Shopping cart
0 Items | US$ 0.00
This actual strong demand for Blockchain implementations from energy com-
panies, shows the willingnes of them to participate in the upcoming changes.
The change from a centralized utility grid, with a few energy suppliers in
monopoly positions, to a decentralized and thereby more transparent utility
grid.
Primarily this book provides a detailed consideration of the Blockchain tech-
nology and its use cases in the energy sector.
In the following it will be examined how the Blockchain solved a mathemat-
ical problem, which was thought to be unsolvable until then, known as The
Byzantine Generals Problem. Furthermore, basic principles of a Blockchain,
like decentralization and its functionality, will be set out.
As most of the Blockchain energy Dapps are build on Smart Contracts, this
thesis takes a closer look on how Ethereum works, as it is the first and
so far only public platform, that enables to easily write Smart Contracts.
Ethereum's Blockchain structure will be analysed and it's differences com-
pared to other Blockchains.
Afterwards an insight into ongoing projects in the field of energy management
is given. This book will outline whether there are reasonable implementations
in the field of energy management. Furthermore, the possibilities of adapting
those implementations to the energy market will be described. However, it
is to mention, that most of the described applications are still in the trial
period.
Finally, an implementation in form of an own peer-to-peer energy market
infrastructure is given, using a private Blockchain on the Ethereum plat-
form.
5
2
Blockchain
Chapter 2 focuses on the Blockchain technology. First we inspect the actual
mathematical problem that is solved by the Blockchain known as The Byzan-
tine Generals Problem. After that we approach the important attribute of
decentralization that comes with the Blockchain. The chapter ends by giving
a definition of the basic Blockchain.
The descriptions in this book are intended to give the reader a fundamental
and vivid understanding of the Blockchain technology, hence highly mathe-
matical proofs are omitted in this chapter.
2.1
The Byzantine Generals Problem
The Byzantine Generals Problem is a computer science problem, that was
known as unsolvable. Lamport, Shostak and Pease were the first one to
point out this issue and released their paper about The Byzantine Generals
Problem in 1982 [19].
It describes the problem that corrupt nodes in a
decentralized network can lie to loyal nodes by passing false information.
This lie spreads among the complete network and could lead to an undesirable
behaviour.
To abstract this problem every node in a decentralized network is represented
as a general, all generals can communicate with each other by a messenger.
There can be corrupt generals among them. The generals can only choose
between two actions "attack" and "retreat".
They win the battle if:
1. Loyal generals, which is assumed to be the majority of all generals,
make the same decisions independently of the decision of the corrupt
generals
2. A small number of corrupt generals can not change the main decision
of all loyal generals
It can be shown that this problem is unsolvable unless we add some more
conditions e.g. two-thirds of the generals are loyal etc. [19].
We inspect an example for a better understanding. Lets imagine there is
a castle of enemies with a 300-men army, surrounded by a 500-men-army,
6
which is subdivided into 5 divisions. Each division consists of 100 soldiers
and got its own general. Every general gets the same priority, so each of
them can suggest when all of them should attack. Surely 500 soldiers would
easily surpass 300 enemies but only if they all attack at the same time. So
the generals have to agree on a time. General 1 decides to attack at 8 PM,
so he sends his rider out with the message "Attack at 8 PM" to deliver it
to general 2. General 2 agrees on this and sends his rider out to share the
message "Attack at 8 PM" with general 3. Now we assume that general 3
is corrupt, so he sends his rider out to general 4 with the false information
"Attack at 9 PM". This false information spreads among the generals 4 and
5. This scenario is shown in figure 2.
What finally happens is, that general 1 and 2 attack at 8 PM with a total
of 200 soldiers and get defeated by 300 soldiers of the enemy. General 4 and
5 attack at 9 PM and get also overpowered by the enemies. Thereby the
corrupt general 3 has reached his goal and the attack failed.
Figure 2: One visualization of the Byzantine Generals Problem
The Byzantine Generals problem basically shows that it is impossible to cre-
ate a consensus in a decentralized network.
Lamport, Shostak and Pease also suggested "A solution with signed mes-
sages" in their paper about The Byzantine Generals Problem.
They added following assumptions [19]:
7
1. A loyal general's signature cannot be forged, and any alteration of the
contents of his signed messages can be detected.
