Contemporary Industry Practice and Future Developments

Introduction

The Internet of Things (IoT) has moved from isolated sensor networks to an essential infrastructure layer for electrical utilities, manufacturing, buildings, transport, and distributed energy systems. Contemporary practice is no longer concerned only with connecting devices. It must also address the integration of information technology (IT) and operational technology (OT), cyber-resilience, interoperability, energy consumption, artificial intelligence, regulatory compliance, and workforce capability.

An IoT-enabled electrical system combines physical assets—such as generators, transformers, motors, meters, batteries, and electric vehicles—with communication networks, data platforms, control algorithms, and human decision-makers. The engineering challenge is therefore socio-technical: a technically successful system must also be safe, maintainable, economically justified, secure, and environmentally responsible.

This chapter examines current industrial requirements and important future developments, including zero-trust security, 5G and 6G, edge AI, autonomous energy systems, transactive energy, vehicle-to-grid operation, blockchain applications, and circular IoT design.

Learning Objectives

By the end of this chapter, students should be able to:

1. Explain the drivers and risks of IT/OT convergence.
2. Apply zero-trust principles to connected electrical systems.
3. Evaluate interoperability, energy-aware computing, and platform choices.
4. Describe the roles of 5G, 6G, edge AI, and autonomous energy systems.
5. Analyse transactive energy and vehicle-to-grid applications.
6. Identify suitable and unsuitable blockchain use cases in energy systems.
7. Explain the principles of sustainable and circular IoT.
8. Assess workforce competencies and governance requirements for future practice.

1. IT/OT Convergence

1.1 Meaning and drivers

Information technology traditionally manages data, enterprise applications, identity, and business processes. Operational technology monitors and controls physical processes. Examples of OT include supervisory control and data acquisition (SCADA), programmable logic controllers (PLCs), protection relays, distributed control systems, and building automation controllers.

Historically, OT networks were isolated, used proprietary protocols, and prioritised availability and deterministic operation. IT networks emphasised data sharing, rapid software change, and confidentiality. IoT has reduced this separation. A smart factory may send motor-condition data to a cloud platform, while a utility may use enterprise analytics to influence feeder operation.

The principal benefits of convergence are:

• improved visibility of physical assets;
• predictive maintenance and reduced downtime;
• integration of energy, production, and business data;
• remote operation and expert support;
• better forecasting and asset utilisation; and
• faster deployment of advanced analytics.

The principal risks are expanded attack surfaces, inappropriate software updates, dependence on external platforms, data-quality problems, and the possibility that a cyber incident produces physical consequences.

1.2 Engineering requirements

A converged architecture should separate functions according to criticality. A typical arrangement contains field devices, a local control layer, an industrial demilitarised zone, supervisory applications, and enterprise or cloud services. Critical protection and emergency shutdown functions should not depend solely on a remote cloud service.

Important design practices include:

• asset inventory and network mapping;
• segmentation using zones and conduits;
• least-privilege access;
• authenticated and encrypted communications;
• secure remote maintenance;
• time synchronisation for event analysis;
• tested backup and recovery procedures;
• change control and configuration baselines; and
• monitoring of both cyber and physical anomalies.

The Purdue Enterprise Reference Architecture remains a useful conceptual model, although modern architectures increasingly use zero-trust principles, distributed edge computing, and software-defined networking rather than relying on rigid network layers alone.

1.3 Safety and reliability

Security decisions must respect process safety. For example, installing a security gateway that introduces unacceptable control latency may be inappropriate for a high-speed protection function. Conversely, leaving a legacy relay permanently exposed because it is difficult to patch is also unacceptable. Engineers must evaluate latency, availability, fail-safe behaviour, and recovery time alongside confidentiality.

A useful reliability measure is availability:

Here, A is steady-state availability, MTBF is mean time between failures, and MTTR is mean time to repair. Both time quantities must use the same units. This expression assumes a repairable system with statistically stable average failure and repair rates. In a redundant system, component availability cannot simply be averaged; the system configuration must be analysed.

