1. Why should you prepare to a volatile energy market?
Don't let your clients down.
How energy prices change in the Netherlands on May 9, 2021 from a negative to a positive extreme
Store energy as electricity or directly in the form that serves you - with less investment.
Last week I published my video Blog about the power grid in view of the increasing volatility of the demand and also the reduction of energy. It is a known fact that the demand has always been volatile. The big power stations had to adapt their output to the demand only, now they have to adapt also to the most volatile renewable sources, which are sun and wind.
But the society wants these sources (including me) and regulation therefore prescribes, every year more, that renewables have priority on the grid. The result is that conventional power plants have more difficulty to compete and to pay themselves back. This will result in more renewables to fill the demand gap or in higher prices, or both. In a blog on May 27 (see Link Tree), I showed already that German electricity prices could go up to 50% in 2030. This price increase was only on the energy part of the whole kWh price, where many other costs (transport, taxes, …) are included, so in reality this 50% is only 0.02 €/kWh. However, in Energyland it is a real shake-up, because for an energy plant, this energy part is the whole revenue source.
And for you, medium or large energy consumer? Do you think to be outside this development? Let me explain it with another commodity market, say fish, and let’s assume you cannot put it in the freezer (more or less the same situation as with electricity). In such a situation, you not only have a big fluctuation in price as the fisher comes in the harbour once in a while with large quantities, but also people that really need fish on a constant supply will be in big problems.
The same is true for you. You cannot do your work without this commodity (electricity) and it is going to be scarce. In other words, it is going to be more expensive during peak times and maybe even not available whenever you want. Do you think that the energy supplier will always be able to deliver on demand, if it, too, is under pressure and has to work with old infrastructure, without the money to modernise it?
Therefore I want to suggest you two ways to come out of these problems and at the same time help the society. Don’t underestimate the convenience of thinking about the interests of the big community. Who does that will go with the stream and improve the own situation, on the long run. These two ways are:
Try to have access to the dynamic energy tariffs. It is in the interest of the energy supplier to make that easy for you and incentivise you to buy on the right moment, like the fisherman has the interest that many of the clients, which come when his market stand is empty, wait instead and come when he has it full of fish. It will avoid him selling fish against dump prices when it is abundant. If the energy supplier doesn’t provide this access to you, you will find so called aggregators, which accumulate customers in order to have enough market power to talk with the grid operator. If you find this access to the electricity wholesale market, then you can enjoy much more volatile prices and get the energy at really low prices. The flexibility that the aggregator offers can come in different flavours. I give here two different examples.
You get access to a kind of spot market, where you actually pay the wholesale market price for energy, which varies each 15 minutes. This allows you to pick the best from higher extremes (rather than the “dull” two-tariff structure);
You promise to lower your power consumption on command of the grid operator with N kW, where you can put your own number for N, but usually more or much more than 100. The grid will pay you for example 50€/kW per year if you give this availability during 6 hours during each working day. You get this money for the promise alone! It is like an insurance premium. If they actually ask you one or more days to lower your demand for a given number of hours, then we are talking about power x time = energy and you can get up to 0.40 € per kWh, in addition to the mentioned kind of insurance premium. This is very similar to the principle according how you pay the supplier: a contribution for the consumed energy and another for the claimed or reserved power.
Look how you can be flexible in your energy consumption so that you can take advantage of these low, and avoid the high prices. Many entrepreneurs say “No” to this idea after half a second. “We need the energy when we need it, period”. If you think a few moments, it is not impossible to be flexible and it opens even the way to unthought advantages. For example, store energy in your enterprise and you will survive a black out of several hours, while other enterprises have a major setback. Thinking again about fish, imagine that your freezing rooms have a greater capacity to keep the temperature constant during a longer time with a blackout than average food stores. Or think about a production stop that will cause big problems and even damage. Or think about the complaints in a hotel, on a hot summer day with a black out. Charging batteries is only one way to come through a blackout and maybe the most expensive. In the example of the freezing room, you immediately see that it is an alternative way to “store” electrical energy, because it enables you to decide when you put electricity into it, as long as the temperature is between a min and max value. The freezing room itself is a kind of energy storage. In the case of air conditioning, you might need a real and additional storage of ice or cold water to store the cold energy, because the space will heat up quickly when no active cooling is present. Another way to “store” energy, creating flexibility in the energy demand is to make more of some kind of intermediate product during low prices of energy. Especially products that require much energy and can be made in an automatic way, can be considered an energy storage. Fully automatic machines can make them even during the night, with only a few or no operators.
