SIMULATING HYBRID ENERGY GRIDS IN SMART CITIES FOCUS ON ELECTRIC ENERGY SYSTEMS

Transcription

1 Sawsan Henein AIT Austrian Institute of Technology Electric Energy Systems Research Group SIMULATING HYBRID ENERGY GRIDS IN SMART CITIES FOCUS ON ELECTRIC ENERGY SYSTEMS Sustainable Places 2015 Savona, Italy, September 17, 2015

2 CONTENT Intelligent hybrid energy grids in smart cities Overall aim of WP4 OrPHEuS approach Interfacing of thermal and electric system simulators Hybrid grid co-simulation example Hybrid energy Grid modeling Electricity domain modeling Electricity Grid modeling Aspects Grid Models Load Profiles Ulm Load Profiles Skellefteå

3 INTELLIGENT HYBRID ENERGY GRIDS IN SMART CITIES Image: National Grid Smart Grid Pilot Proposal ( 3

4 OVERALL AIM OF WP4 System Modelling and Simulation for Evaluation of OrPHEuS Control Strategies WP 4 supports OrPHEuS STO 3: Extended hybrid energy network modeling of cities Hybrid Energy Networks, implementing the following main objectives: Development of an enhanced multi domain simulation environment (focusing on electricity and district heating), on extension of existing energy grid simulation environments with fine granulated level of multi domain system modeling Definition and Analysis of the technical requirements for cooperative control strategies, as well as for validation of related control concepts Simulate and analyze the developed cooperative control strategies with different scenarios and use cases 4

5 OrPHEuS APPROACH Co-simulation approach dynamic coupling of domain-specific simulation tools Biggest advantage is modularity use best available tool for modeling and simulation of sub-system modelers of different domains can continue using their preferred tools Approach faces two main challenges interfacing of models/applications data access, start/resume/stop execution of model, etc. orchestration of simulation components during runtime synchronization of models/applications, data flow, parallelization, etc. Avoid re-inventing the wheel, but rely on existing state-of-the-art solutions application interfacing: Functional Mock-up Interface (FMI) specification simulation orchestration: Ptolemy II (simulation framework) The OrPHEus simulation framework is developed on top of Ptolemy II and the FMI++ library, an open-source software utility library developed at the AIT Austrian Institute of Technology. 5

6 INTERFACING OF THERMAL AND ELECTRIC SYSTEM SIMULATORS The FMI++ library facilitates the easy integration of models and tools District heating system simulation district heating networks are modeled using Dymola Dymola is capable of exporting models according to the FMI specification these models can be directly accessed using the FMI++ library Electric system simulation: electric networks are modeled with PowerFactory PowerFactory provides its own simulation interface (API) that is not in compliance with the FMI specification PowerFactory has been extended with FMI-compliant wrapper interface using FMI++ functionalities 7

7 HYBRID GRID CO-SIMULATION EXAMPLE Scenario: electrical grid (PowerFactory) district heating network (Dymola) coupled via hybrid domestic heat water supplies in residential buildings Graphical representation of co-simulation setup (Ptolemy II view) Ptolemy II hides complexity of the setup behind lean graphical user interface blocks represent full models designed by domain experts tools accessed via FMI-compliant interfaces 9

8 HYBRID ENERGY GRID MODELING Electricity domain Modeling The implementation of the electricity systems of the two demo sites Skellefteå, Sweden and Ulm, Germany (generation units, distribution lines and demand models) were done using the simulation environment DIgSILENT PowerFactory (DIgital SImuLation and Electrical NeTwork calculation program PowerFactory), which is a commercial tool for power system design and analysis. The network models contain both physical properties of single components (e.g. properties of cables, transformers, etc.) and a graphical visualization of them. The models have been divided into: Network Diagrams: Overview, Single Line and Substation Diagrams of all networks. Network Data: network components, switches, topology, controller models, etc. Network Variations: e.g. reinforcements, new lines or stations. Additional data like: Areas, Zones, Feeders, Routes, Circuits, Paths, Boundaries, Owners, Operators. 10

