A Data-Driven Framework for Assessing Cold Load Pick-Up Demand in Service Restoration

Bu, Fankun; Dehghanpour, Kaveh; Wang, Zhaoyu; Yuan, Yuxuan

doi:10.1109/tpwrs.2019.2922333

Cited by 22 publications

(5 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…α k L (t) and DP k L (t) are influenced by the time duration of power outage [22], external environment [23], and user types [24]. For example, the cold load pickup (CLPU) demand is an increased load in the restoration process due to the loss of load diversity, whose predict model can be established by data-driven [25] or model-driven [26] methods, and the load demand in the restoration process can be dynamically updated online. In addition, the restoration income per unit load and the load demand will be affected by load demand response (DR), which should also be modeled [27] and reflected to the time-varying load predict model.…”

Section: B Mathematical Model 1) Objective Functionmentioning

confidence: 99%

Decision Support System for Adaptive Restoration Control of Distribution System

Hao

Xue

et al. 2022

Journal of Modern Power Systems and Clean Energy

View full text Add to dashboard Cite

Aiming at the high-dimensional uncertainties of restoration process, an optimization model for distribution system restoration control is proposed considering expected restoration benefits, expected restoration costs, and security risks of the overall restoration scheme. In the proposed model, the effect of security control on restoration process is actively analyzed considering the security control costs of preventive, emergency, and correction controls. A two-layer decision support framework for distribution system restoration decision support system (DRDSS) is also designed. The upper layer of the proposed framework generates the pre-adjustment schemes of operation mode for energized power grid by load transfer and selects the optimal pre-adjustment scheme and the corresponding partitioning scheme based on the partition adjustment results of each pre-adjustment scheme. In addition, it optimizes the spatial-temporal decision-making of the inter-partition connectivity. For each partition, the lower layer of the proposed framework pre-selects the units and loads to be restored according to the pre-evaluated restoration income, generates the table of alternative restoration scheme for coping with uncertain events through simulation and deduction, and evaluates the risk and benefit of each scheme. For the uncertain events in the actual restoration process, the current restoration scheme can be adaptively switched to a sub-optimal scheme or re-optimized if necessary. Meanwhile, the proposed framework provides an information interaction interface for collaborative restoration with the related transmission system. A 123-node test system is built to evaluate the effectiveness and adaptability of the proposed model and framework.

show abstract

Section: B Mathematical Model 1) Objective Functionmentioning

confidence: 99%

Decision Support System for Adaptive Restoration Control of Distribution System

Hao

Xue

et al. 2022

Journal of Modern Power Systems and Clean Energy

View full text Add to dashboard Cite

show abstract

“…With the global temperature increasing year by year, a large amount of temperature control equipment is utilised, and the problem of cold-load pickup (CLPU) characteristics during the system restoration process is noteworthy [22]. The time-varying characteristic of the CLPU brings great challenges in the practical application of power system restoration schemes [23].…”

Section: Introductionmentioning

confidence: 99%

A bi‐level approach to load restoration strategy considering variant length of time steps

Sun

Yang

et al. 2021

IET Generation Trans & Dist

View full text Add to dashboard Cite

A bi‐level load restoration optimisation strategy is proposed for the transmission system with wind farm‐energy storage combined systems (WESs), taking the variant length of time steps into account. The upper level model is proposed to maximise weighted restored loads, and formulated as a mixed‐integer linear programming model. After solving the upper model, the optimal load pickup and transmission line restoration scheme can be obtained and delivered to the lower level model. The lower level model adopts a non‐linear model to minimise the length of the current time step, which is delivered to the upper lever model. By iteratively solving the upper and lower level models, the optimal load pickup and transmission line restoration scheme as well as the optimal length of current time step can be obtained. To minimise the gap between the scheduled generation of the WES and its actual power generation, a real‐time energy storage (ES) dispatch strategy is proposed taking maximum charging‐discharging cycles of the ES into account. The entire load restoration strategy can be obtained by iteratively solving the proposed model with updated operational conditions of power systems. Finally, two test systems are employed to verify the validity and correctness of the proposed model.

show abstract

“…In fact, in order to predict the CLPU demand in service restoration, there is a need for a technique that is able to characterise the CLPU behaviour of all loads together in a feeder level without the need to model each of them individually. In this regard in [23] a data‐driven framework is proposed for assessing CLPU demand of residential customers using smart meter data. The recent large‐scale deployment of smart meters has provided the industry with a huge amount of data that is highly granular, both temporally and spatially [23].…”

Section: Introductionmentioning

confidence: 99%

“…In this regard in [23] a data‐driven framework is proposed for assessing CLPU demand of residential customers using smart meter data. The recent large‐scale deployment of smart meters has provided the industry with a huge amount of data that is highly granular, both temporally and spatially [23]. Accordingly, in [23] by using smart meter data and non‐linear auto‐regression model, CLPU demand is predicted without the need for developing specific load models.…”

Section: Introductionmentioning

confidence: 99%

“…The recent large‐scale deployment of smart meters has provided the industry with a huge amount of data that is highly granular, both temporally and spatially [23]. Accordingly, in [23] by using smart meter data and non‐linear auto‐regression model, CLPU demand is predicted without the need for developing specific load models. Also, in [13] a realistic parameter set for the CLPU curve (current or power) model is provided.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation