In the electric power system the balance between generation and consumption must be exact at all times. The integration of a high concentration of fluctuating renewable energy sources such as wind and solar therefore plays an important role. In order to ensure proper integration of DER, there are two methods currently in use: first, price signals on the wholesale market  to provide rough balancing and second, the control power at transmission system and control area level respectively is procured uniformly through control power markets . This kind of balancing based on market balancing mechanisms throughout the whole marketplace or within a particular zone creates pricing incentives, promotes good frequency control and manages the operational requirements in transmission networks, but not in distribution networks. In distribution networks the problems are local. In urban distribution networks the primary need is monitoring of network asset utilization, and in rural networks it is maintaining the permissible voltage bands  at each network node. These demands require different operation and control concepts which cannot exclusively be optimized by using just price signals. Since capacity utilization and voltage band management are strongly driven by local network and load conditions, regional differences must be taken into account. While in one section of the network there might be enough reserves available, another section might be about to reach the allowable limit, thus a reaction to price signals may effects security of supply at distribution level. In these critical areas of the network it is therefore necessary to strive for a reaction time of 30 seconds to one minute. The DG Demo Net Validation, ZUQDE and DG Demo Net Smart Low Voltage Grid projects as well as V2G strategies offer ways to manage voltage bands and provide reactive power control of DER in low- and medium voltage networks  in different applications. The network control based on price signals will be enhanced with regional and timely differentiated approaches for voltage band and asset utilization management. Regulating reactive power locally and using on-load tap changing transformer at individual substations significantly improves voltage band management and utilization of existing grid structure (see Figure 1).
Figure 1: Improved voltage band utilization through new network control
Voltage control in MV networks
Pilot projects up to now have produced a method  whose implementation in medium voltage networks works well on a technological, operational and economic level. In the 30 kV Grid in Lungau, the ZUQDE system showed that an additional increase in generating capacity of around 20 % in critical network sections / branches is realistic. Based on the assumption that this amount of energy can replace direct electricity generation by fossil fuel power plants, around 24,300 MWh of thermal energy and therefore approx. 2,200,000 liters of crude oil as a primary source of energy could be replaced. Not burning this much crude oil would lower CO2 emissions by approx. 5,000 tons per year.
The more "flexibility”, which means controllable generation units as well as flexible loads or consumers available for network operation, and the more evenly distributed these are along the distribution lines in a network section, the better the voltage and reactive power in a medium voltage network can be controlled. Because of these findings, the goal should be to maximize the distributed generation in the existing electricity network and to integrate them into future control approaches.
Of course, the limits of physics will determine the boundaries to integrating decentralized generators. Once the maximum transmission capacity in the grid has been reached, investments to strengthen or to replace existing units will be unavoidable.
With regard to cost effectiveness, in particular to the question of whether this solution is cheaper than expanding the grid, it must be noted that these calculations are based on individual cases on which the following factors depend on:
- Investments in reinforcing the grid, which cannot generally be avoided, but rather only deferred, and
- duration, which depends on the existing structure of the grid, the structure of consumers and generators and especially their development over time.
One example of the economic advantage that can result is the Turrach power plant. Without the power plant’s contribution to voltage control, it would have been necessary to lay a 14- km-long cable  to the next suitable connection point. Using the new voltage control concept (ZUQDE or DG Demo Net) reduced the length of the cable to 50 m, thereby significantly reducing the cost to the power plant. In total, the power plant saved 1.67m euros against connection and adaptation costs of between 30,000 and 50,000 euros .
This type of solution is a benefit for new generators connecting to the grid since the connection costs can be significantly reduced. The better maintenance of voltage bands and the improved security of supply, the increased share of renewable energy and the aforementioned reduction in emissions that result from this solution constitute an intangible advantage for society as a whole.
The control concepts for medium voltage networks described here can in principle also be used in low voltage networks, as field tested in the flagship project DG Demo Net Smart Low Voltage Grid in Köstendorf. However, this method is much more complex in low voltage networks because of the greater number of electricity producers and consumers, which are also often prosumers, and the additional technical challenges (see Figure 2).
Figure 2: Schematic illustration of the control concept in the field test in the low voltage network in Köstendorf 
The challenge lies in answering the question of how specific or complex the control concepts need to be in order to ensure voltage band management in unbalanced 4-wire low voltage networks with strongly varying loads. Simulation based investigations have shown that the number of metering points necessary to characterize a low voltage network is extremely high in comparison to a medium voltage networks. Within only a few seconds, the voltage in the grid can change significantly. Since load conditions are complex to measure , critical nodes or feeders are difficult to identify. They also vary from one moment to the next. This is why central or aggregate measurements such as the total current of feeders or low voltage network sections deliver almost no information about conditions in the network. But the problem of maintaining voltage also becomes more severe under unbalanced conditions. The effectiveness of the voltage control concepts implemented in the PV inverters depends on the conditions in the network. The individual control of every single unit  affords the greatest flexibility, but it is very expensive. On the one hand, it would require extremely complex planning and on the other, each individual controller would need to be configured (either on site or remotely). The alternative would be to connect with network users in bundled groups using broadcast commands. It is also always necessary to place an aggregator or flexibility operator  between the units and the network to check the conditions on the network and act on this basis.