2. Zero-Trust Architectures for Smart Electrical Systems

Zero trust is a security model based on the principle “never trust, always verify.” It does not mean that every communication is blocked. It means that trust is not granted merely because a user, device, or application is inside a corporate network.

2.1 Core principles

A practical zero-trust architecture includes:

1. Continuous verification: authenticate users, devices, applications, and services.
2. Least privilege: grant only the access needed for a defined task and time period.
3. Assume breach: design for containment, detection, and recovery after compromise.
4. Explicit policy enforcement: base decisions on identity, device condition, location, time, sensitivity, and operational context.
5. Continuous monitoring: collect logs and telemetry sufficient to detect abnormal behaviour.
6. Micro-segmentation: restrict lateral movement between devices and services.

In an electrical substation, for example, an engineer may be permitted to read relay data but not change protection settings. A vendor may receive temporary access through a monitored gateway, with approval, multifactor authentication, session recording, and automatic expiration.

2.2 Identity and device trust

Identity must apply to machines as well as people. Device certificates, secure boot, hardware-backed keys, signed firmware, and software bills of materials can help establish device provenance. However, authentication does not prove that a device is safe. A validly authenticated device may still be compromised. Therefore, authorisation should also consider firmware status, vulnerability state, observed behaviour, and operational role.

Zero trust is particularly difficult in legacy OT environments because some devices lack modern cryptography, cannot be patched without outages, or use shared accounts. Compensating controls include protocol-aware firewalls, unidirectional gateways, network monitoring, restricted maintenance windows, and physical controls.

3. Platform Interoperability and Data Governance

Interoperability is the ability of independent systems to exchange information and use it correctly. It has at least three dimensions:

• Technical interoperability: systems can physically and electronically communicate.
• Syntactic interoperability: exchanged data follows a common structure or format.
• Semantic interoperability: systems interpret the meaning, units, timing, and quality of data consistently.

A temperature value without a unit, sensor location, timestamp, calibration status, or uncertainty is not sufficient for reliable engineering analysis.

3.1 Open standards and models

Interoperability can be supported by open protocols and information models such as MQTT, OPC UA, Modbus, IEC 61850, and common energy-data standards. Each serves different purposes. MQTT is lightweight and useful for publish–subscribe telemetry, but it does not by itself define the meaning of every measurement. IEC 61850 provides rich models and communication services for substation automation, but implementation requires specialist knowledge.

A platform-selection process should examine:

Interoperability does not imply that every system should use one universal platform. A federated architecture may be preferable, provided that interfaces, data ownership, and responsibility are clearly specified.

3.2 Data quality

For a measured variable x, a simple measurement error is:

where x_m is the measured value and x_r is a reference value, both in the same unit. Relative error is:

expressed as a fraction or multiplied by 100 for a percentage, provided x_r is not zero. These equations describe error, not uncertainty. Engineering decisions should also consider calibration, sampling, missing data, drift, and uncertainty distributions.

4. Energy-Aware Computing

IoT computing consumes energy in sensing, processing, communication, storage, cooling, and manufacturing. In battery-powered devices, communication may dominate operational energy. In data centres, computation and cooling are major loads. Energy-aware computing therefore seeks to provide required service with minimum energy and carbon impact.

A basic energy relationship is:

where E is energy in joules, P is power in watts, and t is time in seconds. If power is measured in kilowatts and time in hours, energy is obtained in kilowatt-hours.

For a device operating in several states:

where P_i and t_i are the power and duration of state i. The model assumes approximately constant average power during each state.

Energy-aware design methods include duty cycling, adaptive sampling, local filtering, event-triggered communication, low-power wide-area networking, efficient machine-learning models, workload scheduling, and renewable-aware data-centre operation. Designers should not optimise only device electricity. A low-power sensor that sends excessive data to a carbon-intensive cloud may not produce a lower life-cycle impact.

A broader metric is energy intensity:

where I_E is energy per service unit, E is energy consumed, and S is useful service delivered—for example, kilowatt-hours per processed transaction or joules per valid measurement. The service definition must be explicit, otherwise comparisons are misleading.

