Fast and accurate load parameter identification has a large impact on power systems operation and stability analysis. This article proposes a novel Imitation and Transfer Q-learning (ITQ)-based method to identify parameters of composite constant impedance-current-power (ZIP) and induction motor (IM) load models. Firstly, an imitation learning process is introduced to improve the exploitation and exploration processes. Then, a transfer learning method is employed to overcome the challenge of time-consuming optimization when dealing with new identification tasks. An associative memory is designed to realize dimension reduction, knowledge learning and transfer between different identification tasks. Agents can exploit the optimal knowledge from source tasks to accelerate the search rate in new tasks and improve solution accuracy. A greedy action selection rule is adopted for agents to balance the global and local search. The performance of the proposed ITQ approach has been validated on a 68-bus test system. Simulation results in multi-test cases verify that the proposed method is robust and can estimate load parameters accurately. Comparisons with other methods show that the proposed method has superior convergence rate and stability.
As microgrids have advanced from early prototypes to relatively mature technologies, converting data center integrated commercial buildings to microgrids provides economic, reliability and resiliency enhancements for the building owners. Thus, microgrid design and economically sizing distributed energy resources (DER) are becoming more demanding to gain widespread microgrids commercial viability. In this paper, an optimal DER sizing formulation for a hybrid AC/DC microgrid configuration has been proposed to leverage all benefits that AC or DC microgrid could solely contribute. Energy storage (ES), photovoltaics (PV) and power electronics devices are coordinately sized for economic grid-connected and reliable islanded operations. Time-of-use (TOU) energy usages charges and peak demand charges are explicitly modeled to achieve maximum level of cost savings. Numerical results obtained from a real commercial building load demonstrate the benefits of the proposed approach and the importance of jointly sizing DER for the grid-connected and islanded modes.
Uncertain power sources are increasingly integrated into distribution networks and causes more challenges for the traditional load modeling. A variety of distributed load components present dynamic characteristics with time-varying parameters. Toward the end, this paper proposes a robust time-varying parameter identification (TVPI) method for synthesis load modeling in distribution networks, including time-varying ZIP, induction motor, and equivalent impedance models. The nonlinear optimization model is developed and solved by the nonlinear least square (NLS) to find the minimum error between estimated outputs and measurements. To cope with TVPI deteriorated by voltage disturbances, dynamic programming is first used to detect the disturbance. Then, a robust TVPI engine is designed to constrain the estimated time-varying parameters within a stable range. Furthermore, advanced tolerance thresholds are also required during iterations of NLS. Numerical simulations on the 9- and 129-bus distribution systems verify the effectiveness and robustness of the proposed TVPI method. Also, this method can be robust to the ambient noise of measurements.
Energy storage can facilitate the integration of renewable energy resources by providing arbitrage and ancillary services. Jointly optimizing energy and ancillary services in a centralized electricity market reduces the system's operating cost and enhances the profitability of energy storage systems. However, achieving these objectives requires that storage be located and sized properly. We use a bilevel formulation to optimize the location and size of energy storage systems, which perform energy arbitrage and provide regulation services. Our model also ensures the profitability of investments in energy storage by enforcing a rate of return constraint. Computational tractability is achieved through the implementation of a primal decomposition and a subgradient-based cutting-plane method. We test the proposed approach on a 240-bus model of the Western Electricity Coordinating Council system and analyze the effects of different storage technologies, rate of return requirements, and regulation market policies on energy storage participation on the optimal storage investment decisions. We also demonstrate that the proposed approach outperforms exact methods in terms of solution quality and computational performance.
Climate change has raised serious concerns prompting urgent and broad actions that extend current operation techniques. Computational methods and artificial intelligence have already shown promising results in power systems applications, including analysis, forecasting and equipment inspection. Nevertheless, the urgency required in the clean energy shift will substantially increase operation uncertainty, as well as control and planning complexity. Leveraging the capabilities of faster computation, better accuracy and stronger decision-making from cutting-edge computational methods and artificial intelligence can be a promising approach to avoid most impacts in the clean energy transition, while improving system reliability, economics and sustainability. This special issue is focused on inviting original research, reviews and experimental evaluations to promote new computational methods and artificial intelligence applications in low-carbon energy systems. In this special issue, there are a total of 19 original research articles to present the state-of-the-art in energy forecasting, situational awareness, multi-energy system dispatch and power system operation. We would like to thank all participating authors for submitting their works to this special section. We really appreciate the anonymous reviewers' valuable efforts. We are thankful to Prof. David Infield, Prof. Tricoli Peitro and Prof. Panos Moutis, who suggested and supported the creation of this special section and nurtured its initial steps, and to Sophie Robinson, Nageen Matlub and Vinay Kumar Nim, Brianna Cooper, and Bhanuchandar Shanthakumar from IET for their administrative and editorial help. Yishen Wang: Conceptualization; writing—original draft. Fei Zhou: writing-review and editing. Josep M. Guerrero: writing-review and editing. Kyri Baker: investigation. Yize Chen: investigation; writing-review and editing. Hao Wang: investigation; writing-review and editing. Bolun Xu: investigation; writing-review and editing. Qianwen Xu: investigation; writing-review and editing. Hong Zhu: investigation. Utkarsha Agwan: investigation.
