Smart Distribution System OperaTion for MAximizing the INtegration of RenewABLE Generation

Final Report Summary - SUSTAINABLE (Smart Distribution System OperaTion for MAximizing the INtegration of RenewABLE Generation)

Executive Summary:
In this report an overview of the main achievements of the SuSTAINABLE project is given and the basis for exploitation of its results are set.
Firstly, a technical reference architecture for smart grids has been proposed based on four hierarchical layers with distributed intelligence. Its main objective is to enable accommodating increasing shares of renewable energy taking advantage of different Distributed Energy Resources (DER) that can be managed by the Distribution System Operator (DSO). Then, advanced tools for supporting distribution network operation have been proposed and developed. This resulted in a set of pre-prototypes for tools that not only provide forecasts and increase the observability of the distribution grid but also enable a coordinated control of its resources. These tools have been tested and validated through simulation and proof-of-concept in controlled conditions exploiting the laboratory infrastructures available and some of them were also demonstrated in the real smart grids pilot located in Évora, Portugal. Furthermore, planning tools that incorporate the possibility of managing assets in order to differ investments in grid reinforcement and exploiting flexibility for power quality planning purposes have also been developed and evaluated via simulation. Additionally, advanced protection schemes have also been proposed and extensively tested and evaluated in laboratory conditions. Moreover, a regulatory analysis has been performed, which involved mapping the functionalities developed and regulatory topics as well as a revision of current regulatory framework in some of the countries of the SuSTAINABLE partners (Portugal, Greece, UK and Germany). Then the main barriers were identified and a set of recommendations was produced in order to overcome these barriers. Also, scalability and replicability issues have been addressed which are important for the large-scale deployment of the SuSTAINABLE concept. This was done through questionnaires that allowed identifying the main barriers for deployment of the functionalities which then enabled analysing the implementation conditions in different European regions and propose mitigation strategies.
Besides, the potential impact of the proposed solutions was analysed. This was done based on Key Performance Indicator (KPI) calculation during the implementation of the designed functionalities, on the Cost Benefit Assessment (CBA) for the different cases and countries based on the Joint Research Commission (JRC) methodology and on the boundaries set by regulation as well as scalability and replicability potential.
Finally, dissemination activities have also been undertaken in order to foster the exploitation of the project’s results. This included several publications in renowned journals and international conferences with peer-reviewing have been produced, which underlines the innovative character as well as the quality of the research conducted. Lastly, the main societal implications have been identified.

Project Context and Objectives:
SuSTAINABLE arises from the need to further push the European R&D for maximization of renewable generation integration in electrical grids, and in particular the distribution grids. The advent of Distributed Generation (DG) faces considerable challenges and requires significant changes in the way the power system is regarded at many levels, from planning to operation, since the networks are changing from passive to fully active networks, shifting towards smarter and more efficient operation of distribution systems.
The main technical challenges to increased DG penetration are:
• Network voltage changes – The voltage effect in particular is a key factor that can limit the amount of DG capacity to be connected to the distribution system, especially in rural networks.
• Power quality issues – Two main aspects are usually considered: transient voltage variations and voltage harmonic distortion. Depending on several issues such as capacity, location, etc. the effect of DG on the network can be either positive or negative.
• Congestion problems – In some scenarios, DG may alter branch flows significantly, which may pose additional problems in terms of managing energy flows. This may ultimately cause branch overload, especially in the case of high levels of renewable-based DG integration, which may inject large amounts of energy into the distribution system.
• Protection issues – Several different aspects can be considered here and need to be carefully addressed: protection of the generation equipment from internal faults; protection of the faulted distribution network from fault currents supplied by DG; anti-islanding or loss-of-mains protection (islanded operation of DG is likely to be allowed in the future as DG penetration increases) and impact of DG on existing distribution system protection.

Following the pathway to the development of integrative distribution energy networks, SuSTAINABLE developed and demonstrated a new operation paradigm, leveraging information from smart meters and short-term localized forecasts of RES and consumption in order to manage distribution systems in a more efficient and cost-effective way, thus enabling a largescale deployment of intermittent renewable based DG. SuSTAINABLE concept is based on a cloud-like principle, where the Distribution System Operator (DSO):

• Collects information from a smart metering infrastructure, other distributed sensors and communications from external partners;
• Processes the information using tools such as state-estimation, forecasting tools, data mining and decision-making applications, where part of this information processed in a distributed way on data concentration devices located in the distribution grid;
• Communicates settings to power quality mitigation devices, protection relays and actuators, distribution components (e.g. On-Load Tap Changers – OLTC, reactive power compensation devices) and distributed flexible resources;
• Assesses its market strategy as a provider of ancillary and balancing services.

