models. Performance evaluation relies on expensive and time-consuming cycle-accurate simulators (CAS), which puts forward strict requirements on the convergence speed and data dependence of the optimization methods. In this paper, based on the characteristics of the design metrics, the processor design problem is modeled as a hybrid black-box and white-box multi-objective discrete optimization problem (BWMO-DOP). In engineering applications, parallel simulation is a common acceleration technology. Therefore, white-box objective, area, is formulated as parallel constraints, while black-box objective, performance, is approximated by order-preserving surrogate objective. And then, BWMO-DOP is simplified into parallel single-objective expensive black-box optimization problems, which are solved by an efficient SOP-MOOA. SOP-MOOA iteratively explores more design points, enhancing the accuracy of the surrogate model while simultaneously updating the Pareto set. Experimental results demonstrate that the proposed algorithm outperforms baseline algorithm in an engineering case and three general numerical cases. In the engineering experiment for performance-area optimization, the proposed algorithm outperforms the baseline algorithm by a factor of more than two when considering the combined effect of performance improvement percentage and area reduction percentage. In numerical tests conducted on three 40-dimensional black-box functions under the same evaluation overhead, the proposed algorithm consistently identified Pareto set of superior quality.


  • IPLAM: A High-Dimensional Expensive Simulation Optimization Method, With Application to Design Space Exploration in Processor

    by Xiaoliang Lv, Qiaozhu Zhai, Yuhang Zhu, Jianchen Hu, Xiaohong Guan

    Design Space Exploration (DSE) in processors is an expensive discrete simulation optimization problem. The data requirements of the regular data-driven methods are so large that it is challenging to converge to a satisfactory solution within a limited simulation budget. Based on binary integer linear programming (BILP), an iteratively piecewise linear approximate method (IPLAM) is proposed for this kind of problems to reduce the dependence of simulation data. IPLAM starts from an initial reference point. Each iteration generates a set of trial points based on the most promising reference point by piecewise shift method. After evaluating the trial points, a local surrogate model is constructed for the unit neighborhood of the reference point. The surrogate model is then used to guide the exploration of the next reference point. In theory, IPLAM can converge to the global optimal point under mild assumptions, which is further verified by the numerical experiments. Meanwhile, the numerical experiments demonstrate that IPLAM outperforms the advanced Bayesian optimization and differential evolution methods on high-dimensional discrete black-box test functions. Besides, the practical effectiveness of IPLAM is validated by an industrial case for processor DSE.

  • A Reinforcement Learning Embedded Surrogate Lagrangian Relaxation Method for Fast Solving Unit Commitment Problems

    by Yuhang Zhu, Gaochen Cui, Anbang Liu, Qing-Shan Jia, Xiaohong Guan, Qiaozhu Zhai, Qi Guo, Xianping Guo

    Unit commitment problems are operation optimization problems solved by independent system operators (ISOs). These problems generally need to be solved within a limited time, and the quality of the solution can significantly impact the benefit of the power system. Due to the combinatorial complexities, quickly solving large-scale UC problems is particularly challenging and proposing an efficient solution methodology is crucial. In this paper, to accelerate solving speed, we embed reinforcement learning (RL) within the surrogate Lagrangian relaxation (SLR) framework. This approach leverages decomposition and machine learning to reduce the complexity of solving UC problems. By relaxing coupling constraints, the entire problem is decomposed into a set of sub-problems, each associated with a unit and significantly reduced in complexity. These sub-problems are then novelly formulated as Markov decision processes (MDPs), and a novel RL algorithm is used to rapidly generate high-quality feasible solutions. Our method substantially improves the overall speed of SLR and is applicable for solving large-scale UC problems. Numerical experiments on the IEEE 118-bus system and the 10K-bus system demonstrate that our method can obtain near-optimal solutions with no more than 3% performance degradation while achieving a speedup of 25$\sim$110 times compared to Gurobi, the state-of-the-practice solver.