The Electricity Rebound Effect: Empirical Evidence From the Chinese Chemical Industry

The Chinese government has adopted many policies to save energy and electricity in the chemical industry by improving technology and reforming its electricity market. The improved electricity efficiency and the electricity reform may indirectly reduce expected energy and electricity savings by decreasing the effective electricity price and the marginal cost of electricity services. To analyze the above issues, this paper employs the Morishima Elasticity of Substitution of the electricity cost share equation which is estimated by the DOLS method. The results show that: 1) There exists a rebound effect in the Chinese chemical industry, but it is quite large because the electricity price is being controlled by the government; 2) the reform of the electricity market reduces the rebound effect to 73.85%, as electricity price begins to reflect cost information to some extent; 3) there is still a lot of space for the reform to improve, and the rebound effect could be reduced further once the electricity price is adjusted to transfer the market information more correctly. In order to succeed in saving electricity and decreasing the rebound effect in the chemical industry, the policy implications are provided from perspectives of the improved energy efficiency and electricity pricing mechanism.

Cost Efficient GPU Cluster Management for Training and Inference of Deep Learning

Expanding the scale of GPU-based deep learning (DL) clusters would bring not only accelerated AI services but also significant energy consumption costs. In this paper, we propose a cost efficient deep learning job allocation (CE-DLA) approach minimizing the energy consumption cost for the DL cluster operation while guaranteeing the performance requirements of user requests. To do this, we first categorize the DL jobs into two classes: training jobs and inference jobs. Through the architecture-agnostic modeling, our CE-DLA approach is able to conduct the delicate mapping of heterogeneous DL jobs to GPU computing nodes. Second, we design the electricity price-aware DL job allocation so as to minimize the energy consumption cost of the cluster. We show that our approach efficiently avoids the peak-rate time slots of the GPU computing nodes by using the sophisticated mixed-integer nonlinear problem (MINLP) formulation. We additionally integrate the dynamic right-sizing (DRS) method with our CE-DLA approach, so as to minimize the energy consumption of idle nodes having no running job. In order to investigate the realistic behavior of our approach, we measure the actual output from the NVIDIA-based GPU devices with well-known deep neural network (DNN) models. Given the real trace data of the electricity price, we show that the CE-DLA approach outperforms the competitors in views of both the energy consumption cost and the performance for DL job processing.

Event-Based Evaluation of Electricity Price Ensemble Forecasts

The present paper considers the problem of choosing among a collection of competing electricity price forecasting models to address a stochastic decision-making problem. We propose an event-based evaluation framework applicable to any optimization problem, where uncertainty is captured through ensembles. The task of forecast evaluation is simplified from assessing a multivariate distribution over prices to assessing a univariate distribution over a binary outcome directly linked to the underlying decision-making problem. The applicability of our framework is demonstrated for two exemplary profit-maximization problems of a risk-neutral energy trader, (i) the optimal operation of a pumped-hydro storage plant and (ii) the optimal trading of subsidized renewable energy in Germany. We compare and contrast the approach with the full probabilistic and profit–loss-based evaluation frameworks. It is concluded that the event-based evaluation framework more reliably identifies economically equivalent forecasting models, and in addition, the results suggest that an event-based evaluation specifically tailored to the rare event is crucial for decision-making problems linked to rare events.