Project Results:
Main Results / Foreground

1) Technical Reference Architecture

The main aim of the proposed control architecture for the SuSTAINABLE project was to enable a coordinated and efficient control of the whole electrical distribution system, taking advantage of its own resources in order to overcome technical problems that may arise in operation especially in scenarios with high integration of RES.
A general framework for the data flow model of the SuSTAINABLE concept is presented in Figure 4. This framework comprises two different types of information: commercial (related to billing information) and technical (related to operational information).
At the top level is the DSO Central Information System. All the billing information from the customers down to the LV level, transmitted by the AMI, must be processed in an Advanced Metering Management (AMM) module located at the central information system level. Also, since several control algorithms that may be envisaged (such as forecasting functions, load flow analysis, state estimation routines) require the knowledge of the exact position of the DER, it will be necessary to have Geographical Information Systems (GIS) at the central information system level that should cover all the territories that the DSO is responsible for. A similar situation occurs for the technical characteristics of the lines, substations, transformers or smart meters, which should be stored in a central database – Enterprise Resource Planning (ERP) / Asset Management (AM). This central database also interacts with the Outage Management System (OMS) of the distribution network assets. A Communication and Event Management System module, at the SCADA/DMS level, is also considered for functions such as outage analysis and pattern detection, communication failures with pattern detection.
Finally, the DSO can be a market facilitator that collects all commercial (e.g. billing measurements) and technical (e.g. consumption, generation, technical constraints) data and provides this information to the market agents (e.g. GENeration COmpanies – GENCO, Energy Service COmpanies – ESCO, retailers, etc.) via DSO Central Information System. The DSO also provides data to the TSO in order to support its operational management processes (e.g. detect reverse power flows from the distribution grid) and facilitate the DSO/TSO cooperation. Ancillary services activated bids are communicated by the TSO and must also be taken into account since it is not realistic that they will communicate directly with each SCADA/DMS.
In order to fully realize the vision of the SuSTAINABLE concept, a technical architecture for management and control of the distribution system as a whole was conceived. This novel architecture, shown in Figure 5, is based on the hierarchical architecture already deployed at the InovGrid test site in Évora, Portugal – InovCity. The main objective of the reference architecture proposed is to allow large scale integration of DER, namely DG based on variable RES such as wind generators and photovoltaic panels, in a secure and efficient way.
The reference architecture proposed is organized in four main layers:
• The upper control level with the Supervisory Control and Data Acquisition / Distribution Management System (SCADA/DMS) located at the control centre of the distribution system (i.e. dispatch level). This layer is under the responsibility of the DSO for managing the whole distribution network and should ensure the interface with the upstream transmission network.
• An intermediate control level located at the HV/MV primary substation – the Smart Substation Controller (SSC) – which is in charge of each MV network and incorporates a set of advanced control and management functionalities that will allow a coordinated and efficient operation of the MV system exploiting the multiple resources that may be available at this level through set-points, namely storage systems (STOR MV), controllable loads under Demand Side Management (DSM) actions (CL MV), Distributed Generation units (DG), On-Load Tap Changing (OLTC) transformers (OLTC) and capacitor banks (CAP MV).
• A lower control level located at the MV/LV secondary substation – the Distribution Transformer Controller (DTC) – which is responsible for a single LV network. This control layer is used to serve as a gateway of data to the upstream systems but will also incorporate some basic control functionalities in order to efficiently respond to technical problems that may occur at the LV network level by communicating set-points to the several smart meters and corresponding DER under its control as well as for MV/LV On-Load Tap Changing (OLTC) transformers and storage devices (property of the DSO) that may be located at the secondary substation.
• A field control level located at the customer premises in which the Smart Meter will serve as a gateway to control its associated resources, namely micro-generation (µG), controllable loads under DSM actions (CL LV), storage devices (STOR LV), and Electric Vehicles (EV).
Furthermore, an evolution of this architecture was foreseen where future developments at the LV level were taken into account, mainly through the integration of domestic customers through a Home Energy Management System (HEMS), which will be responsible for coordinating the DER at the level of each LV client.

2) Tools for Distribution System Operation
Within the SuSTAINABLE concept, the development of a set of advanced functionalities to enable maximizing the integration of RES in distribution networks was foreseen. This intended to provide the DSO with tools that could assist him in distribution system management and support daily operation. The rationale behind the proposed approach is supported by the triad “PREDICT – OBSERVE – CONTROL” as shown in Figure 6.
This ultimately resulted in a set of pre-prototype tools that were evaluated through proof-of-concept via simulation and in a controlled laboratory environment and ultimately demonstrated in the real smart grids’ pilot in Évora – Inovcity. The main features of each of these tools are summarily described in the next sub-sections.

2.1) Advanced forecasting tools for reliable prediction of load at the MV level
Load surveys were performed in order to characterize the availability of customers for demand side actions (controllable loads), consumption strategy, energy use (residential, industry, commercial, etc.). This enabled identifying typical profiles (demand curves) for classes of consumers based on their daily diagram shape and consumption magnitude were determined. Also, load forecasting tools (e.g. ANN, ANFIS) were applied to provide accurate estimates of load, with a time horizon of 15 minutes to 3 hours.
The proposed approach is innovative as it not only targets an accurate forecast of the amount of load demand, but also addresses the composition of the forecasted load and the expected evolution of its components, distinguishing particularly between controllable and non-controllable loads, at different network buses including aggregate load static and dynamic behaviour following network disturbances or control actions. Another particular feature of developed methodology is the capability of predicting not only amount of controllable load, but also its dynamic response to network disturbances over varying time scales.
In addition, a load forecasting tool was developed in order to provide improved hourly, 24 hour ahead predictions. Random forests are employed in four modeling architectures that form a committee of independent forecasters tuned at different prediction horizons. The prediction committee is fused in a combination unit that produces predictions in a 24 hour forecasting window. Particular emphasis has been paid to the time required for building the committee of prediction units in the proposed forecasting tool in view of the large amount of metering points.

2.2) Advanced local forecasting tools to predict renewable generation
A novel neural architecture using radial basis functions was developed to solve complex non-linear problems like solar and wind power forecasting. The novel network consists of multiple neuron groups. Same information is saved to several neuron groups and processed with different ways. Namely, a neuron group has a different perception for an instance. A neuron group provides its own prediction and contributes to the final prediction depending on its activation. The network is built from Radial Basis Functions (RBFs) with different widths to their kernel point position. Subsequently, it can analyze the information in great detail retaining its generalization ability. In this application, the network was applied together with conventional RBF neural networks consisting a powerful forecasting tool.
A solar power forecasting system was also developed that uses different types of statistical models that combine information from weather predictions and time-series data collected by smart meters geographically distributed in space.
The forecasting system generates point and uncertainty forecasts for the next 60 hours and with time resolutions ranging between 15 and 60 min. and combines past observations of time-series distributed in space, collected by smart meters and other sensors, in order to capture the effect of cloud movement in solar generation. Moreover, its parameters are time-adaptive and cope with time-varying operating conditions and it includes an upscaling module to generate forecasts for the total solar power in each secondary substation. Finally, it has data quality control and pre-processing modules that generate alerts in case of communication failure or when the protection system of the PV installation is trigged.