2. Anyone can verify the authenticity of a general's signature.
If we find an algorithm, that can assure those two additional assumptions,
than we have an solution for the Byzantine Generals Problem.
The Blockchain technology provides such a solution for the Byzantine Gen-
erals Problem [24]. In a Blockchain consesus is reached by using a Proof-of-
Work algorithm.
In short this algorithm defines a complex in calculating task whose solution is
easy to verify for others and its calculation time can be estimated in advance.
The information of a node is only forwarded after it has solved this task. In
most Blockchain applications this is realised by computing hashes until one
node finds a hash that complies with the conditions set by the application.
The Proof-of-Work method, which will be described in the next sections, can
be seen as a way of fulfilling the prior described assumptions.
Lets go back to our example and see how the scenario changes if we use the
Blockchain technology.
First of all we define three requirements that every transmitted message need
to fulfil before it is considered as trustful.
1. The preparation of a message have to take a certain time (eg.
10
minutes)
2. Every new message must include the complete history of all transmitted
messages
3. The message is broadcasted to all nodes simultanously
The first requirement simply ensures, that the content of a message takes 10
minutes to create. As an important consequence out of this, it also takes 10
minutes to copy or falsify that message. In our example this serves as the
Proof-of-Work task for all generals.
The second requirement assures, that the effort to falsify a message is in-
creased because it is also necessary to forge all previously transmitted mes-
sages to create a transmittable false message.
The third requirement ensures that all nodes in the network receives a mes-
sage at the same time.
8
Now we will adapt this, to our previous explained example. What happens
is, that the first general needs 10 minutes to create a message. His riders
bring that message to all other generals who know that the message took 10
minutes to create. They all start preparing a message that embed general 1's
message themselves. After one of them finish we get two messages chained
together, a so called Proof-of-Work Chain. Let's assume general 2 finishes his
message first and broadcasts it to the other generals. They will immediately
stop writing on their actual messages and start writing a new message that
embed general 2's message, which already includes general 1's message. And
so on until they all agreed upon a plan.
The type of consensus that is
reached, can be described as an emerged consensus as there is no specific set
time where all generals agree on a message. Rather, the generals express their
acceptance by embeding the obtained message in their own. The scenario
using signed messages, is shown in figure 3.
What about our corrupt general 3? In every possible case his false informa-
tion won't lead to an unwanted behavior. If he is the first one to transmit
a message, all generals will paste his message in theirs and attack at the
time he suggested. Alternatively, if he is not the first one, in every step he
not only must prepare a message that includes all previous messages, he also
has to tamper all prior messages, before a loyal general finishes his message.
Satoshi Nakamoto showed in his whitepaper, that the probability of an at-
tacker archieving this is "[..]exponentially diminishing" [25], which means it
is nearly impossible. This proofs, that a message coming from a general can
be seen as immutable. Additionally, the loyal generals will always trust on
the longest Proof-of-Work chain, which is the message that most generals
signed on as there is the most effort put into.
The solution with Blockchain complies in its sense with the suggested one,
using signed messages [19]. Consequently the Blockchain basically solves a
mathematical problem and makes it possible to create consensus in a decen-
tralized network.
9
Figure 3: The Byzantine Generals Problem solved by signed messages
10
2.2
Decentralized Networks
The Blockchain technology is based on decentralized networks, as mentioned
before. In general, a network is a system consisting of multiple nodes, that
can communicate and share resources together. The way how networks in-
teract with each other vary and can be classified into two types. For the
scope of this thesis we will focus on both, the decentralized network and the
nowadays most common type the centralized network. [2]
These days most of the internet is centralized [16]. That means information
and services are stored in a server node and can be requested by a client node.
A big disadvantage of centralized networks is the Single Point of Failure
structure. That means a defect of the central node leads to cancellation of
the communication between the end stations. Furthermore, multiple requests
at the same time on the same server lead to a longer server respond time.
Another disadvantage is, that if it comes to trusted information transmitted
over the server a loss of privacy is nearly always the case. This is due to
the fact, that the information goes trough a third controlling party, a so
called intermediary. The intermediary is able to store those information in
local databases. This makes centralized networks very vulnerable, because
attacking the central node could harm all participants of the network. Figure
4 shows how a centralized network is built up.
Figure 4: Representation of a centralized network
11
However, centralized networks can offer advantages. The data stored in the
server is more dependable because all modifications are stored locally. More-
over the server-node keeps the network secure, because any malicious request
from a client gets refused and reduces data loss [16].
