arXiv subjects

Bingqing Chen

Publications and source records attributed to Bingqing Chen.

23 records · Page 2Linked to original sources

arXiv2021

Enforcing Policy Feasibility Constraints through Differentiable Projection for Energy Optimization

While reinforcement learning (RL) is gaining popularity in energy systems control, its real-world applications are limited due to the fact that the actions from learned policies may not satisfy functional requirements or be feasible for the underlying physical system. In this work, we propose PROjected Feasibility (PROF), a method to enforce convex operational constraints within neural policies. Specifically, we incorporate a differentiable projection layer within a neural network-based policy to enforce that all learned actions are feasible. We then update the policy end-to-end by propagating gradients through this differentiable projection layer, making the policy cognizant of the operational constraints. We demonstrate our method on two applications: energy-efficient building operation and inverter control. In the building operation setting, we show that PROF maintains thermal comfort requirements while improving energy efficiency by 4% over state-of-the-art methods. In the inverter control setting, PROF perfectly satisfies voltage constraints on the IEEE 37-bus feeder system, as it learns to curtail as little renewable energy as possible within its safety set.

eess.SY↗

arXiv2020

Learning to Solve AC Optimal Power Flow by Differentiating through Holomorphic Embeddings

Alternating current optimal power flow (AC-OPF) is one of the fundamental problems in power systems operation. AC-OPF is traditionally cast as a constrained optimization problem that seeks optimal generation set points whilst fulfilling a set of non-linear equality constraints -- the power flow equations. With increasing penetration of renewable generation, grid operators need to solve larger problems at shorter intervals. This motivates the research interest in learning OPF solutions with neural networks, which have fast inference time and is potentially scalable to large networks. The main difficulty in solving the AC-OPF problem lies in dealing with this equality constraint that has spurious roots, i.e. there are assignments of voltages that fulfill the power flow equations that however are not physically realizable. This property renders any method relying on projected-gradients brittle because these non-physical roots can act as attractors. In this paper, we show efficient strategies that circumvent this problem by differentiating through the operations of a power flow solver that embeds the power flow equations into a holomorphic function. The resulting learning-based approach is validated experimentally on a 200-bus system and we show that, after training, the learned agent produces optimized power flow solutions reliably and fast. Specifically, we report a 12x increase in speed and a 40% increase in robustness compared to a traditional solver. To the best of our knowledge, this approach constitutes the first learning-based approach that successfully respects the full non-linear AC-OPF equations.

cs.LG↗

arXiv2020

COHORT: Coordination of Heterogeneous Thermostatically Controlled Loads for Demand Flexibility

Demand flexibility is increasingly important for power grids. Careful coordination of thermostatically controlled loads (TCLs) can modulate energy demand, decrease operating costs, and increase grid resiliency. We propose a novel distributed control framework for the Coordination Of HeterOgeneous Residential Thermostatically controlled loads (COHORT). COHORT is a practical, scalable, and versatile solution that coordinates a population of TCLs to jointly optimize a grid-level objective, while satisfying each TCL's end-use requirements and operational constraints. To achieve that, we decompose the grid-scale problem into subproblems and coordinate their solutions to find the global optimum using the alternating direction method of multipliers (ADMM). The TCLs' local problems are distributed to and computed in parallel at each TCL, making COHORT highly scalable and privacy-preserving. While each TCL poses combinatorial and non-convex constraints, we characterize these constraints as a convex set through relaxation, thereby making COHORT computationally viable over long planning horizons. After coordination, each TCL is responsible for its own control and tracks the agreed-upon power trajectory with its preferred strategy. In this work, we translate continuous power back to discrete on/off actuation, using pulse width modulation. COHORT is generalizable to a wide range of grid objectives, which we demonstrate through three distinct use cases: generation following, minimizing ramping, and peak load curtailment. In a notable experiment, we validated our approach through a hardware-in-the-loop simulation, including a real-world air conditioner (AC) controlled via a smart thermostat, and simulated instances of ACs modeled after real-world data traces. During the 15-day experimental period, COHORT reduced daily peak loads by an average of 12.5% and maintained comfortable temperatures.

eess.SY↗

arXiv2020

Learning a Distributed Control Scheme for Demand Flexibility in Thermostatically Controlled Loads

Demand flexibility is increasingly important for power grids, in light of growing penetration of renewable generation. Careful coordination of thermostatically controlled loads (TCLs) can potentially modulate energy demand, decrease operating costs, and increase grid resiliency. However, it is challenging to control a heterogeneous population of TCLs: the control problem has a large state action space; each TCL has unique and complex dynamics; and multiple system-level objectives need to be optimized simultaneously. To address these challenges, we propose a distributed control solution, which consists of a central load aggregator that optimizes system-level objectives and building-level controllers that track the load profiles planned by the aggregator. To optimize our agents' policies, we draw inspirations from both reinforcement learning (RL) and model predictive control. Specifically, the aggregator is updated with an evolutionary strategy, which was recently demonstrated to be a competitive and scalable alternative to more sophisticated RL algorithms and enables policy updates independent of the building-level controllers. We evaluate our proposed approach across four climate zones in four nine-building clusters, using the newly-introduced CityLearn simulation environment. Our approach achieved an average reduction of 16.8% in the environment cost compared to the benchmark rule-based controller.

eess.SY↗

arXiv2020

Damage-sensitive and domain-invariant feature extraction for vehicle-vibration-based bridge health monitoring

We introduce a physics-guided signal processing approach to extract a damage-sensitive and domain-invariant (DS & DI) feature from acceleration response data of a vehicle traveling over a bridge to assess bridge health. Motivated by indirect sensing methods' benefits, such as low-cost and low-maintenance, vehicle-vibration-based bridge health monitoring has been studied to efficiently monitor bridges in real-time. Yet applying this approach is challenging because 1) physics-based features extracted manually are generally not damage-sensitive, and 2) features from machine learning techniques are often not applicable to different bridges. Thus, we formulate a vehicle bridge interaction system model and find a physics-guided DS & DI feature, which can be extracted using the synchrosqueezed wavelet transform representing non-stationary signals as intrinsic-mode-type components. We validate the effectiveness of the proposed feature with simulated experiments. Compared to conventional time- and frequency-domain features, our feature provides the best damage quantification and localization results across different bridges in five of six experiments.

cs.CE↗