A real energy storage is also a way to increase the power that you have available. I remember a factory of Thyssen Krupp (Switzerland), which had problems with cooling a rather hot place of 3’000 square meters. The Energy Manager adopted an ice bank of 1’400 kWh, charging it during the night and adding its cooling power to that of the chiller during the day. How expensive would it have been to increase the transformer, the supply contract and adding a chiller to keep the place cool? The money you save with the flexible tariffs can be invested into a form of energy storage and you get Security of Supply as a bonus.
So the answer on the question in the title above is that yes, you should prepare to volatility and why? To save money and more importantly to decrease the risk of a black out, or even that the existing power is not enough because the summer is particularly hot, the winter extraordinarily cold or you just have got a lot of clients all at the same time.
In the future, the circumstances will make this preparation more important.
2. Energy consultancy on a distance - Applying the data
In an earlier Newsletter (May 5 of this year) I described how to do the analysis of the energy consumption in the enterprise. I told about the data that can be given by the client to the energy consultant, like:
The energy bills;
The contract with the energy supplier or at least the tariff structure:
A list of big energy consumers, like lighting systems, air compressors, production lines, etc;
During which times these are operating;
What has been done in the past with Energy Management;
Which Government programs and financial support is available;
Very important is also other information about problems that are going on in the enterprise, and which might be related to energy questions. A few examples are:
Safety in an area that is not well illuminated;
Problems with productivity because of a low comfort;
Other (technical) problems with low productivity;
A continuing increase of legislative pressure related with your environmental footprint.
Then we saw how the Distant Energy Manager (DEM) might take time measures from the most consuming parts of the enterprise, so that it is known, where most of the energy goes and when. The DEM can help you to discover also the why. This Time and Space aspect is what lacks in many enterprises. It can be difficult to make these measurements and decide where. We saw in the previous newsletter that these measurements can be made by the client under guidance of the DEM, with mobile measurement devices that you may hire and which might even be displaced between subsequent measurement campaigns so that more data points become available. Taking measurements in different periods may eliminate the advantage of synchronous measurements, like with the cardiologist’s ECG, where it is important that all 12 heart signals are taken on the same timeline.
Another way is to take the Energy Consumption Graph (ECG™), developed by Birdseye Energy Consulting GmbH, which makes sync measurements very simple, without need for configuring electronic devices, cabling between the data points and the logger or between logger and (your) internet provider.
I will tell here how a Distant Energy Manager can use these data to compose the story of your enterprise, in a way that you will not only save energy and enjoy collateral benefits, which together are able to improve the situation in your enterprise considerably, but also to learn in the same time, so that you start to understand how energy actually works. I see my mission not only in helping as many industries as possible in order to make the Energy Transition reality (nearly 50 years after the first broad consensus on energy crises), but also to help the large consumers understand energy. After all, I created the Birdseye Energy School, so capacity building in Energy Management is a big part of what drives me.
OK, here we go.
The first thing we want to know is how energy flows after running through your main meter. With the time measurements or also with only the power data of the biggest plants and their running times, it becomes possible to know not only where but also when the energy is flowing and where the large part of it is consumed. That will tell us where to focus, because where much energy is consumed, usually the most can be saved. We can also see such problems like peak demands in this way, because the time graphs will reveal if big consumers want their energy all at the same time, for example on Monday morning. Peak demands can cost much (because the grid operator must be prepared for such a peak demand even if it occurs only once a year), knowing which plants contribute to this peak can help make decisions about shifting the demand of some of them. It makes a lot of sense and you could think of it yourself to consume at a different time, but that is the point of experience and data acquisition expertise, where Birdseye Energy Consulting can help you for a few times and then you can figure out some of it by yourself. That was number one. Here you see how information from your list of equipment and the time schedule of your enterprise, together with a few measurements can help you to connect the dots between facts and the solution. This is the general way to do this and solve one problem at a time.
