Project Description:
1 Executive Summary
Electric vehicle (EV) adoption is accelerating across APEC economies, with implications that are particularly pronounced within the context of existing building. A single DC quick charger (e.g. 150 kW) can instantly double the peak electrical demand of a medium-sized commercial building, while a cluster of four superchargers (e.g. 360 kW each) can impose a load exceeding 1.4 MW, which frequently exceeds the building’s original electrical de-sign capacity. As a result, property owners seeking to install significant EV charging infrastructure must often pursue comprehensive electrical upgrades, including new transformers, switch-gear, cabling, and revised utility service agreements. However, these upgrades typically cost over several million and require 18 to 24 months to complete. In this connection, for older buildings, heritage structures, and properties in dense urban environments, such enhancements may be physically infeasible due to space limitations, structural constraints, or bottlenecks in the utility distribution network.
This R&D project solves this problem at the building level. The Smart Building Energy Management System with battery-boosted EV charging is a deployable technology solution that allows existing buildings to provide high-power EV fast charging to their occupants, tenants, and visitors without upgrading the building's incoming electrical capacity. The system comprises:
A Battery Energy Storage System (500 kWh) integrated directly into the building's EV charging infrastructure, which slowly absorbs power from the building's existing electrical supply during offpeak hours and discharges at high power rates to EV chargers during peak demand;
A Data-Driven Building Energy Management System powered by machine learning algorithms that predict both building base-load demand and EV charging demand, dynamically allocating the building's fixed electrical capacity between core building systems (HVAC, lighting, elevators) and EV charging to ensure that the building's total demand never exceeds its existing incoming supply limit;
A Cloud-Based Centralized Management Platform with building operator dashboards, tenant/driver mobile applications, and automated demand response capabilities that integrate EV charging into the building's overall energy strategy.
The outcome is a transformative approach to building energy management. A facility with 500 kW of incoming electrical capacity, which originally experienced a peak demand of 350 kW, can now simultaneously support 360 kW of EV supercharging. It represents a total potential load of 710 kW without ever exceeding the 500 kW grid supply. The battery system bridges the gap between supply and demand, while the AI-driven energy management system ensures the battery remains optimally charged and ready to respond. The building thus operates entirely within its existing electrical envelope.
This constitutes a fundamental paradigm shift for low-energy building operations. Rather than treating EV charging as an external load requiring costly infrastructure expansion, the system reframes EV charging as an integrated, actively managed load within the building’s existing energy budget.
2 Project Background and Problem Statement
With the anticipated influx of electric taxis and buses due to government policies, there is a growing demand for more comprehensive e-public transport infrastructure development to fulfill their daily operation needs. To fulfill the operational needs of the eTaxi and eBus, a substantial amount of super chargers are necessary. During the traffic peak hours, the eTaxi and eBus usually consume more than 80% of battery capacity. In order to maintain the daily operation, only super chargers can efficiently charge the vehicles in a short period of time and at a quick charging rate.
While the EV industry has focused primarily on roadside charging networks and purpose-built charging hubs, the majority of EV charging will ultimately occur at buildings, including workplaces, residential buildings, shopping malls, hotels, hospitals, and transport hubs. In Hong Kong alone, over 85% of parking spaces are located inside buildings. Yet, the electrical infrastructure of these buildings was designed decades ago to accommodate conventional loads such as lighting, HVAC, elevators, and plug loads only, with no allowance for the substantial additional demand imposed by EV fast charging.
Under the current paradigm, property owners face a dilemma, either undertake costly electrical upgrades that span several years and disrupt tenants, or forgo EV charging and accept a loss of competitiveness in the commercial property market. For older buildings, heritage structures, and properties in dense urban environments, physical upgrades may not even be feasible due to spatial, structural, or utility infrastructure constraints. This project was developed to provide a third alternative that enhances building intelligence rather than expanding its physical footprint, but also support Hong Kong Government’s policy and to catch the opportunity ahead.