9 ELECTRICITY GRID MODELING ASPECTS Electricity domain modeling Detailed level of the modeled Grid An important aspect is the level of details of the modeled power system. The models were designed and parameterized on the basis of three phase balanced connection of both loads and photovoltaics. Simulation type Depending on the focus of the analysis, the tools and the Modeling approach might vary. As one of the most relevant aspects the type of simulation has to be defined. The two main simulation types used in OrPHEuS are steady state simulations (load flow) and dynamic simulations (RMS simulations). RMS Simulation (time-domain simulation for stability analysis), and EMT Simulations (time-domain simulation of electromagnetic transients). 11

10 ELECTRICITY GRID MODELING ASPECTS Electricity domain modeling Load Flow Analysis Load flow calculations are used to analyse power systems under steady-state non-faulted (short-circuit free) conditions. Steady-state is defined as a condition in which all the variables and parameters are assumed to be constant during the period of observation. The load flow calculations were applied in normal system conditions as follows: Calculation of steady-state initial conditions for stability simulations. Calculation of branch loadings, transformer loading, system losses and voltage profiles. Simulation of normal operating conditions: it is sufficient for the load flow calculation to provide the active and reactive power of all loads. Balanced RMS Simulation (Dynamic Simulations) The balanced RMS simulation function uses a symmetrical, steady-state representation of the electrical network. Using this representation, only the fundamental components of voltages and currents are taken into account. Because of the symmetrical network representation, the basic simulation function allows the insertion of symmetrical faults only. 12

11 GRID MODELS The models have been implemented in the simulation environment DIgSILENT PowerFactory, based on the element library of global types of PowerFactory and user defined types as described the DSO s. Electric system components models in PowerFactory are divided into two main parts: the first one is element general description, where general data as name, connection points/ nodes and reference to types are defined. the second one is the element type where specific parameters are defined according to the components manufacture data sheet and/or specifications defined by the distribution network operators. In the second part which is the so called element type, the component is parameterized in different manners and ways corresponding to the technology used, the type and target of the simulation done, and also the technical constraints and requirements to be investigated 13

12 GRID MODELS Transformer Model The two-winding transformer model is a detailed model for various kinds of threephase, two-winding transformers in power systems. It is used to describe the element general description part of the model. the general model is described and is valid for all PowerFactory calculation functions. Particular aspects such as saturation or capacitive effects (only relevant for some calculation functions like Load Flow and RMS simulations) are described in the type of element part. All parameters are defined according to the specifications of the network operators. The required parameters depend on the purpose, the type of the simulations and also on the operational and technical conditions required to be investigated. 14

13 GRID MODELS Transformer Model Parameter Technology Rated power Table 2: Transformer model basic data (type and element) Unit MVA Nominal frequency Hz Rated voltage HV side kv Rated voltage LV side kv Positive sequence impedance: short circuit voltage uk % Positive sequence impedance: Copper losses kw Positive sequence impedance: SHC voltage (Re(uk)) ukr % Positive sequence impedance: Ratio R/X Vector Group: HV side Vector Group: LV side Vector Group: Phase shift *30deg Zero sequence impedance: short circuit voltage absolute uk0 % Zero sequence impedance: short circuit voltage Resistive part ukr0 % Tap changer: at side Tap changer: additional voltage per tap % Tap changer: neutral position Tap changer: maximum position Tap changer: minimum position 15

14 GRID MODELS Distribution lines models As aforementioned the distribution lines models are also divided into two parts. the general one which is the element data (ElmLne) the specific part which is line type (TypLne), where parameters are defined according to the specifications of DSO s are done for the simulation types. The model uses the equivalent PI-circuit to represent AC transmission lines with lumped parameters over phase technology (3ph, with/without neutral conductor and ground wires) see D4.1 Annex II. The ElmLne is element used to represent transmission lines/cables. It requires a reference line type. The electrical parameters are defined per unit-length of the line at power frequency. These parameters remain unchanged; if the frequency of the simulation changes i.e. differs from the power frequency, then the program will adjust the reactance and susceptance of the line according to the new frequency. The inductances and capacitances remain however unchanged. 16

