MERLIN Newsletter Issue 2, December 2015 MERLIN Project Project MERLIN Making Energy Management in Rail System Smarter A summary of the latest achievements Background Energy management is currently and will remain a key issue for railway systems in the foreseeable future. Operating a railway network is a complex operation that does not run as smoothly as scheduled with permanent disturbances and adjustments to the planned operations. As safety and respecting arrival times of trains are the drivers of operations, efficient energy management of an entire network is a key issue for a single supplier, operator or infrastructure manager (as large as they may be). Hence, only through a collaborative approach such as MERLIN can effective solutions for this issue be developed. Appropriately, the MERLIN consortium brings together key rail stakeholders from across Europe. This project has received funding from the European Union s Seventh Framework Programme for research, Technological development and demonstration under grant agreement n 314125 MERLIN is co-funded by EU 7th Framework Programme of the European Commission and is coordinated by UNIFE and is under the technical leadership of CAF. Inside this issue you will find details of what has been developed in the various technical work packages of MERLIN during the last 21 months of the project. You will also find the facts and figures of the MERLIN project and as well as information of the partners collaborating in this important project. The overarching aim of the MERLIN project is to investigate and demonstrate the viability of an integrated management system to achieve a more sustainable and optimised energy usage in European electric mainline railway systems.
What are the key deliverables of MERLIN? MERLIN provides an integrated optimised approach to support operational decisions leading to a cost-effective intelligent management of energy and resources via: Improved design of railway distribution networks and electrical systems and their interfaces; Better understanding of the influence of railway operations and procedures on energy demand; Identification of energy usage optimising technologies ; Improved traction energy supply; Understanding of the cross-dependencies between technological solutions; Improving cost effectiveness of the overall railway system; Contribution to European standardisation (TecRec). MERLIN outcomes have also been developed through the application of solutions to realistic scenarios (see section Scenarios Evaluation & Proof of concept MERLIN results in a nutshell The project is finished and several key milestones have been achieved including: Analysis and identification of components in the rail system, which have an impact on energy usage; Application of the MERLIN main outcomes through concrete simulations/demonstrations in real scenarios; Delivery of two Proposals for Technical Recommendations on Railway Energy Management System s architecture and location of Energy Meters in the grid and Energy and power related information protocols at operational level ; Final release for the MERLIN architecture for Railway Energy Management System; Several meetings of the Rail Reference Group involving railway operators to help guarantee the applicability of the MERLIN outcomes to real operating conditions; Successful hosting of the MERLIN mid-term conference in parallel with UIC Energy Efficiency days on 17 June 2014 and the Open Workshop MERLIN opens its doors. Be witness to the Spanish Scenario Demonstrator held on 5 October 2015 in Malaga at the occasion of field tests. Figure 1 The MERLIN Project Work Packages and Structure
MERLIN Newsletter Issue 2, December 2015 Reference Architecture for Smart Energy Use (WP2)- led by D Appolonia The main objective of this work package is the definition and development of the reference architecture for an integrated solution aimed at achieving a more sustainable and optimised energy usage in European electric mainline railway systems. This goal has been achieved by developing a system that monitors the energy consumptions of the different subsystems of the railway network and their components and suggests a smart solution to optimise energy use in the different components of the system. The reference architecture includes Strategic Decision Making Tool (SDMT) as well as an operational Railway Energy Management System (REM-S). The strategic decision making architecture proposed defines and develops the components and interfaces related to the specific strategic decision making functionality, focusing on three main areas: Considerations of the influence of strategic decisions on energy usage within railway systems in terms of objectives, scope and requirements of the decision making process; Based on the previous, Multidisciplinary System Design Optimisation has been identified as well-suited for MERLIN and its specific application to a decision support tool for railway energy use has also been covered; Methodology for the Strategic Decision Making Tool has been developed, for implementation of the tool in WP5. On the operational side, optimisation of energy, power and costs is performed by the REM-S at different levels, as shown in Figure 2 below. The first optimisation level is performed on a daily basis, by calculating the optimum behaviour of the network for the next 24/48 hours; then a minute ahead optimisation is performed, and finally a real time operation generates indications to be fulfilled by the different actors and/or components. Interaction of the REM-S with the electricity market is also foreseen, through the development of the Energy Buyer Decision Maker which is a module of the REM-S that determines the best way to purchase/sell the energy consumed/generated by the railway system managed by the REM-S. The architecture of the REM-S has been prepared according to the Smart Grid Architecture Model (SGAM), a reference model of smart grid architectures for different sectors of application issued by the CEN- CENELEC-ETSI Smart Grid Coordination Group. After the definition of the REM-S architecture, the definition of the new business model applicable to the system has been carried out, in order to evaluate the system usability and applications. Starting from the analysis of business models already used in the electricity market on the one hand and from the conventional business model in the railway domain on the other hand, the new MERLIN business model has been defined. This is based on the REM-S, which brings additional values in the railway domain, allowing an optimisation of energy, power and costs related the different actors involved in the whole system. Figure 2: Operational Railway Energy Management System (REM-S)