3 Technical Solution and Innovation
As announced in the Policy Address 2023, there are 700 electric buses and about 3,000 electric taxis will be deployed in Hong Kong. Therefore, the number of super chargers in Hong Kong is the key of successful in the full deployment of these electric commercial vehicles. To tackle these problems, an Energy Storage Boosted Super Charger with Data Driven Energy Management System is proposed in this project. The proposed system leverages the dependency of high-power grid infrastructure in super charge application and makes the supercharging is possible in any old buildings or facilities without any restriction on grid infrastructure. Moreover, management of stored energy in the cluster of battery boosted supercharger is especially essential. Without the smart and predictive demand management, charger clusters will not be able to provide energy to meet the charging demand peak hour.
In this project, an energy boosted super charger with data driven energy management system will be built. The proposed energy boosted super charger is designed as a single unit that comprises (i) a battery energy storage, (ii) high power super charger, (iii) data driven energy management system and (iv) data driven energy management system algorithm.
The battery system is used as the energy storage to support sudden high electricity demand and reserve the energy produced during demand valley into the battery when another peak comes. The energy management system will be developed to monitor the status of the super chargers, including the energy consumption, temperature, vehicle information, voltage and current, etc.. The system would also allow the monitoring of the battery status, including the battery SOC, battery health, voltage and current. These real-time data will be sent to the control system for analysis so as to allocate proper amount of power for super chargers. This system utilizes historical data analytics to optimize the charging and discharging of the energy storage devices. By using historical data, the system can predict periods of high and low demand and adjust the energy storage and usage accordingly. This ensures that the energy storage is used in the most efficient way, reducing the need for additional building power grid capacity.
The system architecture is as below:
1. Grid Connection:
- The super charger draws power from the electrical grid through a typically three-phase connection for high power levels.
2. Power Conversion:
- The rectifiers (subsystem) convert grid AC power to high-voltage DC power for battery storage and EV charging.
3. Energy Storage System (ESS):
- Consists of large-capacity battery storage using technologies like lithium-ion.
- Battery modules (subsystem) consist of individual battery cells grouped into modules.
- Battery Management System (BMS) (Subsystem) monitors battery module metrics, ensuring safe and balanced operation.
4. Power Management System:
- Manages electricity flow from the grid to the storage system and from there to the EVs.
- DC/DC Converters (Subsystem) may adjust voltage from the storage system for EV charging.
- Communication System (Subsystem) facilitates interaction between EV and supercharger for authentication and billing.
5. Charging Stations (Dispensers):
- Manages electricity flow from the grid to the storage system and from there to the EVs.
- DC/DC Converters (Subsystem) may adjust voltage from the storage system for EV charging.
- Communication System (Subsystem) facilitates interaction between EV and supercharger for authentication and billing.
6. Cooling System:
- Maintains optimal temperatures for power electronics and batteries during high-power charging.
7. Infrastructure Support:
- Physical structures like canopy and lighting, as well as payment systems and network connectivity, support the operation and user experience.
A battery-boosted EV super charger can charge EVs at a faster rate than would be possible if relying on the grid connection alone, especially during times of peak demand when the grid may not be able to supply sufficient power. By using a large battery system as an intermediary, these superchargers can offer consistent and rapid charging without requiring massive and costly upgrades to the grid infrastructure.
Apart from the hardware systems mentioned above, the data driven energy and charger cluster management cloud software system for battery-boosted EV super chargers is a sophisticated platform that integrates with the physical infrastructure to optimize the operation and performance of the charging network. This cloud-based system leverages data analytics, machine learning, and Internet of Things (IoT) technologies to manage energy resources efficiently, ensure charger availability, and provide a seamless user experience. The system architecture is as below:
1. Cloud Infrastructure:
- Hosted on servers provided by cloud service providers.
- Incorporates security layers with firewalls, encryption, access controls, and regular security audits.
2. Data Collection Layer:
- Gather real-time data from superchargers, energy storage systems, grid data, weather forecasts, and user inputs.
- Includes the Monitoring Subsystem to track charger status and performance metrics.
3. Communication Layer:
- Ensure secure, bidirectional communication using protocols like MQTT, HTTP/HTTPS, or WebSockets.
- Facilitates the Firmware and Software Update Subsystem for remote updates to charger firmware and software.
