Jian Lan's research while affiliated with Tsinghua University and other places

Publications (5)

Article
Nontechnical losses (NTLs) are estimated to be considerable and increasing every year. Recently, high-resolution measurements from globally laid smart meters have brought deeper insights on users' consumption patterns that can be exploited potentially by NTL detection. However, consumption-pattern-based NTL detection is now facing two major challen...
Article
With the technological advancement in the fields of advanced metering infrastructure (AMI), a massive amount of customers’ electricity consumption data is collected. Meanwhile, the energy providers need to make informed decisions based on power consumption strategy of demand side to reduce overall operational cost. So how to generate demand side lo...

Citations

... According to [26] and [27], the manual creation of features is not sufficient to properly detect the NTL behavior because of stochastic changes in EC profiles. In [28], the problem of maintaining temporal correlation in the existing ML models is highlighted. Moreover, the learning algorithms are unable to learn the potential features from 1D raw EC data. ...
... Furthermore, research in the direction of EMS with RL showed similar problems [203]. One promising AI solution that addresses these general challenges is GAN (see Section 2.5), which has been proposed to increase the forecast accuracy [204] but more importantly has been implemented both against cyber-attacks [205] as well as for privacy protection [206]. ...
... The system coordinator solves the optimization problem to obtain a pricing strategy (i.e., prices for each area) to maximize the total utility, which includes passenger service profits, charging station incomes, and penalties for unmet demands of energy and traffic. Fig. 3 shows how the operational framework performs on the time axis, which is an extension from our previous work [27]. Each day, users' decision characteristics are recalculated if the newly sampled data deviate too much from the previous result. ...
... For performing high-quality speech synthesis, Prenger et al. [19] combine the melspectograms presented in WaveNet [17] with an Invertible Neural Network. Furthermore, to generate electrical consumption time series for scheduling and energy management, Lan et al. [16] condition their Wasserstein GAN on temporal and consumer information. ...