Inverter and PV System Technology
Industry Guide 2013

The portion of solar power that can neither be absorbed by the grid nor used directly on site needs to be temporarily stored. Batteries are the primary contenders for this task, as they have proven their worth over decades of use and can be employed in decentralized systems. Owing to topographical restrictions in Germany, cross-regional storage in reservoirs (pumped storage hydroelectric power stations) is only possible to a very limited degree. Compressed air energy storage represents one alternative technology that, in principle, holds great potential, but its efficiency is still in need of improvement.

Converting solar power into chemical energy (e.g. by means of electrochemical hydrogen generation) incurs relatively high losses, but does bring with it the advantage that energy can be stored for long periods of time. Solar hydrogen can be converted into either electricity or heat and can also be used as fuel, directly substituting petroleum products. Converting hydrogen into methane would achieve an even higher energy density and thus tap into an even greater storage capacity. In this case, the efficiency of converting the energy does drop somewhat, but this would still be acceptable given that free sunlight is the energy’s source.

The latest developments in technology do not allow us to foresee which storage systems will triumph in the long term. The most likely scenario will involve a mixture of small, distributed, short-term storage systems and large, seasonal storage systems. Ultimately, if the intention is to make photovoltaics a mainstay of German power supply, a solution must be found to store the surpluses generated in summer for use during winter.

The only type of instant storage currently available is that of secondary electrochemical cells, generally known as (rechargeable) batteries. However, the unavoidable phenomenon of self-discharge in batteries means that they are only suited to storing solar power for short (from a few hours to a few days) and medium (a few weeks) periods.

Moreover, the lifespan of a battery is limited by its cycle life, not forgetting that the number of possible charge cycles falls as the depth of discharge increases. The battery therefore needs to be protected against over-discharge. In lead-acid storage batteries, for example, full discharge converts the lead sulfate into a crystalline form which is only partly dissolved when the battery is charged again, causing permanent damage.

What is more, the capacity that can be extracted from an accumulator decreases as the discharge current becomes more powerful.

Lead-acid storage batteries are cheapest and are therefore most frequently used. They are filled with an electrolyte of dilute sulfuric acid, meaning that if the final charge voltage is exceeded, gassing may occur. When this happens, oxygen forms on the positive electrode and hydrogen on the negative. These two gases then form explosive oxyhydrogen. Gassing also leads to the gradual loss of water, which needs to be regularly refilled. Overall, the cycle life of lead-acid storage batteries is relatively low.

In order to increase its lifespan, the electrolyte can be thickened using additives to form a gel. Lead-acid gel batteries can be assembled fully sealed, meaning that they are leak proof. In this case no gas is able to escape, but lead-acid gel batteries may dry out as a result of gassing. A special charge controller is therefore necessary to manage the final charge voltage very precisely. Lead-acid gel batteries have double the lifespan of lead-acid storage batteries with liquid electrolytes. They allow around 2,000 cycles, provided no more than 30% of the capacity is discharged each time. If 50% of the capacity is drawn on a regular basis, lead-acid gel batteries will need to be changed after around just 1,000 cycles.

Lithium-ion batteries achieve markedly higher cycle lives. If discharged and recharged daily, they can reach a lifespan of 20 years, equating to 7,000 charge cycles. Their special features include high energy densities and low self-discharge rates. They also withstand high charging currents, and can therefore be charged very quickly. These advantages currently make them ideal storage batteries for homes and electric cars. Prices for such batteries are still high, however, and will not fall until mass production levels are achieved.

Redox flow batteries

Both types of storage battery (lead-acid and lithium-ion) share the common feature that their electrodes undergo chemical conversion during charging and discharging, and therefore slowly degenerate. Redox flow batteries avoid this. A relatively new development, these batteries combine the properties of the accumulator with those of the fuel cell.

The reactants are each dissolved in an electrolyte and circulate separately. These two electrolytes are pumped through a cell in which ions are exchanged. This cell is divided by a membrane that only allows ions to pass through it, thus preventing the reactants becoming mixed.

The electrolytes that store energy in redox flow batteries are kept in separate tanks. As a result, the quantity of energy and the output can be scaled independently of one another. Redox flow batteries are characterized by their high efficiency and long life expectancy.