5. Workforce Competencies and Professional Practice

Future electrical engineers require hybrid competence. Core capabilities include:

• power systems, machines, protection, and control;
• embedded systems and industrial networking;
• programming, databases, and cloud–edge architectures;
• cybersecurity and secure systems engineering;
• data analytics and machine learning;
• systems engineering and requirements management;
• standards, safety, privacy, and regulation;
• life-cycle assessment and sustainable design; and
• communication with operators, managers, vendors, and policymakers.

Employers increasingly value the ability to translate between disciplines. A data scientist must understand sensor limitations and operational consequences. An electrical engineer must understand identity management, software deployment, and data governance. Professional development should therefore include laboratory work with realistic networks, incident-response exercises, commissioning practice, documentation, and ethical decision-making.

6. 5G, 6G, and Industrial Connectivity

5G provides improved mobile broadband, reduced latency in suitable configurations, greater device density, and features such as network slicing and private industrial networks. Its value in smart electrical systems depends on deployment conditions, spectrum, backhaul, coverage, resilience, and the actual control requirement.

Potential applications include mobile maintenance, augmented-reality support, asset monitoring, connected vehicles, distributed energy resources, and factory automation. However, not every control loop should be placed on a public mobile network. Protection functions requiring deterministic and highly dependable response may need local wired or dedicated wireless systems.

6G remains a developing research and standardisation area. Expected themes include integrated sensing and communication, higher-frequency bands, AI-native network management, extreme device density, and tighter integration of terrestrial, aerial, and satellite networks. These claims should be treated as development goals rather than guaranteed commercial capabilities. Engineering decisions must be based on tested specifications, not promotional descriptions.

7. Edge AI and Intelligent Control

Edge AI places inference or limited learning close to sensors and actuators. It can reduce latency, bandwidth, and dependence on cloud availability. Examples include detecting abnormal motor vibration, identifying power-quality events, estimating battery state, and recognising unsafe operating conditions.

A typical edge-AI workflow is:

1. define the operational decision and acceptable error;
2. acquire representative and labelled data;
3. clean data and identify bias or missing operating regimes;
4. train and validate a model;
5. compress or adapt the model for the target hardware;
6. test under normal, abnormal, and adversarial conditions;
7. deploy with version control and rollback capability; and
8. monitor performance after deployment.

A model’s accuracy on a test set does not establish safety. Distribution shift, sensor failure, adversarial inputs, and rare events must be considered. In safety-related applications, AI should generally support—not silently replace—validated protection and control logic unless the complete system has been rigorously assured.

8. Autonomous Energy Systems and Transactive Energy

An autonomous energy system coordinates local generation, storage, flexible demand, and grid interaction. It may use forecasts, optimisation, and local control to maintain reliability while reducing cost or emissions. Examples include microgrids, community batteries, remote renewable systems, and industrial energy-management systems.

A simplified power-balance condition is:

All terms are powers in the same time interval and sign convention. The equation is a conservation relationship, not an optimisation objective. A feasible controller must also respect voltage, frequency, thermal, ramp-rate, state-of-charge, and protection constraints.

Transactive energy uses economic or incentive signals to coordinate distributed resources. A price signal may encourage a battery to discharge during a high-demand period or a building to shift cooling load. A basic operating objective over intervals t can be represented as:

where c_t is import price, P_import,t is imported power, r_t is export remuneration, P_export,t is exported power, and C_degradation,t is an estimated battery-degradation cost. The objective is meaningful only when bounded by equipment and network constraints. Prices alone cannot guarantee physical feasibility or fairness.

Transactive systems raise questions about market power, customer consent, price volatility, data privacy, and the treatment of households unable to provide flexibility. A technically efficient market may be socially unacceptable if it shifts risk to vulnerable consumers.

9. Vehicle-to-Grid Integration

Electric vehicles (EVs) can act as flexible loads and, where bidirectional hardware and appropriate standards are available, as distributed storage. Vehicle-to-grid (V2G) operation exports energy from an EV to the grid. Related modes include vehicle-to-home and vehicle-to-building operation.