2.3) Advanced local distribution grid monitoring / state-estimation
A new method for the generation of pseudo-measurements was developed that only uses information coming from LV assets to provide accurate pseudo-measurement at the correspondent MV/LV substation. Therefore, the method can take advantage of the stronger correlation that exists between the pseudo-measurements generated and the electrical variables of the correspondent downstream LV network.
The obtained results indicate that with auto-encoders properly trained, accurate pseudo-measurements can be achieved, even when a low number of real-time measurements are available. As expected, the addition of real-time measurements leads in general terms to a decrease in the error of the pseudo-measurements generated. The proposed approach could be particularly useful either when substations are not being telemetered in real-time or in case of telemetry equipment failures. In these circumstances, the missing real-time measurements could be replaced by the pseudo-measurements generated and later used to guarantee observability in a DSE conventional algorithm.

2.4) Advanced coordinated voltage control
Within the SuSTAINABLE concept, advanced voltage control involves a coordinated management of the several DER connected at the MV and LV levels in order to ensure a smooth and efficient operation of the distribution system as a whole.
Therefore, an innovative approach for voltage control at the MV level was proposed based on a preventive day-ahead analysis using data from forecasting tools for load and RES to establish a plan for operation for the different DER for the next day and a corrective intraday analysis aimed at minimizing the deviations from the day-ahead plan. Two different implementations of the proposed algorithm were developed and tested with good results. At the LV level, a centralized scheme was developed and embedded in the DTC based on a set of rules that is able to send set-points to inverters interfacing the DER located to solve a voltage violation exploiting data collected from smart meters (for instance, sending a set-point to a storage unit that is geographically close and on the same phase as a client with an overvoltage). In addition, a local control scheme based on droops is embedded in the inverter of the DER in order to quickly react to sudden voltage drop/rise phenomena (for instance, to locally act on a PV unit that has an overvoltage at its terminals due to a change in the primary resource – sun, in order to avoid technical violations).

2.5) Assessment of TVPP contribution to Distribution / Transmission coordination
The concept of the VPP as a market actor and technical supporter for active distribution networks was developed, with emphasis on the services that can be provided by the TVPP. The modelling suite addresses day-ahead, intra-day and close to real time operation.
In day-ahead operation, the VPP’s flexibilities that concern the variability of the RES generation were defined by doing a statistical analysis of the RES forecasting error and the allocation of the DER generation levels takes the DER units’ cost functions and other constraints. Moreover, the VPP is able to re-dispatch its resources accordingly, in case congestion is detected in the distribution network which assumes an active participation from the side of the DSO in the technical validation of the VPP schedules. In the intra-day and close to real time operation, the provision of firm capacity from a TVPP is proposed using decision trees and shows the re-dispatching the TVPP should do in case its largest DG generation is lost. Moreover, the concept of providing primary and secondary reserve from the TVPP’s units was also developed as well as the scheduling of the minute reserve from a VPP. Finally, the concept of a VPP that includes a hybrid station in island systems was also proposed and a simulation tool was developed that enables simulating possible operation scenarios with regards to different RES profiles.

2.6) Provision of differentiated quality of supply
Methodologies for optimal, close-to real time provision of differentiated Quality of Supply (QoS) were proposed for different groups of customers of the network with DG (in particular intermittent and stochastic power electronic interfaced DG). In order to provide differentiated Power Quality (PQ), an approach for zonal PQ classification and assessment of global PQ performance was developed. The proposed methodology is based on development of Zonal PQ thresholds and Unified Power Quality index.
A classification of PQ zones was proposed with different premiums assigned to each grade depending on location, time and type/class of customers. Moreover, a new index, Unified Bus Performance Index (UBPI) was proposed to evaluate the overall PQ performance of the network considering several PQ phenomena simultaneously, namely harmonics, unbalance and voltage sag. In addition, a new approach to combine the comparative scores of actual state and critical state was proposed in order to come up with a numerical score which can represent the overall PQ performance with different weights assigned to different phenomena.

3) Planning and Protection of Distribution Systems
3.1) Network reinforcement planning considering management of distributed flexibility
A multi-stage reinforcement planning tool was developed for distribution networks for the purpose of minimizing investment and operation cost and maximizing the amount of allowed DG and microgeneration, while respecting the technical constraints on voltage levels and branch loading, and considering the management of the distributed flexibility. The tool is applied to MV distribution networks with high penetration of RES. The uncertainty of the processes that are not directly controlled by the DSO (like DG and flexible loads) and that may have negative impacts on future operation of the network is also considered in planning process.
A set of realistic scenarios was considered to assess operation expenditures (including non-delivered energy costs) incurred while meeting various technical constraints over the time horizon of 20 years. Scenarios take into account the representative days during the year, with an hourly time frame, corresponding to a variety of realistic conditions in terms of load demand, DG injection, reserves and balancing prices and contingencies on distribution networks.
Proposed multi-objective approach, as the core of the planning tool, incorporates classical optimization techniques and meta-heuristics to combine multi-objectives using reference trade-offs set by the user. The multi-criteria analysis procedure allows the planners to incorporate their preferences about valuing RES penetration increase within the planning exercise, while considering different DER options to accommodate load and DG growth.
Analysis and validation of the results were performed on a typical distribution network, where DER are assumed to be available. The results obtained demonstrate that by incorporating the active management of the renewable generation, the investment costs on distribution lines and substations can be reduced by 2.5% to 35%. Furthermore, by controlling the active and reactive power of the RES, a reduction of 3%–8% of the total annual energy losses is achieved and the hosting capacity of the network is increased by 3%–21%. The results show that there is a potential for deferring investments in network reinforcement for a few years, while reducing the overall cost of the investment plan.