4. Data Processing and Analytics:
- Process and analyzes data using big data frameworks and machine learning.
- Predictive Analytics Subsystem predicts future charging demands and energy prices.
- Reporting and Analytics Subsystem provides detailed insights for stakeholders.
5. Optimization Engine:
- Use data analytics insights to optimize energy usage and manage charging load.
- Energy Management Subsystem manages the flow of energy efficiently.
- Demand Response Subsystem interacts with grid demand response programs.
6. User Interface and Application Layer:
- Provide mobile apps, web dashboards, and APIs for EV drivers, station operators, and utility companies.
- Billing and Transaction Management Subsystem handles payments and billing.
- Customer Relationship Management (CRM) Subsystem manages customer interactions and support.
7. Control Layer:
- Send commands to supercharger stations to control charging rates and battery storage state.
- Maintenance Management Subsystem schedules and manages service needs for the charging infrastructure.
- Integration Subsystem:
- Allow integration with renewable energy sources, vehicle telematics, and smart city infrastructure to create a cohesive ecosystem and extend functionalities.
The cloud management system ensures efficient operations, maximizes uptime, improves the user experience, and supports the scaling of EV charging infrastructure to meet the growing demand for electric vehicles. The result shall encourage higher adoption of super charger applications in the community. It will provide a technology solution to support Hong Kong Government’s policy and to catch the opportunity ahead.
The core innovation is original and emerges directly from a building-centric analysis of the EV charging problem. While the EV industry has focused primarily on roadside charging networks and purpose-built charging hubs, this project recognizes that the majority of EV charging will ultimately occur at buildings, including workplaces, residential towers, shopping malls, hotels, hospitals, and transport hubs.
4 Project Innovation
4.1 Integrated Technical Breakthroughs
The core innovation is original, emerging directly from a building-centric analysis of the EV charging challenge. This novelty is expressed through three integrated breakthroughs:
Building-Level Energy Storage as an Electrical Capacity Multiplier
Instead of upgrading a building’s incoming electrical supply, which is a costly, time-intensive, and often infeasible process, the project system integrates a 500 kWh battery energy storage unit within the building’s existing electrical room or parking structure. The battery charges gradually during periods of low building demand (typically overnight at 23:00 – 07:00, when building loads decline by 40–60%) and discharges rapidly to support EV charging during peak periods (typically morning at 08:00 – 10:00 and evening at 17:00 – 19:00). As a result, the building’s peak grid demand remains unchanged, even as its total energy services expand significantly. This application of energy storage as a means of multiplying effective electrical capacity, without physical infrastructure upgrades, represents a novel contribution to the building energy management.
AI-Driven Building Load Balancing
The machine learning system developed in this project is trained on both EV charging demand patterns and the building’s comprehensive load profile, including but not limited to HVAC schedules, elevator usage, lighting loads, tenant occupancy, seasonal variations, and weather data. The algorithm continuously forecasts total building demand over a 24-hour horizon and dynamically allocates available capacity headroom to EV charging. It coordinates battery charge and discharge cycles to ensure that the building’s contractual maximum demand is never exceeded. This whole-building optimisation approach advances beyond conventional EV smart charging, which typically addresses charger-level scheduling only, by integrating EV charging as a managed component within a holistic building energy budget.
Modular and Retrofittable Hardware Design
The battery storage and power conversion system comprises of cascadable 5 kWh modules at a nominal 480 VDC, with 100 modules used for the 500 kWh configuration. These are assembled in a configurable enclosure designed for installation in existing building electrical rooms, parking basements, or rooftop plant rooms. The system connects via a standard three-phase building supply and requires no transformer upgrades, no new utility service entrance, and no trenching or civil works, thereby enabling straightforward retrofitting in a wide range of existing buildings.
Example Case
A commercial office building operates with an incoming electrical capacity of 600 kW and experiences a typical peak demand of 420 kW, driven primarily by HVAC loads during summer afternoons. The building owner sought to install four DC fast chargers in the basement parking rated at 150 kW each, representing a total potential load of 600 kW, to attract and retain EV-driving tenants.