15 GRID MODELS Distribution lines models Table 3: Input parameters of the line element and type Description Unit Range Default Rated voltage kv x>=0 0 Rated current ka x>0 1 Rated current in air ka x>0 1 Nominal frequency Hz x>=0 50 Cable / OHL (overhead line) Cab System type AC:DC AC No. of phases 01:02:03 3 No. of neutrals 00:01 0 Parameters per Length 1,2-Sequence: Resistance R' (20 C) Ω/km x>=0 0 Parameters per Length 1,2-Sequence: Reactance X' Ω/km 0 Parameters per Length 1,2-Sequence: Inductance L' mh/km x>=0 0 Parameters per Length Zero Sequence: Resistance R0' Ω/km x>=0 0 Parameters per Length Zero Sequence: Reactance X0' Ω/km x>=0 0 Parameters per Length Zero Sequence: Inductance L0' mh/km x>=0 0 Parameters per Length Neutral: Resistance Rn' Ω/km x>=0 0 Parameters per Length Neutral: Reactance Xn' Ω/km x>=0 0 Parameters per Length Neutral: Inductance Ln' mh/km x>=0 0 Parameters per Length 1,2-Sequence: Max. operational temperature C x>=20 80 Parameters per Length 1,2-Sequence: Conductor material Al Nominal cross section mm*2 0 17

16 GRID MODELS Load models The loads are represented as static load models considering voltage dependency. The general load element in PowerFactory may be used in conjunction with the general load type. The loads are specified as symmetrical for running balanced load flow. the input parameters for the load is specified based on the load data available from network operators For load flow analysis, it suffices to only specify the electrical consumption of the load called general load mode (real power P). Other data characterizing a load, such as the number of phases and the voltage dependency factors are defined in the general load type assigned to the load element. If no Type is specified, a balanced, three-phase load is assumed, having default parameters for voltage dependency e_cp=1.6 and e_cq=1.8. (constant impedance) see D4.1 For performing load flow analysis, the technology has to be defined which is (3-PH D or 3-PH-N). 18

17 GRID MODELS Generation models Photovoltaic Ulm Photovoltaic generation models are considered as negative load flows with profiles generated based on measured solar radiation and PV technology used (data according to Hochschule Ulm). The solar radiation is taken from the online-available MACC-RAD service and based on satellite-measurements according the Heliosat-4 method. The solar irradiance is calculated on the module plane of each PV system for the existing systems or on the roof orientation for potentials. Further details can be found in deliverable and The same models described above for modeling loads are used for modeling the photovoltaic units characterizing their generation with negative profiles. 19

18 LOAD PROFILES ULM Load profiles refer to both the profiles of photovoltaic electricity generation (supply side) and to the profiles of the electric energy demand (demand side). Supply side Photovoltaic generation models are considered as negative load flows with profiles generated based on measured solar radiation and PV technology according to the data specifications of Hochschule Ulm. Demand side Load profiles were used and generated on the basis of annual measured energy consumption and synthetic normalized load profiles according to the method of BDEW for households and non-residential usage (Ulm, 2014). 20

19 LOAD PROFILES SKELLEFTEÅ Load profiles refer to both the profiles of CHP plant (H2) electricity generation (supply side) and to the profiles of the electric energy demand (demand side). Supply side Biomass combined heat and power plant H2 generation model is considered as negative load flows with profiles generated based on measured data Demand side Load profiles were used and generated on the basis of annual measured energy consumption and synthetic normalized load profiles on the basis of Austrian standard profiles for households and for non-residential usage (APCS, 2012) (no standard profiles available from Skellefteå Kraft). Electric boiler model The electric boiler is modeled as an additional consumption (load) and the same models described above for modeling loads are used for modeling the electric boiler. Generation models Biomass combined heat and power plant H2 The CHP generation model is considered as a negative load flow with profiles generated based on measured data. The same models described above for modeling loads are used for modeling the CHP generation units characterizing its generation with negative profiles. 21

20 FINAL GRID MODELS Electricity domain (Skellefteå) MV network 130/10-30kV Electricity domain (Ulm) LV network 10/0.4kV 22

21 CONTACTS AND DISCLAIMER Project coordinator Ingrid Weiss & Silvia Caneva WIP Renewable Energies Project partner DI. Helfried Brunner, MSc Deputy Head of Business Unit AIT Austrian Institute of Technology GmbH Energy Department Electric Energy Systems T +43(0) F +43(0) The OrPHEuS project is co-funded by the European Commission within the 7th Framework Programme 'Smart Cities' The sole responsibility for the content of this publication lies with the authors. It does not necessarily reflect the opinion of the European Commission. The European commission is not responsible for any use that may be made of the information contained therein. 23