3.2) Power quality planning for flexible distribution systems
Optimisation based Power Quality (PQ) mitigation infrastructure was developed for distribution networks with the presence of stochastic and intermittent power electronics interfaced DG and power electronics interfaced storage devices in the network. PQ phenomena, including harmonics, voltage imbalance and voltage sags, could lead to interruption to equipment or industrial processes and thus cause significant financial loss to both utilities and customers in distribution networks, hence, the inclusion of these phenomena in development of PQ planning. In reality, requirements about specific PQ performance in the network vary from area to area (e.g. commercial, residential and industrial areas), depending on the sensitivity of customers’ processes and equipment to specific PQ phenomena. Considering different PQ requirements by different parties involved in electricity supply chain, costs associated with PQ mitigation and willingness to contribute to PQ mitigation by different market players, the idea of provision of differentiated levels of quality of supply to different customers in different zones is becoming more and more acceptable. This approach will improve the efficiency of electricity / energy distribution by only offering the PQ performance as required.
In SuSTAINABLE, various methodologies and indices were investigated for evaluating the severity of PQ phenomena in distribution networks. It is considered that these phenomena under consideration vary with time and location due to activities of industrial customers and because of the stochastic and intermittent DG in the network. New PQ gap indices were proposed to evaluate the level of attainment of PQ performance compared to the customer specified thresholds, aiming to enable the provision of differentiated PQ levels in different zones of the network. Based on these indices, temporal and spatial analysis of PQ performance was performed first on a case study of a 295 bus generic distribution network with stochastic and intermittent power electronic-interfaced DG. Following this, a range of PQ mitigating solutions was investigated to ensure cost-effective management of PQ in the network. Flexible AC Transmission System (FACTS) devices and network/plant based mitigation techniques were tested as the potential solutions to the PQ problems at hand. The effectiveness of these mitigation techniques were discussed based on the aforementioned severity indices and further tested in generic distribution network.
An optimisation methodology was then proposed for optimal, cost-effective, PQ mitigation in the network with the focus on the type of mitigating solutions and the optimal level of the mitigation. In this methodology, greedy algorithm is applied to search the optimal mitigation scheme in order to enable the provision of differentiated PQ levels. Taking the gap indices as the objective functions, the mitigation strategy is optimised to meet the zonally specified PQ requirements. The optimality and robustness of the obtained solutions were validated through extensive simulations in DigSILENT/PowerFactory.
Finally, the economic assessment of PQ mitigation at planning level was performed, by considering the resulting benefits of PQ mitigation over the entire life span of the deployed solution. The methodology calculates the overall net present value of future benefits (including initial investment and maintenance cost in solution as well as the cost of remaining PQ losses) with planning or deployment year as the reference. Using the proposed optimisation methodology, PQ mitigation is planned based on the economic analysis while taking the provision of differentiated PQ levels as the constraints. The methodology was illustrated on a large scale generic distribution network, and both the financial and technical benefits of the optimal mitigation solution demonstrated and discussed.

3.3) Planning of Advanced System Protections
Faults in the grid are a major concern for any DSO as they are a threat to human life or to the grid assets if the fault is not quickly detected and cleared. Faults are detected and cleared by protection relays, spread in the network, that follow a carefully selected group of settings that can ensure the needed speed and selectivity so only the faulty lines or assets are disconnected. An advanced protection system was proposed by incorporating flexible schemes for distribution network protection and grid interconnection protection of DG units in order to minimise/avoid protection misoperation and failure.
For the proposed protection schemes, fault-ride-through capability is allowed to be adopted by all installed DG without compromising selectivity in the detection of faults at the feeders where these units are installed. The identification, and adoption of additional protection relays and protection strategies with adaptive settings e.g. Rate of change of Frequency (ROCOF), tele-protection, etc. were investigated. In order to ensure required selectivity of different protection relays, some of pre-defined settings need to be changed taking into account the configuration of the network and the presence of DG units in the grid, as well as the neutral to ground connection solution adopted in the distribution grid (e.g. direct neutral to ground connection or connection through an impedance). The location and characteristics of protection devices were particularly investigated in order to assess the costs effectiveness of the advanced protection functionalities and protection strategies. These new protection relays and protection strategies were evaluated through a range of characteristic tests on an RTDS (Real Time Power System Simulation) hardware / software simulation platform where several failure conditions can be evaluated namely: unsuccessful reclosing sequences due to DG units connected to faulted sections of the network; no operation of feeder protection due to protection blinding; loss of directional criteria in relay coordination; change of protection reach and fault resistance coverage due to changes in short circuit power.
The following specific protection problems were addressed:
• Failed reclosing – DG unit may interrupt the auto-reclosing sequence performed by the feeder relay. The reclosing settings used must be coordinated with the operation of DG protection to avoid problems.
• Sympathetic tripping – It may occur when a fault is located outside the feeder including DG. In such a case, the DG unit contributes to the fault and feeds a fault current ‘upstream’ towards the fault, which may trip the relay located at the beginning of the DG feeder. Coordination algorithms for co-operation of feeder relays and DG protection were also developed.
• Protection blinding – The operation of feeder over current protection may become interrupted when DG unit is located between the fault point and the feeding substation. To avoid increase of fault levels, selective coordination is adapted.
• Loss-of-mains protection – DG units are not able to maintain an adequate level of quality in the network during unintended islanding. Islanding will be detected by voltage and frequency relays located at the DG unit terminal. Methods based on ROCOF (rate of change of frequency) and vector surge were developed to provide reliable islanding detection.
• Impact of fault ride-through – Because DG-units interfere with the protective system of the distribution grid, they should be immediately disconnected in case of a fault, restoring the operation with only one source of supply and functioning the protective system as it was designed to do. Under-voltage relay protective scheme is proposed to detect the voltage dip propagating through the grid and disconnect the DG-units.
The advanced protection systems developed and implemented in SuSTAINABLE enable tackling the aforementioned protection problems. The adequate scenarios for the validation of the implementations were considered by using the limited but extremely flexible settings that allowed all the functionalities to be tested. The results and proof-of-concept validation showed that the developed protection functions are adequate to tackle the documented challenges. The results also showed how the current protection relays can be complemented with new functionalities to make them adaptive to the operational specificities introduced by DG.