A simplified battery-energy update is:

Here, E_t is stored energy, P_c is charging power, P_d is discharging power, η_c and η_d are charging and discharging efficiencies, and ∆t is the interval duration. Powers and energy must use compatible units. The equation assumes constant average power and efficiencies during the interval and omits self-discharge and battery-temperature effects.

V2G design must consider battery warranty, user mobility requirements, charger interoperability, communications security, distribution-transformer loading, protection coordination, and compensation. A fleet operator may reserve a minimum state of charge for mobility and offer only the remaining capacity to grid services.

10. Blockchain and Distributed Ledgers

A blockchain is a type of distributed ledger in which records are replicated among participants and linked through cryptographic mechanisms and consensus. It may be useful when multiple organisations need a shared, tamper-evident record but do not want one party to control the database completely.

Possible energy use cases include renewable-energy attribute tracking, settlement among microgrid participants, equipment provenance, and automated execution of contractual rules. Permissioned ledgers are generally more suitable than energy-intensive permissionless systems for regulated electrical applications.

Blockchain is not automatically secure or sustainable. It cannot guarantee that an input meter reading is truthful, and smart-contract errors can produce incorrect transactions. A conventional database may be faster, cheaper, easier to govern, and entirely adequate when a trusted operator already exists. A use case should therefore be justified by governance and audit requirements, not by novelty.

11. Sustainable and Circular IoT

Circular IoT aims to reduce resource extraction and waste by extending product life, enabling repair, re-use, refurbishment, remanufacture, and responsible recycling. Important design principles include modular hardware, replaceable batteries, durable enclosures, open interfaces, long-term software support, secure decommissioning, and material traceability.

Life-cycle assessment considers impacts from raw-material extraction, manufacturing, transport, operation, maintenance, and end-of-life treatment. Operational energy is important, but it is not the only impact. Replacing thousands of functioning sensors merely to obtain a small efficiency improvement may create unnecessary embodied emissions.

A circular system must also address cybersecurity. Re-used devices require secure erasure of credentials, verified firmware, ownership transfer, and removal of residual personal or operational data. Extended software support is essential: a physically durable device that becomes insecure after two years is not genuinely sustainable.

12. Integrated Case Study: Illustrative Smart Industrial Microgrid

Background

Consider an illustrative manufacturing plant with rooftop photovoltaic generation, a battery energy-storage system, variable-speed motors, an EV fleet, and an existing SCADA network. The plant experiences demand charges, occasional outages, and limited visibility of motor condition.

Problem

The plant manager proposes connecting all devices to a cloud platform and allowing automatic optimisation. The engineering team identifies several risks: legacy controllers use shared credentials, the EV chargers use different protocols, the battery vendor retains some operational data, and the proposed AI system has been trained on data from a different factory.

Analysis

The plant first creates an asset inventory and classifies functions. Emergency shutdown, protection, and essential motor controls remain local. Non-critical monitoring and optimisation may use edge and cloud services. A segmented architecture is introduced, with an industrial demilitarised zone between control and enterprise networks.

The team defines a canonical data model containing asset identity, measurement units, timestamps, quality indicators, and cybersecurity metadata. Protocol gateways are used where necessary, but each gateway is documented as a potential failure and attack point.

A zero-trust policy requires named accounts, multifactor authentication for remote access, device certificates where supported, temporary vendor privileges, and continuous logging. Legacy devices are isolated and monitored rather than directly exposed.

For energy management, the controller forecasts photovoltaic production and plant demand, then schedules battery charging, EV charging, and flexible loads. It enforces minimum battery state of charge, inverter limits, feeder limits, and production constraints. The AI model recommends maintenance actions but cannot directly disable a protection function.

Outcome

The plant obtains improved energy visibility and can shift selected loads away from peak-price periods. Operators retain local control during loss of communications. The project also identifies the cost of software support, data ownership, and battery degradation—costs that were absent from the original proposal.

Learning points

1. IT/OT convergence should be staged according to operational risk.
2. Interoperability requires shared meaning, not merely network connectivity.
3. Zero trust must include machines, vendors, and software services.
4. AI should be bounded by validated engineering constraints.
5. Energy and cybersecurity performance must be assessed over the complete life cycle.