Under a conventional approach, this installation would necessitate upgrading the electrical service to 1,200 kW. Such an upgrade would require a transformer replacement costing approximately HKD10 million, entail 18 months of construction, and cause significant disruption to tenants.
With the proposed battery-integrated system, the machine learning algorithm identified that the building maintains approximately 180 kW of available capacity headroom for most of the day. This available capacity is used to slowly charge the 500 kWh battery. When multiple EVs arrive simultaneously, the battery discharges at up to 360 kW, supplementing the 180 kW drawn in real time from the grid. As a result, the building's total demand never exceeds its 600 kW grid supply.
4.2 Financial Model and Public-Private Partnership (PPP) Framework
The building-centric approach to EV charging infrastructure creates natural opportunities for publicprivate partnership (PPP) structures, enabling risk-sharing and value creation across multiple stakeholders.
Building Owners as Partners
Building owners benefit from enhanced property value and tenant attraction, as EV charging availability is increasingly becoming a competitive necessity in commercial leasing markets. By adopting the proposed system, owners can offer this amenity without incurring the prohibitive costs of traditional electrical upgrades. The technology provider can structure leasing or revenue-sharing arrangements under which the building owner contributes space and an electrical connection, the technology company supplies and operates the hardware and software, and charging revenues are shared between parties.
Government Incentives for Building Retrofits
Hong Kong's Buildings Energy Efficiency Ordinance (Cap. 610) mandates regular energy audits and efficiency improvements for commercial buildings. The integration of battery storage with EV charging infrastructure qualifies for various energy efficiency incentive schemes. Additionally, the demonstrated reduction in peak building demand can attract rebates through utility demand-side management programmes, further strengthening the financial case.
Utility Demand Response Participation
Both CLP Power and HK Electric operate demand response programmes that compensate building owners for reducing grid demand during peak periods. The battery system enables active participation in these programmes by shifting EV charging load to off-peak hours, thereby generating supplementary revenue streams that improve overall project economics.
Green Building Certification Uplift
The system contributes to credits under multiple green building rating schemes, including BEAM Plus, LEED, and WELL. Achievement of these certifications can increase a building's market value and support rental premiums, delivering additional long-term benefits to building owners.
4.3 Policy Alignment and Future Development Trends
The project system is positioned at the intersection of multiple emerging trends in building energy management, urban policy, and transportation technology, ensuring its long-term relevance and scalability across APEC economies.
Net Zero Building Standards
As building codes increasingly mandate net-zero energy performance, the integration of energy storage becomes essential for managing intermittent on-site renewable generation, such as rooftop solar, while simultaneously serving EV charging loads. The battery system serves a dual purpose for buffering building energy demand and enabling high-capacity EV charging, thereby supporting the transition toward netzero buildings without requiring grid capacity expansion.
Building-as-a-Grid-Asset
Future building energy systems will participate actively in electricity markets as distributed energy resources. The bi-directional power conversion hardware developed in this project positions buildings to provide multiple grid services beyond EV charging, including frequency regulation, peak shaving, and emergency backup power. This functionality transforms buildings from passive energy consumers into active grid assets capable of generating additional revenue streams.
Autonomous EV Fleet Charging
With the emergence of autonomous electric vehicles, buildings are poised to become primary charging destinations. Autonomous taxis and delivery vehicles will require charging during idle hours within building parking lots. The predictive AI algorithm is designed to accept fleet telemetry data, enabling buildings to serve as automated charging hubs that optimise charging schedules based on vehicle availability, electricity pricing, and building load constraints.
Mandatory EV-Ready Building Codes
Multiple APEC economies are implementing "EV-ready" building codes that require minimum charging provisions in new and renovated buildings. For existing buildings, the battery-boosted system offers the most practical pathway to compliance without undertaking costly and disruptive electrical overhauls.
5 Project Potential Future Inspiration
5.1 Inspiring Subsequent Projects
This project establishes a replicable model with the potential to inspire and inform a range of downstream initiatives across the Asia-Pacific region and beyond.
