PaperlayerView exact revision on arXiv

Source-linked AI summary

Privacy-Preserving Transactive Energy Management for IoT-aided Smart Homes via Blockchain

Qing Yang, Hao Wang

arXiv:2101.03840v2eess.SYcs.CRcs.DCcs.SYmath.OC

TL;DR

Smart-home transactive energy management must address efficiency, privacy, and implementation on IoT devices. The paper combines grid and peer transactions with a blockchain-based distributed algorithm; the algorithm is guaranteed to converge to the optimal solution because the TEM objective is convex.

Abstract

from arXiv · show

With the booming of smart grid, The ubiquitously deployed smart meters constitutes an energy internet of things. This paper develops a novel blockchain-based transactive energy management system for IoT-aided smart homes. We consider a holistic set of options for smart homes to participate in transactive energy. Smart homes can interact with the grid to perform vertical transactions, e.g., feeding in extra solar energy to the grid and providing demand response service to alleviate the grid load. Smart homes can also interact with peer users to perform horizontal transactions, e.g., peer-to-peer energy trading. However, conventional transactive energy management method suffers from the drawbacks of low efficiency, privacy leakage, and single-point failure. To address these challenges, we develop a privacy-preserving distributed algorithm that enables users to optimally manage their energy usages in parallel via the smart contract on the blockchain. Further, we design an efficient blockchain system tailored for IoT devices and develop the smart contract to support the holistic transactive energy management system. Finally, we evaluate the feasibility and performance of the blockchain-based transactive energy management system through extensive simulations and experiments. The results show that the blockchain-based transactive energy management system is feasible on practical IoT devices and reduces the overall cost by 25%.

I. INTRODUCTION

The introduction frames blockchain-based transactive energy around grid efficiency, user privacy, and IoT-device feasibility. It proposes holistic smart-home transactions supported by distributed privacy-preserving management and an IoT blockchain.

I. INTRODUCTION

  • The proposed system combines grid-facing feed-in PV and demand response with peer-to-peer trading, while a distributed algorithm preserves privacy and the blockchain platform is validated on IoT devices.The paper identifies household cost reduction and grid support as benefits of transactive energy.Source: §I · pp. 1–2

II. RELATED WORKS

  • Related work is assessed across privacy, management algorithms and performance, and blockchain design; the paper contrasts its holistic options with V2G-focused work.The proposed scope includes demand response, feed-in PV, and peer-to-peer energy trading.Source: §II · pp. 2–3

III. SYSTEM DESIGN

  • The system design covers smart-home loads, generation, and electric vehicles, transactive-energy mechanisms, and an IoT-tailored blockchain.Its stated transaction mechanisms include feed-in tariff, demand response, and energy trading.Source: §III · p. 3

A. The Smart Home Model

  • Each smart home includes electrical loads, local PV or wind generation, and an electric vehicle, and connects with the grid, aggregator, and other homes.The smart meter manages supply and load and handles energy and information exchange; users are scheduled over discrete time slots.Source: A. The Smart Home Model · p. 3

1) The Electric Loads of the Smart Home:

  • The load model distinguishes shiftable, curtailable, adjustable HVAC, and inflexible appliances, with discomfort costs modeled for shifting and curtailment.The shiftable-load cost uses a sensitivity coefficient for behavior change; HVAC comfort reflects deviation from a setpoint and is bounded by a tolerable temperature range.Source: 1) The Electric Loads of the Smart Home · pp. 3–4

2) The Power Supply Models:

The power-supply model combines grid purchases with renewable generation and allows users to trade surplus energy.

2) The Power Supply Models:

  • Smart homes can obtain electricity from the grid or their renewable generation, and may trade surplus energy with other grid parties.Grid power is capped by a shared maximum powerline capacity; renewable availability varies with each home's solar and wind conditions.Source: 2) The Power Supply Models · p. 4
  • The grid pricing model includes regular and peak usage prices, with a bill component for peak usage.The stated pricing strategy is intended to incentivize peak shaving.Source: 2) The Power Supply Models · p. 4

3) The Model of the Electric Vehicle (EV):

The EV model represents home availability, battery charging and discharging, and the cost and constraints associated with using the EV to serve household demand.