So, here we have a general solution to avoid a peak load on the energy supply. From here we go into a deeper analysis because we still don’t know how much we can gain with Demand Side Management (DSM), as this flexibility of the consumer’s side is called. We have also to figure out how much this is going to cost, because flexibility sounds good, but in reality you just need all this energy together on Monday morning, to start up. So you don’t have this flexibility … if you don’t create it yourself. And there is the solution: you can create flexibility in cooling your space of heating your process by storing cold or heat during a time that it is more convenient and then use it later, when you need it. And what is exciting about Energy Efficiency Technology, is that here an unexpected collateral benefit comes into sight: the storage (of cold in an ice storage, for example) allows us to be independent from a unstable grid for a couple of hours. Moreover, the storage can increase the power available, without need to replace your existing chiller with a bigger one (which anyway would not resolve the need for a very consistent surge of power if this double size chiller starts really to work). How is this possible? You remember from above the example of the factory of Thyssen Krupp in Switzerland, making metal forms. They had a rather big building that was difficult to cool in summer. So they used their chiller to charge a 1’400 kWh ice storage during the night and used that to add more power to the chiller when it was really hot. So they had the sum of the cooling power of ice storage and chiller at their disposal, with only the chiller’s electrical power as the maximum electrical load, as far as cooling was concerned.
So, summarising this example of how to come from data to solution, what started only as an idea to shift chiller consumption from the morning to the night (to avoid peak demand and exploit the night tariff), actually delivered not only that in an excellent way, but as a collateral benefit also:
1. Security of Supply (cold), which can really be a stress factor. Indeed, decision makers act often more to avoid risk than to save money. In the case of for example food damage, comfort in an hospital, pensioners home or hotel, this risk tolerance is really small.
2. Extra cooling power, which can be added to that of the chiller, without adding another chiller (which will anyway ask for additional electrical power.
You will note that this ice storage will have also its cost, but you don’t have to justify it with the savings on the day to night tariff and avoiding the peak demand charge. It can also be justified with the Security of cold Supply and the avoided investment into an additional chiller, if the existing chiller is too small. That makes decision-making easier!
Once decided for such a solution, the subsequent step is to analyse these options, so that you can express the advantages in numbers. This is something for the next Newsletter. I will show you also how you can simulate that in a spreadsheet.
3. What adds storage to the simulation of dynamic tariffs?
In the last article above we saw that a simulation can finalise the feasibility analysis of a Demand Side Management (DSM) solution, because the particular load curve of the enterprise can be multiplied with the tariffs, even if these are volatile, so that we can assess the financial impact of the solution.
Don’t forget that an energy storage can be an electric battery, an ice storage with an electric chiller or a container of warm water with an electric heat pump. It is all about storing electrical energy or another commodity that is directly related with it.
I am working on a spreadsheet model, the same as shown in the Newsletter of May 19 (about simulation of dynamic energy tariffs). In the last Newsletter we saw how these tariffs can be interesting even if your enterprise is not able to offer any flexibility. However, that is a chance, it can also become more expensive. Anyway, you will know it if you run a simulation.
If you are interested to take some form of storage to improve the Security of Energy Supply (which for most entrepreneurs will be the main driver, I imagine), then the simulation becomes still more complicated. In any case, even if the economics of dynamic energy tariffs are not on the foreground for you as entrepreneur, it will be advantageous to know how easy you the savings will pay your additional infrastructure (energy storage + control system) with the savings. In this case, you need to take the following parameters into account and use for the simulation:
You want to charge the energy storage during cheap hours and consume it when the supplier asks a high price.
The storage must not be fully charged when you expect a low tariff, because in these conditions you will like to charge it. In the same way, it must not be depleted when energy prices are high, during which you would like to get energy from the storage instead of from the grid.
Another Business Model, available for DSM is that you agree a time slot during the day, during which the grid operator can ask you to decrease power demand by N kW (mostly N is far beyond 100) so that they pay you a kind of insurance premium, even if they never call you. Under this model you want to have charged the storage before the time slot starts, so that a decrease of the demand can be realised without negative consequences for your business.