Centralized structures can be found in finance or energy institutions. A
bank for example works centralized. The bank enables a secure transaction
between two parties, so that the bank branches function as trusted third party
intermediaries for their customers. Therefore parties who wish to transact
with each other need to trust on a central system. Sensitive information of
the clients such as assets, transactions, acces keys etc. are stored in central
databases of the bank. This is done to mediate between two customers, if
there are any types of disputes. Therefore attacking those databases can
harm many customers.
Energy suppliers record their customers energy consumption and delivers
constantly energy from power stations to the clients house, acting as an
intermediary between power producer and consumer.
The second network structure is the decentralized one, which offers direct
Peer-to-Peer (P2P) connections between all nodes. The essential difference
between decentralized and centralized networks is, that there is no central
point of control, hence no one owns the network and no intermediary is
needed. This results in information being directly and faster transmitted to
the recipient. Furthermore all nodes are equal and store the same informa-
tion. In contrast to a centralized network, the decentralized network still
works, even if some nodes break down, with no loss of data or integrity.
Figure 5 shows, how nodes in a decentralized network can communicate with
each other via a direct link.
Figure 5: Representation of a decentralized network
12
Decentralized networks showed disadvantages in the past. They were known
as very vulnerable, because there is no controlling entity and all nodes can
act anonymously. Thus any corrupt node can directly transmits malicious or
false information to another one. However, after the Blockchain technology
solved this issue many interesting applications were found for decentralized
networks e.g. Smart Contracts, Cryptocurrencies etc. The advantages of
this applications are, that there can be an exchange of values anonymously,
so all private information stick to oneself, and without having an controlling
entity. Rather all nodes have to reach together consensus, to keep the network
secure. How this works in a Blockchain will be described in chapter 2.3.
If we adopt this to our previous described examples, transactions in the form
of money or energy, could take place completely without intermediaries such
as banks or energy suppliers. For the customers this accompany with many
advantages, which will be described in chapter 4.
Summing up, this chapter showed, that the main purpose of decentralization
in a Blockchain is to enable a free exchange of values without the need of an
intermediary.
13
2.3
Blockchain
The following chapter provides an overview about the functionality of a
Blockchain. First of all hashing and its importance in relation to Blockchain
technology is described. Afterwards the basic Blockchain technology will be
discussed.
2.3.1
Hashing
The encryption of events and the mining of new blocks in a Blockchain, is
based on the usage of hash functions. A hash function assigns every input
string of arbitrary lenght, an output value (hash) of fixed length and consis-
tent random value. Modifications in the input string result in a new hash,
which makes it possible to detect tampered information. As one can see in
figure 6, hashing messages with small differences, results in big differences in
their hashes.
Figure 6: Hashing messages with the cryptographical hash function SHA256
In concrete, cryptographic hash functions are applied, which are known to
be one-way functions. As Xuejia Lai and James L. Massey showed, recre-
ating a message out of a n-bit long hash takes at least 2
n
computations.
This is infeasible to calculate, if we consider that most hashes got several
hundred of digits [17]. That means it is impossible to reassemble the input
string, only by knowing the hashed value. This is on one hand important
to ensure the anonymity of the Blockchain participants. On the other hand
arbitrary large amount of data can be mapped on a single hash. This brings
many advantages, in terms of storage savings as will be shown in chapter 3.
Furthermore calculating complex hashes is the main idea behind the Proof-
of-Work method, which is used to generate new blocks and guarantee the
network security, as will be described in the next section.
14
2.3.2
The Basic Blockchain Technology
This section is called "The Basic Blockchain Technology", as content and
structure of a Blockchain differs for every application. The elementary Blockchain
concept can be implemented in various ways. Although there can be still
found commonalities in the fundamental functionality of most known Blockchain
applications, which we try to summarize here. Furthermore it will be shown
why Blockchain applications are more vulnerabe without an intern economy
and how a Blockchain can be attacked. Chapter 3 discusses the Ethereum
Blockchain in detail, hence this chapter focuses more on the functionality
than on the technical implementation.
For a better understanding, we need to differentiate between a Blockchain
and the Blockchain technology.