Building Retrofit Programmes
Many APEC economies countries face the challenge of aging building stock where conventional electrical upgrades are cost-prohibitive or logistically impractical. This project demonstrates that battery storage can serve as a virtual electrical upgrade for any building. This is an approach readily transferable to diverse building typologies across the region. The model offers a scalable pathway for accelerating EV charging deployment without awaiting major grid infrastructure investment.
District-Level Energy Management
The concept of using distributed building-level storage to manage aggregate district demand in this project has significant potential for scaling to smart district and smart city applications. A network of battery-equipped buildings could collectively absorb, store, and redistribute EV charging loads across a district, reducing strain on local distribution networks and enabling more efficient utilization of existing grid capacity.
Extension to Other Building Load Types
The battery-integration approach developed in this project is not limited to EV charging. The same methodology could be applied to other high-power building loads that strain existing electrical capacity, including commercial kitchen equipment, data center backup power, and industrial machinery in mixeduse buildings. This versatility enhances the project's replicability and long-term impact.
5.2 Domains Enlightenment
The project offers transformative insights across multiple domains and establishing new perspectives for integrating EV charging within the built environment.
Building Energy Management
The project demonstrates that building electrical capacity need not be regarded as a fixed physical constraint. Through the strategic integration of energy storage and intelligent load orchestration, capacity can be dynamically expanded to accommodate EV charging without requiring traditional infrastructure upgrades.
Property Development and Asset Management
For building owners and property developers, the project introduces a compelling new value proposition: existing buildings can achieve "EV-ready" status without electrical rewiring. This enhances asset competitiveness and future-proofs properties against evolving tenant expectations and regulatory requirements.
Utility Infrastructure Planning
From a utility perspective, the project illustrates that building-level energy storage can defer or eliminate the need for distribution network upgrades. By enabling buildings to maintain peak demand within existing supply limits, the approach reduces strain on grid infrastructure and supports more efficient capital allocation by utility operators.
Green Building Certification
The project introduces a novel pathway for achieving energy performance credits under green building rating schemes. By integrating EV charging management within a holistic building energy strategy, it enables certification outcomes that were previously difficult to achieve in existing buildings without major retrofits.
Urban Transport Electrification
The project reframes the narrative around urban transport electrification by demonstrating that buildings are the natural location for EV charging infrastructure. This insight removes the assumption that highcapacity supercharging requires dedicated roadside or off-site installations, unlocking a vastly larger addressable market for charging deployment within the existing urban structure.
6 Project Knowledge Sharing and Stakeholder Engagement
The project adopts a comprehensive approach to knowledge dissemination and stakeholder engagement, ensuring that findings, methodologies, and operational insights are widely accessible and capable of informing policy, practice, and further research across APEC economies and beyond.
Sharing Platform
Project information will be uploaded to organisation website for wider knowledge sharing and publicity. This open-access approach ensures that policymakers, building owners, and technology developers across the region can draw upon the project's experience when designing their own EV charging initiatives.
Building Operator Dashboard
To engage building occupants and the broader public, a real-time building operator dashboard could be displayed publicly within building lobbies. The dashboard shows live EV charging activity, battery state of charge, building energy balance, and cumulative carbon savings, making visible the building's sustainability performance and demonstrating its commitment to environmental responsibility to tenants, visitors, and passers-by.
Stakeholder Engagement
The project team maintains regular consultations with key government and industry stakeholders, including Electrical and Mechanical Services Department, the Building Authority, and utility companies CLP Power and Hong Kong Electric. These engagements ensure alignment with evolving building codes, grid operation requirements, and utility programme design, while also providing feedback channels to inform regulatory development.
Comparison with Similar Approaches
Table 1 Comparison with Similar Approaches
7 Project Measurement
7.1 Actual Implementation
The project is an ongoing initiative with a duration of 24 months.