Energy efficient components and energy management technological solutions (WP4) led by D Appolonia The activities in this work package were aimed at developing the component layer of a holistic Railway Energy Management System in order to obtain new smart grid structures and new controllable components at infrastructure grid level as well as at vehicle level. The technological solutions that have been chosen follow a system-wide energy management approach. First, the Operational REM-S software (SW) tool was developed based on the reference architecture defined in WP2. A generic algorithm including Day Ahead Optimisation (DAO), Minutes Ahead Optimisation (MAO) and Energy Storage System (ESS) features was implemented for architecture concept test proof. The Operational REM-S tool was distributed and tested within the scenarios that were identified in WP6. An example of the Operational REM-S SW tool Graphical User Interface (GUI) is provided in Figure 3 below. The work of Task 4.2 focussed on the structure of a railway smart grid applied to the Paris-Lyon high speed line of scenario 1. This scenario studied three solutions that could be combined in order to maximise energy savings and power reductions: Solution #1: First, the timetable is drawn offline in order to be less energy consuming and to avoid simultaneous trains accelerating or braking at the same time. Solution #2: Then the Automatic Train Supervision (ATS) solves the conflicts induced by real operation by avoiding useless braking, stopping, re-accelerating, Solution #3: If the first two software solutions are not sufficient to ensure optimal energy feeding into the line, a hardware solution is necessary. Therefore instead of building a new substation, it is possible to implement intelligent energy storage systems and/or alternative energy sources such as renewables. Figure 4 represents the Hybrid Railway Power Substation concept applied in Solution #3. Finally, the study carried out under task 4.3 addressed the analysis and dimensioning of on-board auxiliary loads and development of intelligent control algorithms. In particular, energy savings are corrected automatically by algorithms on the one hand, and on the other hand, submitted for approval to the Infrastructure Manager who is informed of the contractual arrangements of energy supplier, network configuration, time-table, and network status. The energy optimisation process can be represented as in Figure 5. The Local Optimisation System (LOS) establishes the power/energy restriction from the infrastructure according to MAO strategies. This evaluation is sent to the Auxiliary Loads Manager (ALM) which manages and implements energy saving actions through the Central Control Unit (CCU). Figure 3: 1st meeting of the Merlin Rail Reference Group, UIC HQ, Paris, 11 September 2013
MERLIN Newsletter Issue 2, December 2015 Figure 4: Illustration of Solution #3: Hybrid Railway Power Substation concept Figure 5: Representation of a serial development for energy saving
Energy management system development and validation (WP5) led by Newcastle University One of the elements of the management system at the core of MERLIN is a Strategic Decision Making Tool (SDMT), intended to be a decision support system for optimisation of energy use in different railway systems, specifically targeting the strategic decisions required when designing new railway systems or carrying out significant modifications to existing systems, such as timetable changes, new rolling stock, electrification infrastructure, energy storage systems or revising contractual arrangements for the supply of electricity. Following the specification defined in the first half of the project and based on the draft architecture proposed in WP2, this work stream has implemented it by developing three new modules, namely: Core module; Contractual arrangements module (CA); Optimisation algorithm (OA). The core module is central to the SDMT effectively providing the user interface and data interfaces for all the functionalities provided by the SDMT performing tasks such as: Visualise the trade-offs between indicators associated with the four missions; Display the parameters describing the calculated configurations, and collect any additional userdefined configurations for the next iteration; Write a set of input files for the next iteration, with the objective of analysing the effects of changes to the system design and finding configurations with better performance, as well as ruling out configurations that would violate constraints As the name suggests, the optimisation algorithm module implements a genetic algorithm aiming to describe a Pareto front of optimal system configurations. The contractual arrangement module main task is to explore the impact of parameter optimisation in the cost of operating each configuration of the railway system referenced to a defined timetable. Once the leading partners for each module had released them, the WP5 team ran a series of validation tests to confirm their accuracy, integration success and compliance with the specification. This iterative process resulted in a final release of the tool to all relevant parties for use in the simulation activities as part of WP6. Specially, the following scenarios have been using the SDMT, providing useful feedback: Scenario 2 (15kV 16.7Hz, Sweden); Scenario 3 (3kV DC, Spain) ; Scenario 4 (mixed traffic with 25kV 50Hz and diesel traction, UK) Scenario 5 (25kV 50Hz, UK) A screenshot (see Figure 6) of the SDMT showing an optimisation iteration being completed can be seen below. The SDMT can be partially deployed depending on the optimisation variables. It has been developed solely for the purpose of the MERLIN project and to implement the proposed strategic decision making architecture. In order to be operational, it requires the use of a multi-train simulator or similar. Figure 6: Implementation of SDMT architecture