3) The Model of the Electric Vehicle (EV):

  • The EV is assumed to be parked at home during its availability period, and it must be fully charged before departure.It may also be discharged to serve household load during that period.Source: 3) The Model of the Electric Vehicle… · p. 4
  • Arrival and departure times define the EV’s home-availability window, while its battery energy level and capacity describe stored energy.The model also gives charging and discharging power limits and battery charging and discharging efficiencies.Source: 3) The Model of the Electric Vehicle… · p. 4
  • Battery energy changes over time according to the EV’s charging and discharging operation.The passage introduces the battery dynamics and identifies charging and discharging efficiencies.Source: 3) The Model of the Electric Vehicle… · p. 4
  • V2H discharge incurs a quadratic battery-degradation cost, and the model does not count degradation when the EV is used outside the home.The passage says deeper discharges carry more significant degradation cost.Source: 3) The Model of the Electric Vehicle… · pp. 4–5

B. The Transactive Energy Management Model

The model lets smart-home users transact vertically with the grid through feed-in tariffs and demand response, or horizontally with peers through energy trading. Vertical transactions include selling renewable generation and reducing scheduled load in response to demand-response signals.

B. The Transactive Energy Management Model

1) Energy Trading of the Vertical Transactions:

  • The utility sets a feed-in tariff price, and users choose how much renewable energy to sell to the grid, bounded by available generation.Renewable energy serving local demand plus energy sold to the grid cannot exceed available renewable generation.Source: 1) Energy Trading of the Vertical… · p. 5
  • Feed-in energy earns a tariff reward, while demand-response load reduction is non-negative and bounded by scheduled grid load.The passages identify rewards for both feed-in energy and reducing demand in response to signals.Source: 1) Energy Trading of the Vertical… · p. 5

2) Energy Trading of the Horizontal Transactions:

Horizontal transactions let users trade energy with peers at transactive energy prices sent by the distribution system. The model treats nearby users’ exchange losses as negligible and defines trades by whether a user sells or purchases energy.

2) Energy Trading of the Horizontal Transactions:

  • For a user pair, positive traded energy means the user sells to the peer, while negative traded energy means the user purchases.The model assumes energy losses are negligible because users are located close to each other.Source: 2) Energy Trading of the Horizontal… · p. 5
  • Users trade at prices sent by the distribution system; sellers receive payments from counterparts and purchasers pay them.The passage identifies the transactive energy price as the basis for peer payments.Source: 2) Energy Trading of the Horizontal… · p. 5
  • The section frames transactive energy management as a distributed algorithmic design for both vertical and horizontal transactions.It states that the transaction model and distributed algorithm are developed in the following section.Source: 2) Energy Trading of the Horizontal… · p. 5

C. The IoT Blockchain on the Smart Meters

The IoT blockchain runs on smart meters as a peer-to-peer platform for energy transactions, but its design must accommodate their limited hardware resources. The system adapts PBFT consensus and the block structure, and uses transaction types and smart contracts to support vertical and horizontal energy management.

C. The IoT Blockchain on the Smart Meters

  • Smart-meter blockchain nodes connect over technologies including powerline, Wi-Fi, Ethernet, LoRa, and 5G Narrowband IoT, and use gossip to transmit transactions.The connected nodes form a peer-to-peer network.Source: C. The IoT Blockchain on the Smart… · pp. 5–6
  • Smart meters have constrained resources, while deploying nodes on them can use existing grid infrastructure and let nodes retrieve trusted trading data directly from meters.The paper notes that typical IoT devices have embedded CPUs, less than 1 GB of memory, and bandwidth below 200 Kbps; existing blockchain software cannot be directly deployed on them.Source: C. The IoT Blockchain on the Smart… · p. 6

1) The Consensus Protocol:

  • The modified PBFT uses round-robin leader selection and message aggregation to reduce complexity and improve robustness for IoT devices.Validators take turns leading consensus after a successful block; aggregation reduces communication complexity from O(n^2) to O(n).Source: 1) The Consensus Protocol · pp. 6–7

2) The Transaction and Block Structure:

IV. PROBLEM FORMULATION AND ALGORITHM DESIGN

The section formulates system-wide transactive energy management for smart homes, balancing each home’s supply and demand while accounting for internal costs and external transactions. It focuses on full transactions and introduces three benchmark scenarios.

IV. PROBLEM FORMULATION AND ALGORITHM DESIGN

  • The formulation introduces three benchmark scenarios for comparison with the transactive-energy management approach.Source: §IV · p. 7

A. Modeling and Optimization of the Transactive Energy Management

B. Distributed Optimization Method for TEM

The method uses ADMM to split transactive-energy optimization between users’ local computations and a smart-contract task. Users optimize in parallel while the contract coordinates trading variables, with convergence to the convex problem’s optimum.