The capacity of the redox flow storage systems that are shortly due to be launched on the market lies between 3 and 13 kWh. Apartment buildings and commercial establishments require larger units, providing an opening for those redox flow batteries that are currently offered in 200 kWh modules. Here, additional modules can be added to increase the capacity.

As both lithium-ion batteries and redox flow batteries are still at an early stage of development and are relatively expensive, the lead-acid battery is still the most economical way to store solar power, despite its short cycle life.

Storage systems

Overall, storing solar power in batteries is a relatively expensive enterprise. It currently costs roughly as much to store a kilowatt hour of electricity as it does to generate it from sunlight. The specific costs of storage (in euro cents/kWh) are not the sole criterion, however. If the goal is to operate a battery system as profitably as possible, cycle life, the output of the PV plant and household energy requirements must also be considered.

In order to incorporate batteries into a PV system, special storage systems are required which consolidate the storage battery with the necessary power electronics. These have only recently become available on the market. They not only differ according to battery type, but also based on how they are installed. Some systems are incorporated into the house’s AC circuit, while others are integrated into the PV plant’s DC circuit.

Integrating the system into the AC circuit has the advantage that as much additional capacity as desired can be added at a later date, irrespective of the PV capacity installed. A battery inverter is needed in addition to the PV inverter, meaning that relatively high outlay is required, but such systems come with the extra advantage that power from the grid can be fed into them more easily, as the battery inverter operates bidirectionally.

Incorporation into the DC circuit also has two advantages: the system costs are lower and the storage efficiency is higher. This method requires the installation of a PV inverter and a pair of DC/DC con verters. They set the voltages of the PV system and the battery at precisely the level that is best for the inverter. To simplify matters, they can be installed in the metal cabinet that houses the battery.

Despite the high investment costs involved, battery capacity should be selected to enable as much solar power as possible to be consumed on site. By way of example, for a 4 kW plant and annual energy consumption of 4,000 kWh, a capacity of 6 to 7 kWh is recommended if the quota of on-site consumption is intended to reach 70%. A quota of over 30% will be virtually impossible to achieve without using storage, unless the solar power is also used to heat water. Combinations of photovoltaics, heat pumps for water heating and intelligent energy management, for instance, can achieve on-site solar power consumption rates of up to 50%, even without storage systems. Of course, checks should always be made to examine whether or not solar thermal installations will represent the most economical solution for the actual needs and conditions of the site.

Maximizing on-site consumption without considering the consequences must, however, be avoided. Care must always be taken to ensure that consumption only increases in order to raise the quota of on-site consumption and that power is not used arbitrarily and unnecessarily, as this would be counterproductive in terms of energy efficiency. On the other hand, replacing fuel, by, for example, using energy-saving electric vehicles on a large scale, would be a highly welcome development.

One very simple way of storing energy is to store heat. As buffer storage is already available in some houses in the form of hot water tanks, surplus solar power could also be converted into heat by conducting it through an immersion heater inserted into the storage tank. As long as the production of solar power is significantly more expensive than producing hot water, this will remain a very wasteful use of energy. Nevertheless, falling solar energy generation costs combined with rising prices for raw materials will slowly close this gap.

A far more efficient application of surplus solar energy is in a heat pump. If this pump is capable of generating 3 kWh of heat from 1 kWh of electricity, 1,600 kWh should theoretically be sufficient to heat a well-insulated house with a living area of 120 m² and heating requirements of 40 kWh/m² . This calculation is unrealistic, however, as supply and demand do not coincide: In winter, when the greatest demand is placed on the heat pump, the PV plant will furnish the least electricity.

The conditions are somewhat different if the heat pump is used to provide cooling via an air-conditioning unit. In this case, the periods of energy production and consumption do correlate if power from the photovoltaic plant supplies electricity for heat pump cooling during the summer months. In the USA, for example, heat pump cooling is already widespread.

Irrespective of their intended use, heat pumps are not suitable for storing energy over the long term, but merely represent a component of good energy management.

Energy management

It is considerably easier to generate solar power than to store it, as this entails relatively complex installation procedures and unavoidable losses. In order to complement the storage options, as much energy as possible must therefore be consumed on site and the conditions for marketing the power must be made as favorable as possible. This situation will then replace the current practice of unreservedly feeding power straight into the grid. If the statutory feed-in tariffs were to be abolished, or sink so low that it became unviable to feed all the solar power generated into the grid, this custom would die away. Grid feed-in will then only be sensible under certain circumstances and should only be considered if other options are not available.