Accurate estimation of customer baseline load (CBL) is a key factor in the successful implementation of demand response (DR). CBL technologies implemented at utilities currently are primarily designed for large industrial and commercial customers. The U.S. Federal Energy Regulatory Commission (FERC) order 745 states that DR owners, including residential customers, can sell their load reduction in the wholesale market. However, since residential load is random and un-schedulable, this tends to inherently degrade the effectiveness of existing CBL technologies. In this paper, a novel SAE based CBL method for residential customers that uses the data reconstruction capability of a stacked autoencoder (SAE) is described. In the model, two SAEs are synchronously trained-one SAE generates a pseudo-load pool and the second one is used to select a pseudo-load to reconstruct a residential CBL. A support vector machine (SVM) classifier is self-trained to conduct the pseudo-load selection. The proposed strategy is validated using a real data set consisting of 328 residential customers' smart meter readings. Benchmarks from other machine learning techniques and existing CBL methods are compared with the proposed method. Test results show that the accuracy of the residential CBL reconstruction significantly improves when compared with existing methods, such as HighXofY and exponential moving average.
Traditional load analysis is facing challenges with the new electricity usage patterns due to demand response as well as increasing deployment of distributed generations, including photovoltaics (PV), electric vehicles (EV), and energy storage systems (ESS). At the transmission system, despite of irregular load behaviors at different areas, highly aggregated load shapes still share similar characteristics. Load clustering is to discover such intrinsic patterns and provide useful information to other load applications, such as load forecasting and load modeling. This paper proposes an efficient submodular load clustering method for transmission-level load areas. Robust principal component analysis (R-PCA) firstly decomposes the annual load profiles into low-rank components and sparse components to extract key features. A novel submodular cluster center selection technique is then applied to determine the optimal cluster centers through constructed similarity graph. Following the selection results, load areas are efficiently assigned to different clusters for further load analysis and applications. Numerical results obtained from PJM load demonstrate the effectiveness of the proposed approach.
Stochastic programming methods have been proven to deal effectively with the uncertainty and variability of renewable generation resources. However, the quality of the solution that they provide (as measured by cost and reliability metrics) depends on the accuracy and the number of scenarios used to model this uncertainty and variability. Scenario reduction techniques are used to manage the computational burden by selecting representative scenarios. The common drawback of existing scenario reduction techniques is that the number of representative scenarios is a user-defined parameter. We propose a scenario reduction algorithm based on submodular function optimization to endogenously optimize the number of scenarios as well as rank these scenarios. This algorithm is compared, both qualitatively and quantitatively, with the state-of-the-art fast forward selection algorithm.
Wind power is playing an increasingly important role in electricity markets. However, it's inherent variability and uncertainty cause operational challenges and costs as more operating reserves are needed to maintain system reliability. Several operational strategies have been proposed to address these challenges, including advanced probabilistic wind forecasting techniques, dynamic operating reserves, and various unit commitment (UC) and economic dispatch (ED) strategies under uncertainty. This paper presents a consistent framework to evaluate different operational strategies in power system operations with renewable energy. We use conditional Kernel Density Estimation (KDE) for probabilistic wind power forecasting. Forecast scenarios are generated considering spatio-temporal correlations, and further reduced to lower the computational burden. Scenario-based stochastic programming with different decomposition techniques and interval optimization are tested to examine economic, reliability, and computational performance compared to deterministic UC/ED benchmarks. We present numerical results for a modified IEEE-118 bus system with realistic system load and wind data.
Probabilistic load forecasting (PLF) has gained widespread attention in recent years because it presents more uncertainty information about the future loads. To further improve the PLF performance, this letter proposes a novel PLF method to leverage existing point load forecasts by modeling the conditional forecast residual. Specifically, the method firstly conducts point forecasting using the historical load data and related factors to obtain the point forecast. Then, this point forecast is used as an additional input feature to describe the conditional distribution of the residual on the point forecast. Finally, the point forecast and conditional distribution of the residual are integrated together to produce the final probabilistic forecast. By comparing different point forecasting and quantile regression models, comprehensive case studies obtained from a publicly available load dataset with multiple zones demonstrate the advantages of our proposed method. This letter also informatively reveals the relationship between point and probabilistic forecast accuracies.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