Basically a Blockchain is a database, distributed over a decentralized net-
work. It consists of blocks, which contains long lists of verified digital events
that have occured between participants of the network. A Digital event can
be for example the execution of a smart contract or transferring cryptocur-
rencies between two parties. Each Block is cryptographically chained to the
prior Block by including the previous block's hash in itself. The first created
block in a Blockchain is called the genesis block. This block does not have a
reference to a previous block.
Every node in the network owns a local copy of the Blockchain and therefore
a record of all occured events in the network. It can be compared to a public
ledger.
Figure 7: Schematic representation of a Blockchain
The Blockchain technology enables it to create consensus in a decentralized
network. New events get cryptographically hashed and broadcasted to all
nodes in the network. Until this point the events are only announced but
not verified. To verify them, some nodes collect new events into a block and
try to find a difficult Proof-of-Work (PoW) for their block. The Proof-of-
Work method was already mentioned in a prior section and can be described
15
as "solving a cryptographical puzzle". Nodes try to find a hash for their
block that begins with a predefinded number of zero bits by adding a nonce.
A nonce is a random number of arbitrary length, which is only used once
to try finding the required hash. If the actual nonce does not delivers the
correct hash, another random number is generated and used for calculating
the next hash.
In figure 8 the procedure of finding a hash for N events with 4 leading zeros
by adding a nonce is shown.
Figure 8: Successful attempt of finding a hash
The average CPU working time is exponential in the number of zero bits
required [25]. Therefore, depending on how fast the network creates blocks,
the difficulty can be increased or decreased by changing the required zero bits
at the beginning of the hash. After one node solves this puzzle, by finding
a hash that matches the condition, his block is broadcasted to all members
of the network. The process of creating new blocks is called mining and the
mining nodes are named miner. Multiple miners can work together on finding
a correct block, by building a mining pool. Other nodes now can check easily
if the new block's hash have the required number of zero bits. They express
their acceptance by adding the received block into their local Blockchain copy
and start working on a new block, including the just received blocks hash as
the previous hash in their new block. In case two miners find simultaneously
a correct new block, the other nodes will work on the first one they receive
and save the other branch. After the next block is found, all miners continue
to work on the longest branch. The longest Proof-of-Work chain is considered
to be the correct one, because the most computational effort is put into it.
16
Loyal nodes will embed new blocks in their local blockchain copy, only if all
events in the block are valid and loyal miners will only put valid events into
a new block, before hashing them.
Figure 7 shows how such a Blockchain can be constructed. Tampering block
n+1 would require to tamper block n, as block n+1 contains it's hash. Block
n can again only be tampered, if block n-1 gets modified and so on.
This means there is no way to retroactively change or falsify an event in a
block, cause this would mean all blocks after the modified one also needs to
get tampered. It is already shown that the probability of an attacker tamper-
ing all following blocks, while new blocks are added, diminishes exponentially.[25]
This technology makes the Blockchain, which can also be called Proof-of-
Work Chain, a from afterward-tampling secured record of all events without
the need of trusted third party intermediaries.
In a nutshell consensus is reached by all loyal nodes following special rules:
· The longest Proof-of-Work chain is considered to be the correct one
· Miners only put valid events into blocks
· Nodes accept only blocks with valid events
The Blockchain gets vulnerable, if the CPU power of corrupt nodes surpass
the computing power of all honest nodes. This happens when corrupt nodes
make up the majority of the network and control over 51% of the networks
CPU Power. This is called the "51 % Hack". It is equivalent to the fact that
most new blocks, containing false events, are generated by corrupt partici-
pants of the network. A block with false events would usually get discarded
by honest nodes and not embed in the longest Proof-of-Work chain. But when
corrupt nodes make the majority of the network, they will accept those false
blocks in their local Blockchain copy.
But why should the participants follow those rules and help enlarging the
network? Here the intern economy comes into play. In the section before
it was not mentioned, that miners get rewarded if their block is added to
the longest PoW Chain. Depending on what the intern economy is based of,
they will get rewarded with it, because they offer computing power and thus
electricity. This reward creates the incentive for all nodes to act in terms of
the network, as it is more profitable for them.
17
Excerpt out of 62 pages
- scroll top
Details
- Quote paper
- , 2017, An Introduction to Blockchain and its Applications. With a Focus on Energy Management, Hamburg, Bedey Media GmbH, https://www.anchor-publishing.com/document/416145
Comments