13. Challenges, Ethics, and Regulation

Contemporary smart electrical systems create several unresolved challenges:

• Privacy: detailed energy data can reveal occupancy, production schedules, or personal behaviour.
• Accountability: responsibility becomes difficult when decisions involve vendors, algorithms, and autonomous agents.
• Equity: dynamic tariffs and automation may disadvantage consumers with limited ability to shift demand.
• Resilience: dependence on cloud platforms, satellites, or telecommunications may create common-mode failures.
• Cybersecurity: interconnected devices increase the scale and speed of attacks.
• Environmental impact: batteries, semiconductors, rare materials, and electronic waste require responsible management.
• Regulatory compliance: requirements may cover critical infrastructure, data protection, radio spectrum, product safety, and cyber incident reporting.

Engineers should apply privacy by design, security by design, accessibility principles, documented risk assessment, independent testing, and transparent communication of system limitations. Compliance is a baseline, not a substitute for professional judgement.

Best-Practice Design Checklist

Before approving a smart electrical system, ask:

1. Are all assets, data flows, dependencies, and owners known?
2. Which functions are safety-critical or time-critical?
3. Can the system operate safely during communication or cloud failure?
4. Are identities, permissions, firmware, and remote access controlled?
5. Are data semantics, units, quality, and retention defined?
6. Is the platform interoperable and reasonably portable?
7. Has energy use and life-cycle impact been measured?
8. Are AI models validated, monitored, and bounded by engineering constraints?
9. Have customers, workers, and affected communities been considered?
10. Is there a tested decommissioning and recovery plan?

Chapter Summary

Contemporary IoT practice in electrical engineering requires more than device connectivity. IT/OT convergence provides visibility and optimisation but expands cyber and operational risk. Zero-trust architectures reduce implicit trust through continuous verification, least privilege, segmentation, and monitoring. Interoperability depends on technical, syntactic, and semantic compatibility, supported by standards and sound data governance.

Energy-aware computing considers the energy used by sensing, communication, processing, and life-cycle activities. 5G and future 6G networks may support mobile, dense, and intelligent connectivity, but their suitability depends on tested performance and operational requirements. Edge AI can improve responsiveness and privacy, yet requires representative data, validation, monitoring, and human accountability.

Autonomous energy systems, transactive energy, and V2G can coordinate distributed resources, but must obey physical constraints and address fairness, market, battery, and privacy concerns. Blockchain is appropriate only where a shared tamper-evident ledger solves a real governance problem. Circular IoT extends product life, supports repair and reuse, and includes secure software decommissioning. Future electrical engineers therefore need integrated competence in power systems, software, cybersecurity, data, sustainability, ethics, and systems engineering.

Key Terms

IT/OT convergence: Integration of information technology with systems that monitor or control physical operations.

Operational technology: Hardware and software used to monitor or control physical processes.

Zero trust: Security approach that continuously verifies access and grants no implicit trust.

Least privilege: Practice of providing only the permissions required for a defined task.

Micro-segmentation: Division of a network into small security zones to limit lateral movement.

Interoperability: Ability of systems to exchange and correctly use information.

Semantic interoperability: Consistent interpretation of data meaning, units, timing, and quality.

Edge AI: Artificial-intelligence inference or learning performed near data sources.

5G: Fifth-generation mobile communication technology with enhanced capacity and service capabilities.

6G: Emerging generation of mobile communication research beyond 5G.

Autonomous energy system: Energy system that coordinates resources with limited continuous human intervention.

Transactive energy: Coordination of energy resources using economic or incentive signals.

Vehicle-to-grid: Bidirectional transfer of electrical energy between an electric vehicle and the grid.

Blockchain: Distributed ledger using cryptographic linking and consensus mechanisms.

Circular IoT: IoT designed for durability, repair, reuse, refurbishment, and responsible recycling.

Energy intensity: Energy consumed per unit of useful service.