4) Proof-of-Concept
The tools for operation and planning, developed in SuSTAINABLE have been evaluated using simulations and real data from existing networks. The overall conclusion is that the tools developed will improve the efficiency and the management of future distribution networks.

4.1) Forecasting
The developed forecasting modules were validated mainly using data from the non-interconnected island of Rhodes according to the evaluation protocol. In this case study, the RES forecasting models were applied to nine PV plants and to four wind farms while the load forecasting model is validated to a HV/MV power station. Due to the high RES penetration to the above substation, the load is recorded as negative making its prediction especially challenging. In addition, the complex terrain of the Rhodes island together with the land-sea breezes reduces the predictability of wind power.
The PV forecasting has also been validated in the Évora case study and the proposed models were tested in different climate conditions. The results from both case studies show that the models have satisfactory performance and are robust to challenging conditions appeared to the energy management in MV/LV level.

4.2) Technical Virtual Power Plant
The TVPP concept was validated through simulation covering the operation from day-ahead to close to real time. Firstly, the tool for assessing a VPP’s internal flexibility to cover RES forecast uncertainties was validated. The results for two days with different RES penetration levels were presented and the internal flexibility for different risk factors was calculated in each case. Moreover, simulations were performed for two different test cases to show the effect of the VPP’s operation on the KPIs for the reduction of carbon emissions and for the share of energy produced by RES. It was shown that the KPI values are strongly dependent on the portfolio of the VPP and the share of RES and storage it contains at each case, however the aggregation and control of different units under one single profile has in both cases a positive impact on these values. Moreover, the ability of the VPP to support the DSO in case of a congestion was validated, showing that the VPP can potentially support the DSO by controlling its DER. The provision of firm capacity by the VPP in the case of sudden loss of a unit with decision trees was also validated and the results were presented with respect to their profitability to the VPP. Furthermore, it was validated that a TVPP can aggregate different distributed resources, as storage systems or controllable loads in order to support primary control reserve. Simulation results in the context of a power system under the ENTSO-E regulatory requirements and for the Continental European power system showed that around 20% to 40% more primary reserve can be aggregated by a TVPP and the frequency response of the system can be improved by around 1 Hz for a loss of generation of a 10%, in a scenario of medium load with high penetration of renewables. Finally, the simulation tool was used to simulate three different scenarios for the island of Rhodes. These scenarios considered the current situation on the island, a scenario with penetration of CSP and a scenario with penetration of CSP and a hybrid station. The yearly energy production for RES, CSP and the thermal units was shown as well as the diesel and heavy oil consumption for 2014-2020, validating that the increase of RES integration will lead to a reduction in fuel costs.

4.3) Advanced coordinated voltage
The multi-objective optimization algorithm that was developed for advanced voltage control in MV networks was validated with simulations on feeders of Rhodes island. Three different case studies were simulated, the case of low load and high RES penetration, the case of large load and medium RES penetration and the case of different MV feeders, as the ones of the previous cases, which are connected to the same HV/MV substation. For all the different case studies KPIs were calculated to assess the improvement of different factors in comparison to the current practice. As a general conclusion it can be stated that optimal control leads to a significant improvement concerning voltage regulation, energy losses and the other optimization targets. Moreover, the performance of the voltage control scheme developed for LV grids was analysed exploiting the laboratory facilities of INESC. Extensive tests were performed to evaluate the behaviour of the control algorithm in different operation conditions and the operation of the smart inverters developed specifically for this purpose. This provided valuable feedback for the demonstration activities of the voltage control functionality at the InovGrid site.

4.4) Advanced Protection Schemes
The advanced protection functionalities were tested in laboratory environment. Firstly, the impact of the penetration of DG in causing protection blinding was evaluated. A test was conducted with and without DG connection to a bus of a distribution network showing that in case of DG unit connection, protection blinding may occur. Moreover, the impact of connection of large DG in sympathetic tripping was studied. The impact of connection of large DG units on the overcurrent protection of neighbouring feeders was firstly studied using the ICCS test bed. Furthermore, the relation of sympathetic tripping to the DG technology was studied, considering synchronous generator based DG units and electronic interfaced DG units. The synchronous generator based DG units showed a larger short-circuit contribution than electronic interfaced units and it was concluded that the first may lead to tripping of the DG unit protection. The adaptability of a commercially relay to abnormal conditions under the control of the SSC was also studied and tests validated that if the SSC is enabled, it can contribute in avoiding sympathetic tripping. Different islanding protection methods, such as UOV, UOF, ROCOF and VS were also tested for various scenarios of power imbalances (∆P, ∆Q). Finally, Fault Ride-through (FRT) tests were conducted on a closed-loop testbed by using an EFACEC digital relay for the control of the DG interface breaker. Finally, additional simulations were performed that validated the new FRT protection function concept.