MERLIN Newsletter Issue 2, December 2015 Scenarios Evaluation & Proof of concept (WP6) led by CAF The main objective of WP6 is to evaluate and validate the solutions proposed by MERLIN through real scenarios. The simulations carried out concluded the following in each study case: For scenario 1, French High Speed 25kV AC, both DAO and MAO modules of the REM-S tool were simulated. First, DAO provides recommendations to use the energy storage systems associated to two conventional substations. Then, MAO suggests power limitations on few trains with flexible schedules. Finally, in combination with existing railway multitrain simulation tools, the REM-S tool provides an optimised timetable with power-limited trains, allowing full traffic on the line, without exceeding railway power supply capabilities. In Scenario 2, Swedish 15kV 16.7Hz Intercity service, simulations demonstrated that application of the MERLIN developments can effectively improve the power flow in 15kV 16,7Hz networks. First, with application of the SDMT, network energy demand could be reduced by optimising converter parameters by more than 4%. Also the second objective has been achieved to cut short-term power peaks by a range of 25% by limiting the train s power temporarily without affecting the timetable. Scenario 3 simulations concluded with the optimum location and capacity for an Energy Storage System (ESS) in a 3300 VDC suburban Spanish line. By applying the operational REM-S only, power peaks could be reduced by up to 15% in two of the three substations. Simulations concluded that flexibility in arrival times was necessary to perform benefits within REM-S, because congested traffic and strict time tables do not allow any operational optimisation prospects. Real tests carried out in Malaga validated the architecture and the correct integration of some components of the MERLIN innovations. Scenario 3 simulations concluded with the optimum location and capacity for an Energy Storage System (ESS) in a 3300 VDC suburban Spanish line. By applying the operational REM-S only, power peaks could be reduced by up to 15% in two of the three substations. Simulations concluded that flexibility in arrival times is necessary to perform benefits within REM-S, because congested traffic and strict time tables do not allow any operational optimisation prospects. Real tests carried out in Malaga validated the architecture and the correct integration of some components of the MERLIN innovations. Figure 7: Timetable and electrical infrastructure illustration for Scenario 1
Scenario 4, mixed traffic with 25kV 50Hz and diesel traction, UK explored the optimisation opportunities on a busy mixed-traffic corridor powered by diesel and electricity. The application of the SMDT tool was therefore aimed at illustrating the potential of finding an optimum combination of diesel and electric traction services. The results obtained suggest that the SDMT can indeed guide the user towards strategic choices that are more optimal in terms of overall energy use, based on the test results of the scenario. For scenario 5, 25kV 50Hz, UK, both SDMT and MAO applications from MERLIN were used for simulations. First, SDMT determined optimum no-load voltage and location parameters to minimise energy use. From this optimal configuration, the MAO manages train driving profiles in a coordinated manner in order to optimise energy consumption required for the operation of the rail system. Figure 8: Peak current reduction in a converter station. The remaining peaks in the optimised current profile are due to the simulation method and would not occur in real implementation Figure 9: Los Prados substations power demand profile optimisation with REM-S (red without and blue with REM-S) Figure 10: Scenario 5 - Bournemouth Weymouth line optimised SST configuration
MERLIN Newsletter Issue 2, December 2015 Recommendations for market uptake and implementation & Standardisation(WP7) led by FFE The objectives of WP7 were to ensure the exploitation of the project results by transferring them into standardization and thus leading to implementation/ market uptake. In order to achieve this goal, WP7 was divided into 5 tasks: Exploitation Plan (Task 7.1) The exploitation plan consisted in a general SWOT analysis of the project which had to be considered in order for the projects outcome to be successfully implemented and exploited. This document also highlighted political and legislative aspects which ought to be considered by the MERLIN project. As a summary, this is a live document, revised and updated alongside the results of the project, which provides a general overview of the context in which the MERLIN project is set. The document also includes an annex detailing the project s dissemination plan drafted to ensure effective exploitation among MERLIN end-users. Position Paper for Agent Integration and Interaction (Task 7.2) Based on the MERLIN project achievements and on an analysis of EU legal framework, the position paper analyses the roles that railway agents could play in the electrical sector. The development of the railway smart grid is pushing forward a change in the paradigm of electricity management in railway sector, with richer interactions among the agents (see Figure 11). The aim of this position paper is to describe the technical and economic aspects of the adaptation of electricity distribution in the railway sector to this new paradigm. Proposal for TecRec: Specification and verification