You must take care of the demand charge, too. Even if energy is very cheap, if in those moments your enterprise is consuming much, you cannot charge the storage at the same time or the demand charge will go up even more.
So in short, you should charge the battery when:
The tariffs are low;
The storage has still the capacity to store it;
The combined consumption of the factory and charging device is not going beyond a level defined by the planned peak demand.
Next Newsletter we will see such a simulation.
4. Internet can give you heat!
Small data centre of in office in Paris, France
Google is shifting to liquid cooling with its latest hardware for artificial intelligence
Last minute picture from https://datacenterfrontier.com
As I told in my blog of June 7, when working at Siemens, I often met Robert of Siemens Milan in Italy, he is now in pension but I think he was the greatest Energy Efficiency designer for data centres in the whole country. He was an artist and did his work with pleasure.
You certainly know that data centres consume a lot of energy, but why should YOU care about them?
First: you might have yourself at least 1 server rack, consuming between 500 and 1200 Watt. Second, maybe there is a datacenter in your neighbourhood, giving loads of thermal energy. An example is described in the article of Peter Gerhardt, May 16, on www.tagesschau.de (in German), saying that the heat from the many data centres in the city of Frankfurt can deliver the heating for the whole city (93’000 inhabitants + enterprises). As reader of the Newsletter I give you the original article, you can always translate it automatically or contact the author.
By the way, on www.computerweekly.com I read that one rack in a big data centre is consuming of over 50kW, much more than what I mentioned for “our own” servers, like on the picture on the left.
How will they recover that heat in this economic centre of Germany?
Nowadays, the heat of the many data centres is discharged with warm air between 25 - 35°C and that will be the enacted technology in Frankfurt, but new technologies work with water cooling, at 60°C. This is much more useful, think about the many applications that come into reach with this shift of 30°C. You can not only use it as warm tap water (where 60°C is nearly enough, considering hygiene rules for water storage) but also objects like in the restaurant (dishes) or hospitals and hotels (laundry). Also in production there are many applications for warm water.
I am only talking about the summer, because in the winter there is no question where to apply the heat at this temperature level: space heating!
Moreover, water at 60°C makes heat easier to transport than air at even the high end of the range, 35°C! It is 4000 times better than air to capture the heat. The problem with water cooling is that there is apprehension to bring water so close to the electronics. It will also cost more, certainly if you want to address this apprehension.
From the other side, new data centres with liquid Heat Recovery technology will be cheaper than the retrofit of existing data centres. Here I can assure you that humankind will continue to build many new data centres, so even if you skip the retrofit opportunities and apply liquid cooling only in the new data centres, you will have a very high impact. For example, one of the largest data centers in Europe is currently being planned in Hanau … yes, near Frankfurt. This plant alone will consume more than twice as much electricity as the entire city with its 93,000 inhabitants. As a technical consideration, I can imagine that water-cooling, together with more concentration of transistors on each square mm and new semiconductor technology with higher temperature limits will allow even higher temperatures than 60°C, making the heat even more useful, for example in absorption cooling machines, but this additional temperature increase is wishful thinking.
Another question is: why use water? We know that water and electricity are not always good friends. In high power transformers, they use a special type of transformation oil for the cooling. One liter of oil can evacuate less heat than water, but still much more than air. So it might be a solution. However, during my research on Internet, it appears that water is still the favourite liquid.
Another advantage of liquid cooling, the one or the other, is that they have much more heat capacity, so you can dissipate more heat with less flow. Less flow means less space. So it is easy to understand that the racks will become much smaller with liquid cooling, because the air channels occupy much space and much energy is needed to blow this air through the racks. Mind you, if you can heat a city like Frankfurt with this air, a few fans will have to move it.
As a conclusion, I can understand the apprehension of technical designers for water but as data centres start to consume more than 1% of all energy in the world (probably the double in 2030), it is a good idea to invest in the technology. There have been bigger challenges. Existing data centres will be difficult to convert as I mentioned, but they don’t live very long and many brand new data centres are being built. Bitcoin mining will become less controversy if we make this shift.