With higher-level communication, the effectiveness of these regulation mechanisms can be improved. As a consequence of the challenges in the planning of electricity networks with a high number of generators with smart-grid functions, the economic evaluation of different approaches is very complex. In general, the objective should be to use automation and the opportunities offered by the market to achieve a maximum level of energy efficiency from residential units as well as the highest load flexibility potential.
Aspects across different voltage levels
On the whole, there are promising solutions both in low- and medium voltage networks. The biggest difference is in the different stages of research. In the medium voltage network, the solutions which have been tested are practicable. For the further development of prototype products and solutions, there are concrete agreements with delivery companies. In workshops with grid operators in Austria and Germany, results and solutions have been presented and discussed. There is real interest in seeing them implemented.
In low voltage networks, the first prototypes are currently being set up and will be tested until early 2014. For this reason, results are not as valid as in medium voltage networks. It can however be assumed that the solutions tested in the low voltage network will be similarly promising and will be able to be implemented as a purely network drive controller in the near future. Due to the numerous applications for residential customers, it will be necessary to bring the control of the power grid into line with the electricity market system. The optimization of consumption in residential units, for example, has to be connected with generation from photovoltaic units and charging strategies for electric vehicles. In addition, everything has to be harmonized with the offers on the electricity market and correspond to the vicissitudes of time and region that will occur. For this reason, it is a greater necessity that the open questions at the system level be answered in order to design products for low voltage networks. Only after these questions are answered can the technological solutions, the devices, the communication and the data exchange be devised in a way that the products address both the needs of residential customers and the demands of the market.
In the flagship projects, realistic assumptions were made in order to be able to put the right technology in place.
It should also be noted for all voltage levels that the solutions were implemented in typical applications used by a rural distribution network operator in order to be able to apply the findings to other distribution networks. This will, however, not rolled out all at once but rather implemented network area for network area based on actual situations and needs. The determination whether to use intelligent network control concepts or to reinforce the grid is influenced by the following criteria and conditions:
- Do investments in existing lines have to be made anyway due to age or condition?
- How should the cost of network control concepts be allocated in the future? Although the generators reap the largest profit, especially in medium voltage networks, the operator will continue to shoulder the investment and operating costs of the new control systems, as will its consumers down the line in the form of system usage fees (network tariff). This question is important especially from a macroeconomic perspective, particularly since too high of an economic burden on generators could undermine targets for renewable energy use.
- Which power plants or consumers should be incorporated in this scheme? For new generators coming on to the grid, active medium voltage network control is an attractive alternative to high connection costs and can easily be integrated in new network connection contracts. Existing units, which may not be suited for new control capabilities, would however cost a lot to retrofit without creating a direct advantage. They could only be integrated into this scheme on a voluntary basis, through financial incentives or by being required to do so by law. An additional question is whether applying control exclusively to new power plants coming onto the grid constitutes unequal treatment  among network users.
- Does equal treatment mean that the grid operator always has to call upon a different participant or can it work with participants who are able to solve the technical problem the most effectively? Due to the electricity network and generating structure, some would be called upon more than others.
In essence, all of the different variations of market rules at distribution level are possible. It is important for overall grid operation that rules are set to answer these questions so that distribution network operators can institute these operating solutions within the appropriate legal framework.
Independent of the voltage network, the question arises how to best integrate ICT to use the synergies with other smart-grid applications as well as the technical and organizational connection of the individual functions on the market.
An integrated view of the aforementioned methodologies for medium- and low voltage networks will be taken in the INTEGRA project.
 This includes day-ahead and intraday trading
 Among the control markets are primary, secondary and tertiary control. Further information is avail- able at http://www.apg.at/en/market/balancing
 In order for machinery and appliances that are connected to the grid to be able to operate trouble free, grid operators have to maintain particular voltage characteristics at all connection locations (according to EN 50160: Voltage characteristics of electricity supplied by public distribution systems). One example of the range of variation of the r.m.s. magnitude of the supply voltage in Un ±10 %, which corresponds to a value of between 207 V and 253 V. at low voltage level
 The ZUQDE system can be expanded and used in a high-voltage network (110 kV) as well
 Further information is available in the ZUQDE project’s final report at www.smartgridssalzburg.at/downloads
 This cable length already requires local voltage dependent reactive power and active power controls in power plants. Without these an even longer connecting cable would be necessary.
 This calculation is based on the 2011 price for cabling across different areas (fields, residential are- as, streets, etc. and without offsets from the grid operator for implementing the ZUQDE system.
 Source: Andreas Abart, Energie AG OÖ Netz GmbH
 The ISOLVES project focuses on monitoring the condition of the grid using power snapshots taken by smart meters
 Among the controllable units in low-voltage networks are generation units such as photovoltaic or charging stations for electric vehicles, heat pumps, electric heating systems, chillers as well as residential units with home automation
 This term was defined by the Smart Grid Coordination Group of the M/490 standardization mandate as follows: The flexibility operator is a general role that pools small flexibilities of customers / network users in order to make use of them in the grid or on energy markets. The concept is often referred to as aggregator, but in this case the name should underline the general role concept of “Using flexibility”. According to the description of the role concept the roles of the flexibility operator might be performed by existing market roles like energy suppliers, aggregators, DSOs, etc.
 The equal treatment of all grid users is enshrined in the Electricity Act [ElWOG (Elektrizitätswirtschafts- und -organisationsgesetz) and provides an important legal foundation for grid operators (see for example ElWOG 2010, BGBl. I No. 110/2010, §5 and 9).