Phase 1 — Building Integration Hardware Development:
• Development of 500 kWh battery storage system optimized for building plant room installation(compact footprint: 2.5 m × 1.5 m × 2.0 m; weight-bearing assessment for typical building floor loads)
• Bi-directional AC/DC and DC/DC power conversion modules for dual-source operation (building grid + battery)
• Multi-standard DC charging dispensers (CCS2 and GB/T) designed for building parking structure environments
• Configurable, cascadable mechanical casing suitable for basement, ground floor, or rooftop installation
• Thermal management system designed for enclosed building plant room environments (limited ventilation)
• Battery Management System (BMS) with real-time SOC, temperature, voltage, and current monitoring • Integration interface with building management systems (BMS/BEMS) via BACnet and Modbus protocols
Phase 2 — Building Energy Management Software Development:
• Cloud infrastructure on containerized microservices architecture (ECS Fargate, DynamoDB, Lambda, API Gateway, CloudFront, Cognito, QLDB)
• OCPP 2.0.1 connection handler for EV charger communication
• Building load monitoring integration via smart meter data feeds (1-minute interval building demand data)
• Whole-building demand prediction ML model trained on: building HVAC load profiles, occupancy patterns, elevator usage, lighting schedules, weather data (temperature, humidity, solar irradiance), EV charging demand history, time-of-day/week/season, and special event calendars
• Dynamic capacity allocation engine that calculates real-time available headroom (building supply limit minus current base load) and allocates it between EV chargers and battery recharging
• Building operator web dashboard showing: real-time building demand vs. supply limit, EV charger status, battery SOC, predicted demand for next 24 hours, historical trends, and financial analytics
• Tenant/driver mobile application with payment, reservation, real-time charging status, and building energy dashboard access
• Billing and transaction management with immutable ledger (QLDB)
Phase 3 — Building Integration Testing:
• Hardware-software integration testing in simulated building environment
• ML model validation (target: whole-building demand prediction MAPE < 12%; EV charging demand MAPE < 15%)
• Critical test: demonstrate that building meter never exceeds supply limit during maximum EV charging conditions
• Safety certification (IEC 61851, IEC 62196, IEC 62619)
• Building code compliance review (Hong Kong Building Authority, EMSD)
• Integration testing with building management system (HVAC coordination: if building demand approaches limit, HVAC setpoint adjusts by 1–2°C before EV charging is curtailed)
Phase 4 — Pilot Building Deployment:
• Installation in selected Hong Kong commercial building parking structure
• 6-month operational pilot with full data collection
• Model retraining with live building-specific data
• Performance reporting: building peak demand maintained within supply limit, EV vehicles served, energy shifted, cost savings, carbon reduction
7.2 Project Target Goal
Table 2 Project Numerical Goal
8 International Applicability
The system is specifically designed for maximum building-sector replicability within APEC economies.
Universal Building Integration
The system connects to any three-phase building electrical supply operating between 200V and 480V at 50 or 60 Hz through standard switchgear, requiring no building-specific customization for the electrical interface. A BACnet and Modbus integration layer enables communication with all major building management systems, ensuring seamless interoperability with existing building infrastructure.
Building Type Versatility
The modular, cascadable design allows system sizing from 100 kWh suitable for small residential buildings with two chargers to 2,000 kWh appropriate for large commercial complexes supporting twenty or more chargers. The machine learning model automatically retrains on each building's specific load profile within four to eight weeks of installation, ensuring optimal performance across diverse building types and usage patterns.
Regulatory Adaptability
All charging hardware complies with applicable IEC international standards. The cloud platform supports multi-jurisdictional billing, taxation, and regulatory reporting requirements, enabling deployment across multiple economies without bespoke software development. Data sovereignty concerns are addressed through flexible deployment options, including cloud region selection or fully on-premises installation.
Language and Market Localization
The mobile application supports multilingual interfaces to serve diverse user populations. Charging hardware supports the full range of EV connector standards deployed across APEC markets, including CCS2 for Australia, New Zealand, Korea, and Singapore; GB/T for China; and CHAdeMO for Japan, ensuring compatibility with all major EV fleets in the region.
9 System Cost Effectiveness Evaluation
The project could potentially influence the production or application costs of EV chargers, making them more competitive compared to existing solutions.
Cost Reduction:
1. Economies of Scale: As the adoption of energy storage boosted superchargers increases, economies of scale could reduce the production costs of both the chargers and the associated battery storage systems.
2. Optimized Energy Consumption: Data-driven energy management allows for intelligent charging during off-peak hours when electricity rates are lower. This can reduce operational costs and make the overall application of these chargers more cost-effective.