of energy and power consumptions of railway systems (Task 7.3) In order to respond to the need to have a better and more standardized integrated energy management, MERLIN partners developed a proposal for Technical Recommendation to analyse the architecture of the Railway Energy Management System (REM-S), including functions/components layers to help identify where measurement equipment (energy meters) should be placed in the grid in order to achieve an integrated energy management. Power flow Economic flow Power System Control (ESO) From Electricity Users (directly/indirectly) Railway System Control (IM) From Railway Infrastructure Users (directly/indirectly) Transmission Grid (TSO) Distribution Grid (DSO) Railway Distribution Grid (IM) Power plants (GenComs) Energy Supplier (Supp) Railway Electricity User (RU, ESS, DES) Electricity Markets (EMO) Figure 11: Agent integration and interaction in electrical and railway systems
Proposal for TecRec: Energy and Power related Information Protocols at Operational Level (Task 7.4) The MERLIN partners also drafted a proposal for Technical Recommendation on Energy and power related information protocols : this proposal identifies what information protocols at operational level should be implemented in order to achieve an integrated energy management system. The document will improve the usage of energy and will have positive impact on the railway system through the opening of the railway market and the possibility of interchangeable spare parts. Guidelines for the implementation of network integration (strategic and operational levels) (Task 7.5) This document is intended to provide guidelines for the practical implementation of the different subsystems of a MERLIN-based railway electrical smart grid. For each of these subsystems, a brief description is provided focusing on their respective functionality, inputs and outputs. Its primary purpose is to identify the main information exchanges, which are inherent to the idea of a Smart Grid. Due to the large amount of investment required, the implementation of a railway smart grid will have to occur progressively: the smart trains and the smart infrastructure will have to coexist with traditional trains and traditional infrastructures. Furthermore, the concept of smart train and infrastructure will evolve as new functionalities are added. In order to provide some criteria for progressive implementation, this guide describes the interdependencies among the subsystems specified in MERLIN and suggests implementation in phases, when it is possible. Furthermore, this guide includes a brief description of the decisions to be made in every temporal horizon and how the different subsystems of MERLIN can support them. WP1.Railway network specification WP2. Reference Architecture for Smart Energy Use WP4. Energy efficientcomponents and energy management technologicalsolutions WP5. Energy management system development and validation WP6. Scenarios Evaluation & Proof of concept Task 7.1. Exploitation plan Task 7.2. Position paper for agents integration and interaction Task 7.3. Proposal for TecRec - Specification and verification of energy and power consumptions of railway systems Task 7.4. Proposal for TecRec -Energy and power related information protocols at operational level Task 7.5. Guidelines for the implementation of network integration (strategic and operational levels) Figure 12: Dependency relationships between WP7 tasks and other packages
MERLIN Newsletter Issue 2, December 2015 MERLIN Conclusions and perspectives for the future MERLIN has proposed an integrated optimisation approach that includes multiple agents, dynamic forecasting and cost considerations to support strategic and operational decisions. First, the project has developed the reference architecture of our proposed railway smart grid concept, namely REM-S, including its functions, interfaces, components and protocols, following the CENELEC-ETSI SGAM model. To serve as a basis for the energy management systems in the railway domain, we have combined the technical developments with new business models to enable and foster their application. The second main achievement is a strategic decision making tool (SDMT) based on MERLIN s architecture, to support the design of the rail smart grid. The tool takes into account the business constraints and suggests modifications for the system layout after analysing (by simulation) the scenario during a complete running cycle. This closes the loop allowing MERLIN to propose a comprehensive solution for the implementation of an interoperable rail smart grid. In that sense the proposed architecture and tools have been evaluated in the high-speed, freight, regional, commuter and mixed freight-passenger traffic scenarios. Significant improvements in subcontracted power, consumed energy and costs have been obtained in most cases. Third, a pilot case of this integrated approach has been deployed in the suburban network of the Spanish city of Malaga for evaluation and assessment, obtaining good results. The deployed components showed the potentials of MERLIN s concept under real operational conditions. Last but not least, MERLIN has issued a number of technical recommendations and implementation guidelines to facilitate the market uptake of the project concepts and results. To sum up, MERLIN has pioneered the holistic management of energy in railways. This opens the door for further studies and a system level demonstrator within future initiatives like Shift2Rail. High Speed REM-S Freight Regional SDMT Mixed Suburban Figure 13: Next steps. All MERLIN public Deliverables and documents are available for download at www.merlin-rail.eu
Project coordinator Technical leader Facts and Figures Total Budget: 7million ( 4.5m EU funded) Duration: 36 months Project Start Date: 1st October 2012 Project End Date: 31 December 2015 Partners: 21 Grant agreement N 314125 Contact us For more information, please visit our website: www.merlin-rail.eu Or contact: andrea.demadonna@unife.org or dekeyzer@uic.org