4.5) Differentiated Quality of Service
Offering a differentiated level of quality of supply helps utilities to price it based on customer’s willingness to pay and minimize overall mitigation costs. For customers located in the same area with shared cost mitigation schemes, several benefits of the differentiated quality zones.
The provision of differentiated QoS was validated using extensive simulations. Firstly, a DSM scheme applied from the VPP to provide temporal power reduction to the DSO during the day was validated with a test case on a representative German feeder. Direct load control was applied for one hour to interruptive loads that are controlled by the VPP by changing the switching of the devices according to their current status, the internal and external temperature and the consumers’ preferences. It was validated that by aggregating and controlling small flexible loads, a temporal power reduction can be achieved, which can reduce possible threats introduced by the variability of RES. Moreover, the methodology for mitigation of voltage unbalance in LV networks was validated through simulation and testing at the laboratory of TUB. The demonstration under experimental conditions validated that voltage unbalances produced by asymmetrical loads in different phases can be mitigated. Two methodologies were validated. On the one hand, a power electronic device in the main feeder was able to inject/absorb zero sequence components. On the other hand, a DG unit injected negative sequence and reactive power to cancel present current and voltage unbalances. The network topology can influence the capability of the devices to mitigate voltage unbalance. One way to overcome that is by having advanced communications infrastructure between DG units, loads, and a centralized or a VPP agent. Finally, the QoS phenomena of voltage sag, voltage unbalance and harmonics were evaluated for different case studies and periods with Monte Carlo simulations on UK based generic networks. The poor performing areas were then identified according to the CBPQI (Compound Bus PQ Index) and a classification of customers based on their QoS requirements (variable spatial QoS thresholds). These levels of QoS performance were used as input to an optimization for QoS mitigation solutions and comparisons were done between mitigated and unmitigated operation. It was validated that by applying the mitigation schemes from the optimization solutions, the different QoS requirements of the customers can be met.

4.6) Reinforcement planning tools
Within the scope of SuSTAINABLE project two planning methodologies have been developed and validated in real distribution networks.
The first methodology, uses complex mixed integer nonlinear programming and was applied to Rhodes distribution network. Two 20 kV distribution feeders of the Rhodes distribution network were analyzed to validate the effectiveness of the developed planning tool for real world distribution networks. The results demonstrate that by incorporating the active management of the renewable generation the investment costs on distribution lines and substations can be decreased by 2% to 35%. Furthermore, by controlling the active and reactive power of the RES a reduction of 2%–6% to the total annual energy losses is achieved and the hosting capacity of the network can be increased by 15%–60%.
The second methodology was applied to the Évora distribution network assuming high levels of PV deployment. The optimization of the distributed flexibility provided by distributed storage devices (DSD) placed at a strategic location enabled managing branch overloads. This entailed optimizing the daily operation strategy of the DSD considering two different economic objective functions: (1) maximizing the economic profit from arbitrage and (2) minimizing the grid’s active power losses. The results show that, in the first case, there is a significant potential for deferring investments in network reinforcement, while reducing the overall cost of the investment plan. However, minimizing the grid’s active power losses does not seem to be rational from an economic point of view, as more energy is consumed by the DSD than the resulting energy savings from the ensuing loss reduction. A multi-criteria analysis procedure allows the planner to incorporate his/her preferences about valuing RES penetration increase within the planning exercise, while considering different DER options to accommodate load and DG growth.

5) Demonstration and Evaluation
The purpose of the demonstration is to validate the developed functionalities in real conditions, using data collected from the network (sometimes, in real-time), resorting to already installed equipment or to equipment installed specifically for SuSTAINABLE demonstration purposes. This is one the most differentiating components of the project, due to the significant development in software, equipment installation and configuration, and mostly, to the effort to follow the proposed technical architecture, especially in live conditions on the field.
Within the demonstration a subset of functionalities developed were validated in the InovGrid site at Évora, Portugal, namely:
• SF1 – Load forecast
• SF2 – Renewable energy forecast
• SF3 – State estimation
• SF4 – Voltage control
• SF5 – Technical Virtual Power Plant

5.1) SF1 – Load forecast
In the case of functionality SF1, the demonstration was in one specific feeder – Casinha. The specific purpose of this demonstration was to forecast active power values for the feeder, for a period that can range from 30 minutes to 24 hours. For this, real data was collected from the network and, together with weather data, was used to perform forecasts according to the algorithm developed. Since SF1 demonstration requires not only network but also weather values, data for Évora region including wind, humidity and temperature values, is provided twice a day. The network data is updated every 30 minutes, while weather data is updated twice a day.
A specific software interface was developed in order to validate this functionality. It allows users to easily obtain forecast values, compare them with real power values and forecast error values, change forecast horizon, etc. The forecast error is calculated according to Mean Absolute Percentage Error (MAPE).

5.2) SF2 – Renewable energy forecast
The purpose of SF2 functionality demonstration is to perform renewable energy forecasts at LV point level and secondary substation level. Since the only available renewable technology generation in Évora region is PV (at LV level), the demonstration focused on this specific technology.
SF2 requires information at the LV level, namely, the energy produced by micro-generators. The collection this information for billing purposes is one of the roles of EDP DISTR as the main Portuguese DSO. However, the frequency of data collection was not sufficient to produce forecasts with the quality necessary for the SuSTAINABLE project, which is focused on short-term forecasts. Therefore, it was necessary to install prototype monitoring equipment in a subset of micro-generators. Criteria such as quality of historical information and geographical dispersion were taken into account. The participation of the customers was entirely voluntary and the scope of the project and of the demonstration was previously explained to them. Through a phone survey it was possible to determine if the customers were interested in participating in the project.
After the criteria analysis and customer survey, a group of 41 micro-generators was selected. These micro-generators are spread throughout 27 secondary substations. Visits were then scheduled to each one of the customer’s houses in order to install the monitoring equipment and to retrieve a protocol signed by the client that states the consent to provide microgeneration data for the purposes of this project. The SF2 application used this information, together with weather data, to process the forecasts using the algorithm developed.