SCADA: Supervisory control and data acquisition system.

Microgrid: Local electrical system capable of coordinating generation, loads, storage, and grid interaction.

Review Questions

Short Answer Questions

1. What are two benefits and two risks of IT/OT convergence?
2. State the central principle of zero-trust security.
3. Distinguish syntactic interoperability from semantic interoperability.
4. What is meant by edge AI?
5. Define vehicle-to-grid operation.

Long Answer Questions

1. Explain how a zero-trust architecture can be applied to an industrial substation.
2. Discuss the engineering and governance requirements for interoperable IoT platforms.
3. Explain energy-aware computing and evaluate its relevance to battery-powered IoT.
4. Describe the opportunities and limitations of transactive energy systems.
5. Discuss how circular-economy principles can be incorporated into IoT product design.

Analytical / Application Questions

1. A factory wants to place all control functions in a cloud platform. Analyse the safety, availability, cybersecurity, and latency issues and propose a safer architecture.
2. Design a control strategy for an EV fleet that provides grid support while preserving minimum mobility energy and limiting battery degradation.
3. Evaluate whether blockchain is justified for renewable-energy certificate tracking when a regulated central registry already exists. State the technical and governance evidence required for your recommendation.

Multiple Choice Questions with Answers

Question 1

Which statement best describes zero-trust security?

A. All devices inside the corporate network are trusted

B. Access is continuously verified and limited according to need

C. Only passwords are required for remote access

D. Security is achieved by disconnecting all systems from the internet

Correct Answer: B. Access is continuously verified and limited according to need

Question 2

Which type of interoperability concerns the consistent meaning and units of exchanged data?

A. Physical interoperability

B. Syntactic interoperability

C. Semantic interoperability

D. Mechanical interoperability

Correct Answer: C. Semantic interoperability

Question 3

A device consumes 2 W for 3 hours. Its energy consumption is:

A. 0.67 Wh

B. 1 Wh

C. 5 Wh

D. 6 Wh

Correct Answer: D. 6 Wh

Question 4

What is a principal advantage of edge AI in an industrial application?

A. It eliminates the need for sensor validation

B. It can reduce latency and dependence on cloud connectivity

C. It guarantees that predictions are always correct

D. It removes the need for cybersecurity

Correct Answer: B. It can reduce latency and dependence on cloud connectivity

Question 5

Which function should generally remain independent of an unreliable remote cloud connection?

A. Emergency protection or shutdown

B. Monthly energy reporting

C. Historical dashboard rendering

D. Non-critical maintenance scheduling

Correct Answer: A. Emergency protection or shutdown

Question 6

In the simplified battery update equation, discharge energy is divided by discharge efficiency because:

A. Efficiency is always greater than one

B. Stored energy required exceeds delivered energy when efficiency is below one

C. Charging and discharging are identical processes

D. Power has no relationship to energy

Correct Answer: B. Stored energy required exceeds delivered energy when efficiency is below one

Question 7

Which is a potential vehicle-to-grid application?

A. Using an EV only as a passive electrical load

B. Exporting stored EV energy to support the grid

C. Removing all charging controls

D. Preventing bidirectional power flow

Correct Answer: B. Exporting stored EV energy to support the grid

Question 8

Why is blockchain not automatically appropriate for every energy-data application?

A. It cannot store any transaction records

B. A conventional database may be cheaper and adequate when a trusted operator exists

C. It cannot use cryptography

D. It always prevents auditing

Correct Answer: B. A conventional database may be cheaper and adequate when a trusted operator exists

Question 9

Which practice best supports circular IoT?

A. Designing devices with short software-support periods

B. Sealing batteries permanently into disposable products

C. Providing modular hardware, repairability, and secure reuse

D. Replacing all devices at fixed annual intervals

Correct Answer: C. Providing modular hardware, repairability, and secure reuse

Question 10

A system has MTBF = 99 hours and MTTR = 1 hour. Its availability using A = MTBF ÷ (MTBF + MTTR) is:

A. 0.01

B. 0.50

C. 0.90

D. 0.99

Correct Answer: D. 0.99