The main dilemma and, at the same time, challenge is that high-frequency metering data which are required for efficient network operations, e.g. managing load, demand side response and management and load shedding or shifting, may expose private information. To preserve user privacy, local gateways should not be able to access the content of consumer's data. To enable them to perform data aggregation, we use homomorphic encryption techniques for encrypting consumer's data. In this technique, a specific linear algebraic manipulation towards the plaintext is equivalent to another one conducted on the ciphertext. This unique feature allows the local gateway to perform summation and multiplication based aggregation on received consumer data without decrypting them. Existing data aggregation schemes regards power use information as one-dimensional information. With smart meters being used, it is however multi-dimensional in nature, for example, including the amount of energy consumed, at what time and for what purpose the consumption was, and so on. Taking into account all the dimensions allows fine grained control and optimization.
Since the Smart grid network is a hybrid of the power system and a communication network, intrusions should be detected that concern either the physical power system or the communication network or both. Intrusion detection in smart grids presents some challenging issues. First, it is necessary to define which data will be collected and analysed about smart grid operation to detect intrusions. The second challenge is related to where the various modules that comprise an intrusion detection system will be placed in the smart grid system. Another challenge is related to the ability to classify attacks efficiently and effectively through the use of robust classification algorithms.
In order to accomplish this, we use at Tech a Support Vector Machine (SVM) as the classification algorithm because of its convenience of usage and high accuracy in classification. We will also apply the artificial immune systems (AIS) as another classification algorithm. In addition we will apply several predictive modelling techniques to understand the IoT based botnets in Smart Grids as well as develop techniques to detect them.
BlockChain, which is a distributed ledger, will help in eliminating the need of a centralized hierarchical system and will enable multiple entities to have the same copy of a record. This will add in redundancy and thus enable cross validation of records. BlockChain based smart grids will be able to validate authenticity of user while maintaining anonymity with the aid of a public/private key pair which will be unique to each entity in the network.
A BlockChain keeps a track of the entities who were involved in a transaction along with the time they agreed upon it and gets this information validated from an entity known as the miner. The miner validates a transaction and signs it with a time stamp thus making it irreversible and unalterable. In a BlockChain setup, an adversary needs to own 51% of the overall network to launch a software attack to alter a BlockChain record, thus making it hard to be compromised. In case of a hardware attack adversaries tend to exploit hardware vulnerabilities and launch attacks such as DDoS attacks. Such attacks will be avoided with the use of smart contracts which will operate on the basis of specific information of the smart meter: vendor, model and firmware version. At Tech, we use this data to verify both whether a smart meter is vulnerable and if a security patch exists. Such verification will be carried out against information stored in a dedicated knowledge base which tracks over time security vulnerabilities and related patches of smart meters. A dedicated smart contract will deal with updating this knowledge base, periodically or on demand. All the assets in the smart grid will be validated through smart contracts consisting of asset registries.
