Program Topic: Distribution Operation

T1-2Thursday 08:20-09:40

08:20 Interaction of Demand Response and Voltage Stability in Smart Grids

Masoud Esmaili (West Tehran Branch, Islamic Azad University, Iran); Ali Vedadi (West Tehran Branch, Islamic Azad University, Iran)

Demand Response (DR) is one of efficient tools in smart grids to manage load profile. When loads alter their consumption to respond to price signals, the voltage dependency of load powers leads to a mismatch between scheduled and obtained power consumption levels; this is called Demand Response Mismatch (DRM). To mitigate DRM and voltage rise after DR, some voltage compensators are used in the inductive mode. However, since DR is implemented in load peak hours, introduction of inductive compensators can jeopardize voltage stability of the power system. In this paper, a method is proposed to manage DRM while enough Voltage Stability Margin (VSM) is retained. The proposed method is tested on a typical 34-bus test system to evaluate its performance.

08:40 Mitigating Overvoltage Scenarios Caused by Large Penetration of Distributed Energy Resources

Alexandre Nassif (ATCO Electric, Canada); Xun Long (FortisAlberta, Canada)

Distributed Generators (DGs) are continuously more prevalent in medium-voltage distribution systems. Steady-state voltage management for these systems has become a major obstacle to power utilities due to DGs’ control mode typically applied, i.e., constant power factor control. Voltage rises are becoming a norm that poses hurdles to accepting many DG connections. This paper addresses this issue and provides existing mitigation strategies utilities are adopting. It also explains the imperative need to develop better mitigation measures.

09:00 Grid Voltage Level Spanning Operational Strategies for Battery Energy Storage Systems in Distribution Grids

Lorenz Viernstein (Technical University of Munich, Germany); Rolf Witzmann (Technische Universität München, Germany); Joachim Przibylla (Technical University of Munich, Germany)

With rising renewable generation and battery storage prices decreasing, storage integration in electrical grids is increasingly discussed as an option to avoid conventional grid extension as well as an option for new business models for utilities and customers. While the majority of available literature focuses only on the low-voltage level, the impact distributed storages can have on the superimposed medium-voltage grid has to be considered. Especially in grids with high distributed renewable generation, battery storages can contribute to a grid operation conforming to standards. In this paper an operational strategy is presented that includes medium-voltage/low-voltage transformers as well as the high-voltage/medium-voltage transformer into the control algorithm. Results obtained by simulations show, that battery storage systems in low-voltage grids can efficiently relief the superimposed medium-voltage grid and the transformer connecting it to the high-voltage level.

09:20 Comparative Evaluations of Regenerative and Electro-dynamic Braking and Power Substations Along Graded Section of a Japanese Suburban Rail Line

Kosuke Kumagai (East Japan Railway Company, Japan); Tetsuo Fujita (East Japan Railway Company, Japan); Masashi Nakahira (East Japan Railway Company, Japan); Yoshiki Mizuguchi (East Japan Railway Company, Japan); Hideki Sonoda (East Japan Railway Company, Japan)

To reduce the environmental impact of the DC feeding system of the East Japan Railway Company (JR-EAST), it is important to have effective regenerative energy transfer from one train to another. Considering previous studies in simulation, we gathered information helpful for effective energy-saving measures. In this paper, such measures were adapted to the DC feeding system on a graded section of the Chuo Line, a rail line originating in the Tokyo metropolitan area. Simultaneous measurements were conducted on Chuo Line Series E257 trains and the adjacent substations.
In particular, from these measurements the relation between regenerative and electro-dynamic braking of trains and behavior of the adjacent substations was specifically confirmed in a track section with a falling gradient. Also, we studied estimation regarding regenerative power constriction, electro-dynamic braking and regenerative energy on trains using the data from substations to estimate the amount of energy-saving per train. This was by changing substation DC bus voltages for more effective regenerative energy transfer in the future.

T4-1Thursday 15:20-17:00

15:20 Automatic Restoration in Distribution Systems Considering DG Transfer and Islanded Microgrids

Mohamed El-sharafy (York University, Canada); Hany Farag (YorkU, Canada)

This paper proposes a new centralized control scheme for automatic back-feed service restoration in distribution systems with high penetration of distributed generation (DG). The proposed restoration algorithm introduces three types of power transfer between the tied adjacent feeders during the restoration process: load transfer, DG transfer and a combination of load/DG transfer. Another feature of the proposed scheme is the ability of forming islanded microgrids (IMGs) in the post-restoration network by utilizing the available DGs. The objectives of the proposed algorithm are to 1) maximize the restored loads, and 2) minimize the switching operations taking into consideration the load variability during the fault occurrence and the system operational constraints. Several case studies have been carried out on a typical distribution system with multiple feeders to show the effectiveness and robustness of the proposed algorithm.