5.3) SF3 – State estimation
The purpose of SF3 functionality and demonstration is to perform state estimation at MV level, but with an innovative feature: the estimator is able to determine MV measurements by incorporating measurements collected at LV level (as long as they belong to the same LV network), through the use of an auto-encoder. LV historical data is used for model training and pseudo-measurements generation. Real data from a subset of smart-meters is necessary every time a state estimation is performed. For the purposes of this functionality, it was necessary to select a MV network and, within this feeder, a LV network for the provision of LV measurements.
The selected substation was Montemor’s primary substation, and the selected LV network is located in Guadalupe village. The system installed at Montemor primary substation includes a workstation that can be used to visualize the network state using synoptic diagrams that have been specially built for this purpose. This workstation also offers an application with a Graphical User Interface (GUI) that is used to control and display results of Power functions.

5.4) SF4 – Voltage control
The purpose of SF4 functionality is to ensure that voltage values remain within regulatory limits, by making use of the several DER spread throughout the network. Control schemes at primary substation level, secondary substation level, and LV level were developed in this functionality. However, the demonstration focused only on the voltage control of an LV network.
SF4 requires resource controllability, and, as of today, there are no resources for the DSO to be able to control either in the MV/LV secondary substation and at the LV level. Also, the inverters that are usually used for micro-generator installations cannot be controlled, due to technical and regulatory reasons. Therefore, it became necessary to install resources that could in fact be controlled. In order to obtain physical space for this purpose, an official protocol was established between EDP DISTR and local government authorities that allowed the use of a government building that is used only for material storage and sporadic local events. Within these installations, the following resources were installed in order to ensure that, in total, there are 4 controllable resources:
• 30 250 WP solar panels, corresponding to an installed capacity of 7.5 kW (two sets of 3.75 kW);
• 64 12V 12Ah batteries, corresponding to a nominal capacity of 4.6 kWh (two sets of 4.6 kWh).
Following the SuSTAINABLE architecture, each DER contains a prototype inverter that is controlled through a smart meter. The smart-meters communicate with the DTC installed at the secondary substation, in which the voltage control algorithm is embedded. The smart meters installed at the SuSTAINABLE dedicated building have GPRS communication with the DTC, and so do the smart meters installed in the LV customers. The smart meters in the SuSTAINABLE facilities are connected to the inverters via a local HAN interface. This interface is present in all of the smart meters installed and allows the smart meter to serve as a gateway to the customer’s house.
Two different prototype inverters have been developed to be used in the demonstration of the SF4 functionality:
• Smart Solar Inverter for interfacing PV generation;
• Smart Battery Charger for interfacing battery storage units.
To interact with the LV controllable resources (batteries and solar inverters), the user must access to the DTC HMI web interface. A set of modules are available for visualization and control the smart meter data.

5.5) SF5 – Technical virtual power plant
The goal of this SF was to evaluate the TVPP concept. Currently, it does not exist, within the Portuguese legislation, regulatory scope to send customers set-points for load reduction or increase, except in very specific situations. For this reason, SF5 demonstration consisted of a methodology that comprised both customer surveys and simulation. For the scope of these simulations, flexible demand was the only DER considered for TVPP control.

6) Regulatory Analysis
The deployment of smart distribution grid solutions is based on two mayor pillars. On the one hand, power distribution is deemed to be a natural monopoly. Consequently, the remuneration and performance of DSOs are subject to some form of regulatory supervision. This implies that regulation will greatly influence the expenditure decisions made by these companies, both in terms of their amount and the type of expenditures incurred (e.g. CAPEX vs. OPEX or copper-and-iron vs. ICT-based solutions). On the other hand, distribution network users conventionally comprised exclusively passive consumers with very little, if any, interaction with the DSO. However, smart distribution grids require a more active contribution from distribution network users so as to reap the benefits of enhanced flexibility. The regulatory analysis performed in the SUSTAINABLE project is based on 3 steps: (i) mapping functionalities and regulatory topics, (ii) reviewed of current regulatory framework in the focus countries, and (iii) barriers and recommendations.
The functionalities and use cases analysed within the SUSTAINABLE project are quite diverse with respect to their goals, technologies involved and stakeholders affected. Consequently, not all of them will be equally affected by the different regulatory topics discussed in the previous section. Thus, in order to present more focused discussions about the impact of regulation of the future deployment of these smart grid solutions, they were mapped against a list of relevant regulatory topics. Moreover, this allowed identifying what regulatory barriers and recommendations are relevant for each of the smart grid solutions evaluated.
It can be observed that some regulatory issues can be considered as cross-cutting and affect all the functionalities. Among these, one may find the ad-hoc incentives for innovation, mainly related to financial risk mitigation to promote demonstration activities, as well as general revenue regulation approach, i.e. how DSO allowed costs are determined and regulated. The former would allow DSOs to test new solutions for the challenges they are facing or expect to face in the near future, whereas the latter would determine the future replication and deployment beyond pilot projects.
A second step for regulatory analysis was the review of the current regulatory conditions in the SUSTAINABLE focus countries. Here major regulatory topics affecting the DSO activities, and those related with connected network users (consumers and generation) were assessed.
Lastly, barriers and bottlenecks for the deployment of the previous smart grid solutions have been identified based on current regulations. Also, regulatory recommendations aiming to overcome these barriers have been proposed.

7) Scalability and Replicability
The scalability and replicability of the SuSTAINABLE functionalities were also analysed. Scalability is the ability of a system to increase its size and replicability is the ability of a system to be duplicated in another location. The tools were developed and validated at different sites with different technical requirements, as well as regulatory and social conditions. Therefore, both scalability and replicability are very important for the large-scale deployment of the SuSTAINABLE concept.
Questionnaires were firstly distributed to the SuSTAINABLE partners, which developed the different tools. The purpose of these questionnaires was to identify the barriers to scale-up and replicate each tool. Some of the most important barriers identified were:
• Limitations of current regulation
• Communication requirements
• Willingness of participants
• Automation limitations
• Lack of commonly accepted practices and techniques
• Cost of hardware (e.g. storage) and measuring devices and their installation.
As a next step, the implementation conditions in four diverse European regions were analysed. These regions were Germany, UK, Greece and Portugal. The social, geographical and regulatory conditions for each country were studied, as well as the stakeholders’ viewpoints and engagement.
For each of these regions and their characteristics, the timescales for the mitigation of the barriers was defined. The high priority functionalities were then identified, as well as the ones that are important in the medium and long term. The results of this analysis were the four SuSTAINABLE deployment roadmaps.