B. Distributed Optimization Method for TEM

C. Benchmark Scenarios

The paper compares full transactive-energy management with three cases that isolate internal scheduling, vertical transactions, or horizontal transactions. It notes that horizontal trading couples users’ decisions, whereas the first two benchmarks decouple them.

C. Benchmark Scenarios

  • Benchmark 1 schedules only each home’s internal energy supply and demand, without transactive-energy activities.Source: C. Benchmark Scenarios · p. 9
  • Benchmark 2 jointly optimizes internal schedules and vertical feed-in renewable-energy and demand-response transactions.Source: C. Benchmark Scenarios · p. 9
  • Benchmark 3 jointly optimizes internal schedules and horizontal peer trading; unlike the first two cases, its coupled trading decisions are solved with a distributed algorithm.Source: C. Benchmark Scenarios · pp. 9–10

V. SYSTEM IMPLEMENTATION AND PERFORMANCE EVALUATION

The evaluation combines a practical IoT-device test of the blockchain design with numerical simulations of the energy-management algorithm using data from practical applications.

V. SYSTEM IMPLEMENTATION AND PERFORMANCE EVALUATION

  • The section has two parts: testing the blockchain design on a realistic IoT-device network and evaluating the management algorithm through numerical simulations.Source: §V · p. 10
  • The blockchain design is tested on a realistic network of IoT devices.Source: §V · p. 10
  • The algorithm evaluation uses numerical simulations with data collected from practical applications.Source: §V · p. 10

A. Performance Evaluation of the IoT Blockchain

The IoT blockchain was tested on a network of Raspberry Pi devices and evaluated for resource use and transaction performance. Tests report that it runs smoothly on IoT devices and provides throughput sufficient for the distributed energy-management algorithm.

1) Experiment Setup:

  • The evaluation network uses 11 Raspberry Pis, with Model 3B+ and Model 2B devices emulating high-end and low-end IoT nodes.The router limits network bandwidth to less than 250 Kbps.Source: 1) Experiment Setup · p. 10, Figure 5 · p. 10

2) The IoT Blockchain Evaluation:

  • On a Raspberry Pi 3B+, the IoT blockchain validator uses less than 50% CPU, while Ethereum mining uses 100% and makes the operating system very slow.The authors report that their blockchain runs smoothly on the IoT devices.Source: 2) The IoT Blockchain Evaluation · p. 10
  • Transaction delay is about 5 ms, block confirmation takes less than 100 ms, and peak throughput is around 700 TPS.The measured performance was reported as sufficient to support execution of the distributed transactive energy management algorithm.Source: 2) The IoT Blockchain Evaluation · p. 10

1) Simulation Data and Parameters:

Simulations use a week of energy and weather data to evaluate scheduling, transactive-energy participation, and costs. The algorithm converges in the tested five- and ten-user cases, and holistic transactive energy reduces total costs by about 25%; the conclusion identifies large-scale deployment as future work.

1) Simulation Data and Parameters:

2) Algorithm Convergence:

  • With convergence error threshold ε = 1 × 10^-10, the algorithm converges at iteration 16 for five users and iteration 22 for ten users.Fig. 6 plots convergence error for the two user counts.Source: 2) Algorithm Convergence · p. 10, Figure 6 · p. 11

3) Power Scheduling in Smart Homes:

  • Over the simulated week, users use daytime PV for demand, buy grid electricity mostly in early morning and at night, and discharge EVs during evening peaks before recharging overnight.Positive EV values indicate charging and negative values indicate discharging.Source: 3) Power Scheduling in Smart Homes · p. 11, Figure 7 · p. 11

5) Benefits of Transactive Energy:

  • About 25% lower overall costs result from holistic transactive energy, compared with about 16% for vertical and 11% for horizontal transactions.Fig. 10 compares total costs for ten users over one week across TEM and three benchmark scenarios.Source: Figure 10 · p. 11, 5) Benefits of Transactive Energy · p. 12

4) Transactive Energy within the Grid and Among Smart Homes:

VI. CONCLUSION

  • The system combines grid and peer transactions with privacy-preserving distributed optimization, and the authors identify reducing algorithmic complexity for networks of tens of thousands of nodes as future work.The conclusion reports feasibility on practical IoT devices and a 25% overall cost reduction.Source: §VI · p. 12
Loading 2101.03840v2…