15:40 Optimal Power Flow with Demand Participation of RESs

Mutlu Yilmaz (Dalhousie University, Canada)

The paper is concerned with a semidefinite program (SDP) to solve the optimal power flow (OPF) problem for integrating renewable energy sources (RESs) with demand participation in electric grids. The electric generation from renewables as a supplier is randomly realized using an algorithm which is intended to demand participation. The demand participation is provided by demand-side resources with renewables to curtail the native loads, which means that the penetration of renewable generation is raised. Thus, a balance between supply and demand response of the grid is maintained. The optimization problem, accommodating the excess of the renewable generation, is introduced as a convex problem. The convex problem is represented in the form of the semidefinite program. Due to the nonconvexity of the optimal power flow problem, the convex semidefinite relaxation has been proposed to solve the optimization problem. We also perform contingency scenarios and test these scenarios by solving the semidefinite relaxation to obtain feasible solutions.

16:00 Applying Machine Learning Techniques for Forecasting Flexibility of Virtual Power Plants

Pamela MacDougall (TNO, The Netherlands); Anna Kosek (Technical University of Denmark, Denmark); Henrik Bindner (Technical University of Denmark, Denmark); Geert Deconinck (KU Leuven, Belgium)

Previous and existing evaluations of available flexibility using small device demand response have typically been done with detailed information of end-user systems. With these large numbers, having lower level information has both privacy and computational limitations. We propose a black box approach to investigating the longevity of aggregated response of a virtual power plant using historic bidding and aggregated behaviour with machine learning techniques. The two supervised machine learning techniques investigated and compared in this paper are, multivariate linear regression and single hidden layer artificial neural network (ANN). Both techniques are used to model a relationship between the aggregator portfolio state and requested ramp power to the longevity of the delivered flexibility. Using validated individual household models, a smart controlled aggregated virtual power plant is simulated. A hierarchical marketbased supply-demand matching control mechanism is used to steer the heating devices in the virtual power plant. For both the training and validation set of clusters, a random number of households, between 200 and 2000, is generated with day ahead profile scaled accordingly. Further, a ramp power (power deviation) is assigned at various hours of the day and requested to hold for the remainder of the day. Using only the bidding functions and the requested ramp powers, the ramp longevity is estimated for a number of different cluster setups for both the artificial neural network as well as the multi-variant linear regression. It is found that it is possible to estimate the longevity of flexibility with machine learning. The linear regression algorithm is, on average, able to estimate the longevity with a 15% error. However, there was a significant improvement with the ANN algorithm achieving, on average, a 5.3% error. This is lowered 2.4% when learning for the same virtual power plant. With this information it would be possible to accurately offer residential VPP flexibility for market operations to safely avoid causing further imbalances and financial penalties.

16:20 Calculation of Line Loss in Low-Voltage Transformer District Based on BP Network Model Optimized by LM Algorithm

Liu Li-ping (Chinese Socite for Electrical Engineering, P.R. China)

A novel method of calculating line loss rate in transformer district is presented and realized by programming, which is BP network model based on LM algorithm. Establish the characteristic index system according to electric characteristics parameters of samples. The classification of samples by K-Means clustering algorithm solves the numerical dispersion of line loss rate in transformer district. On this basis, each class is learnt and trained by BP network optimized by LM algorithm. BP network is adopted to map complex non-linear relation between line loss rate and electric characteristic parameters. Variation of transformer district line loss rate is obtained under different grid structures and load characteristic parameters. The transformer districts in the real system as an example, simulation and calculation are performed to verify the accuracy of the proposed method. This method has the advantages of fast convergence and high accuracy.

16:40 Current Modulated Messages Generated by PFC Circuits for Allocating Appliances with Management Systems in Smart Grid Applications

Christoph Aldejohann (TU Dortmund University, Germany); Thomas Wohlfahrt (TU Dortmund University, Germany); Christian Rehtanz (University of Dortmund, Germany); Johanna Myrzik (Technische Universität Dortmund, Germany)

This paper shows a technical solution to locate and to identify an electrical load and to allocate it with controllers for interchanging clearing and control information e.g. in smart grid applications. It is a solution where the controller can be easily integrated in smart meters, substations and in external control devices. The appliance uses its integrated switching power supply for establishing a communication link. Therefore the control of the integrated power factor correction (PFC) is modified. Thus plugging to a normal socket-outlet is sufficient for allocating the device with controllers in a way like a Plug’n’Play system. The technique does not need new lines and can be installed in existing electrical installations with minimal effort. The allocating technique uses two different communication channels. The first one is a unidirectional channel with a very low frequency which uses the electrical current as a carrier, and the second one is a bidirectional channel. The management system detects, if an appliance is connected to the pathway behind the controller and it establishes an individual communication link. The controller can check its responsibility because grid’s impedance for low frequencies is much lower than the impedance of the connected loads so that the current flows only on the pathway to the transformer station. The method has some more features such as setting up a hierarchized system with several management units for different tasks such as power management, clearing services or congestion management.