Potential Impact:
Potential Impact and Main Dissemination Activities and Exploitation Results
The evaluation of impact of the SuSTAINABLE concept and an adequate dissemination of the main achievements obtained was a constant concern throughout the project. In this context, the main goal was always to disseminate the SuSTAINABLE project in a consistent and ample way, so that all interested parties, stakeholders and decision makers are informed at all times about progress, findings and results. It was considered that dissemination is key to ensure maximum usefulness of SuSTAINABLE concept achievements, throughout other regions and projects across Europe and a vehicle to ensure a smoother path for the implementation of the results of the project. Moreover, the potential impact of the solutions developed has been thoroughly assessed at different levels. In the next sections, the potential impacts and main dissemination activities are described.
The adoption of the SUSTAINABLE concept will not only affect the distribution network business but will also impact the overall energy system and its participants, particularly final consumers and distributed energy sources promoters. In the project, its potential impact has been evaluated in three different stages: KPI calculation during the implementation of the designed functionalities, cost and benefit assessment for the different cases and countries based on the JRC methodology, and the boundaries set by regulation and scalability and replicability potential.

1) Overall system impact
• Reduced electricity system costs, as all SUSTAINABLE functionalities derive in a more efficient use of the current assets (electricity generation and networks) and helps the integration of new renewable generation.
• Improved market and system performance: the use of forecasting tools has a direct impact on a system level with a reduction in the power system balancing needs and costs. It has been assessed that a reduction of 38% in the forecasting error implies a reduction of 34% of balancing cost. This is directly translated into a reduction the final electricity price for all electricity consumers. A second benefit is the reduction of the price volatility resulting in systems with high penetration of intermittent generation. Therefore, the reduction of balancing needs will also allow the integration of higher penetration levels of renewable intermittent generation.
• Introducing automation and smartness in the operation of the distribution networks also increasing the hosting capacity of new distributed generation. Within the SUSTAINABLE concept, all electricity consumers and distributed renewable would be the beneficiaries.
• Aggregating the effects indicated previously, there would be a clear reduction of pollutant emissions. The combined effect of higher renewable generation levels and lower energy use and a modified demand profile, mainly in peak periods, will results in a reduction of conventional thermal generation, and therefore lower emissions. If the social cost of the emissions is internalized, these benefits will be seen directly by all energy system participants. The improvement of RES forecasting under the SUSTAINABLE concept reduces by 1% emissions.
• Improving standardization of smart grid technologies, based on the implementation of the SUSTAINABLE pilots. Therefore, there will be a reduction in time and costs of implementing in future these functionalities.
• Rules for scale and replicate the testing functionalities in a selected group of countries will easy its implementation, and to understand the basic boundary conditions.
• The software and hardware technologies developed in the SUSTAINABLE project will certainly increase EU export potential in High-tech electricity industry, and therefore increase of employment in this sector.
• The project’s regulatory recommendations will allow for a swifter and more efficient deployment of smart grid functionalities across Europe, thus unlocking the aforementioned technical potential.
• Testing the functionalities in a real environment helps to identify which functionalities certainly provide benefits and which are the technical challenges for those that did not succeed.

2) Distribution network operator
• Reduced energy losses as a result of an optimal voltage control strategy.
• Reduced maintenance and operational costs, achieved using a better observability/predictability that enables the network to anticipate failure events and plan maintenance programs.
• Improved quality of supply using higher observability and control of the distribution network. There is a reduction in the time needed to solve power interruptions, as the time to locate the faults is reduced thanks to real-time and distributed metering, and also time reduction in service restoration of healthy network thanks to remote control and identification of the failed part (if a maintenance crew is required).
• Helps identify the optimal level of automation, based on the failure rates, network structure, etc. In particular, a 7% automation degree is identified as optimal in Évora, while in Rhodes the optimum is 3%.
• Reduced network investments. The increased observability/predictability of the main distribution network parameters, an active voltage control and a flexibility-based network planning methodology gives the distribution network operator inputs for a more efficient and safe operation. Therefore, active grid operation strategies will enable a more efficient use of the current installations, being able to defer or avoid costly network reinforcements. Furthermore, by exploiting alternatives such as load shifting using a VPP, it was shown that the network upgrade can be deferred.
• Help design an efficient communications network. Definition of the bandwidth and latency for sending control signal to distributed resources in order to implement voltage control.
• Provision of services to DSO via VPP, such as congestion management and voltage control.

3) Distributed energy resources impact
• Increased network hosting capacity, by a more efficient real time operation of the network. This would result in lower costs of grid connection and avoiding delays in the grid connection process.
• Reduced RES power curtailment due to network constraints, by using advanced voltage control strategies and support in congestion management using a VPP.
• Additional revenue streams through the provision of services to the DSO via VPP. Market integration of distributed energy resources by utilizing aggregation through a VPP.

4) Electricity consumer
• Lower electricity cost as compared to a business as usual scenario where DER are integrated through a fit and forget approach.
• Increasing the quality of electricity supply, both continuity of supply and power quality, and therefore to reduce the costs associated to poor quality or supply interruptions. This will have a direct impact on the production cost in the industry, competitiveness of the services sector, and quality for domestic consumers.

List of Websites:
info@sustainableproject.eu

Final Report Summary - SUSTAINABLE (Smart Distribution System OperaTion for MAximizing the INtegration of RenewABLE Generation)

Related documents

Share this page

Download