Load shifting can help to increase on-site consumption rates. For example, large household devices that do not require power at a given time might only be switched on when solar power is in plentiful supply. Washing machines, tumble driers and freezers can thus contribute to improving the coordination between demand for power and supply. These energy management systems need to succeed in changing the consumption patterns of the average consumer, i.e. to drive them away from simply using household power at any time at the push of a button. They must clearly indicate the costs per kilowatt for each device at a given moment and ideally offer alternative operating times within the shortest possible time frame.

Favorable sales conditions will become possible if the tariff for purchasing power from the grid increases while the solar power produced on house roofs becomes cheaper at the same time. This power can then be sold to neighbors in close distance at a price below the electricity tariff, as the short distances will mean that grid fees are waived.

Only once these two channels have been exhausted should solar power be used in hot water tanks or heat pumps. Using solar power for heating means using solar energy under its value. This is why it ranks third in the hierarchy of solar power exploitation.

Given the fact that storing power in batteries will remain relatively expensive for the foreseeable future, it should be avoided if at all possible. If several other possibilities for use are in place, only a small battery capacity will be required.

As a last resort, feeding power into the grid remains an option. This would struggle to contribute to PV plant profitability, however, as solar surpluses are produced by many PV installations simultaneously.

Increasing on-site consumption through load shifting and solar heating, selling power, storing it, and feeding it into the grid are the options available to the energy management systems of the future. If possible, these systems should be able to independently decide on which type of use would be most beneficial to PV plant profitability at which times, and to activate consumption devices and storage systems as required. This concept demands a great deal of the systems technology, which will inevitably change over time. Gradually inverters will be replaced by energy management systems. Developing these is the task that now lies ahead.

Power storage has always been a key requirement to achieving self-sufficient energy supply in locations situated away from the grid. Short-term power fluctuations, caused for example by clouds passing overhead, have particularly negative effects on these stand-alone systems and appropriate storage capacity must be made available to counterbalance this. For instance, batteries capable of storing 250 kW are required to compensate for unexpected power fluctuations in solar plants with outputs of 1 MW. The batteries used must also be capable of discharging quickly. Lithium-ion batteries, and in particular those with medium energy densities, have been found to be highly suitable for this.

The solar power supply of stand-alone systems therefore makes additional demands on energy management. Load shifting plays a vital role in this area.

Outlays for storage are determined by the system’s investment costs and lifespan. When calculating these, focus is placed on the capacity-related costs (euros/kWh) as opposed to the performance-related costs (euros/kW), with the efficiency of the battery system also playing a role.

The lifespan of batteries used to store solar power should ideally be at least as long as that of PV plants, i.e. 20 years. The desired lifespan of the first mass produced batteries is ten years. Traction batteries, which are currently being developed at breakneck speed for the automotive industry, have much shorter lifespans of between five and eight years, making them less suitable for storing solar power. What’s more, they remain too expensive. A lithium battery currently costs around 400 euros/kWh. Although this amount could fall by half over the next five years, lithium batteries would still remain more expensive than their lead counterparts. A way of overcoming this problem would be to combine large lead batteries with small lithium-ion batteries. To protect lead batteries from peaks in demand, which could result in damaging over-discharge, lithium-ion batteries should be employed during peaks in demand, while lead batteries should be used to cover base load.

As the cost of storage falls with increasing lifespans, a battery’s cycle life is of great significance. Taking into consideration the levels of insolation normally seen in Germany, batteries need to be charged and discharged around 3,000 to 4,000 times in the space of 20 years. Lithium-ion batteries have already achieved a cycle life of this extent. Lead batteries, on the other hand, are only capable of performing around 1,000 cycles, outweighing the advantage of their lower investment costs – unless, of course, they are combined with lithium batteries (see above).

Overall, the drop in prices brought about by the technical development of storage media and the economies of scale resulting from mass production, as well as the coordinated interplay between the various forms of storage, shall contribute to ensuring that the share of photovoltaics in the power supply continues to grow.