3. Reduced Peak Demand Charges: Energy storage can help mitigate peak demand charges from utilities by storing energy when demand and prices are low and releasing it during peak times. This can lead to significant savings on electricity bills.
4. Efficient Resource Allocation: By using predictive analytics and real-time monitoring, the system can optimize resource allocation, potentially reducing the need for overbuilding infrastructure to meet peak demand and thus lowering capital expenditure.
5. Maintenance Cost Savings: Predictive maintenance strategies enabled by the data-driven system can help prevent costly repairs and downtime, thus reducing the long-term maintenance costs of the chargers.
Balancing Cost with Value:
1. Value-Added Services: The advanced capabilities of these superchargers may allow operators to offer premium services, justifying a higher price point or creating new revenue streams, such as selling stored energy back to the grid during peak demand.
2. Long-Term Savings: While the initial costs might be higher, the long-term operational savings due to increased efficiency and reliability can make the total cost of ownership more competitive over time.
3. Incentives and Subsidies: Government incentives for renewable energy and advanced infrastructure projects could offset some of the initial costs and promote quicker adoption.
9.1 Cost-Effective Implementation
Building-Level Capital Cost Comparison
A comparative cost analysis between the traditional grid upgrade approach and the battery-integrated system developed in this project demonstrates the significant economic advantages of the proposed solution. Under the traditional approach, installing four DC fast chargers requires a substantial upfront investment for charging hardware, accompanied by major electrical infrastructure upgrades including building transformer replacement, utility service upgrades, and extensive cabling and switchgear work. Civil works and tenant disruption add further costs, along with estimated disruption-related expenses for building occupants. The traditional approach also incurs ongoing cloud management fees and requires an extended construction period spanning many months.
By contrast, the battery-integrated system involves higher upfront costs for charging hardware and the battery energy storage system, as well as moderately higher cloud management fees. However, it entirely eliminates the need for building transformer upgrades and utility service upgrades, while substantially reducing cabling, switchgear, civil works, and disruption-related expenses. Tenant disruption is minimized, and construction time is compressed to a matter of weeks rather than months or years.
The total capital required for the traditional approach is several times higher than that of the batteryintegrated system. This represents significant savings to the building owner, reducing the capital outlay required under conventional approaches by a substantial margin.
Table 3 Cost Comparison between Traditional Approach and Battery-Integrated System
Building-Level Annual Operating Benefits
The selling price of system shall be subject to the scale of the charging station. For every charging, the estimated charging price is HK$3.8/kWh, implying the profit is HK$2.3/kWh. Given that the estimated charging time per EV is 0.6 hour, assuming 8 operation hours per EV per day, the annual revenue for each charger will be HK$2.3/kWh x 8 hours x 60 kW x 365 days =~ HK$400K. The price excludes the power and communication cable laying cost, which shall be subject to distance between the charger and the distribution board (power supply for the charging station) and the complexity of the carpark layout.
Table 4 Benefit Comparison between Traditional Approach and Battery-Integrated System
9.2 Environmental Impact Estimation
• Measurement Method
The cloud management system continuously logs:
- Building total demand at 1-minute intervals (smart meter data)
- EV charging energy delivered per session (kWh), with source breakdown (grid or battery)
- Battery charge/discharge cycles with time-of-use tariff mapping
- Building peak demand events (count of times building demand exceeds 95% of supply limit)
- Grid carbon intensity factor at time of energy consumption (utility data)
• Estimated Annual Building-Level Impacts (per installation)
Table 5 Estimated Annual Building-Level Impacts
10 Consistency and Long-Term Sustainability
10.1 Alignment with Building Energy Policy
The project demonstrates strong alignment with Hong Kong's policy framework for building energy efficiency, climate action, and smart city development, ensuring its relevance and support within the local regulatory environment.
Buildings Energy Efficiency Ordinance
Under the Buildings Energy Efficiency Ordinance (Cap. 610), commercial buildings are required to conduct regular energy audits and implement efficiency improvements. The system's whole-building energy optimization approach, which includes coordination of HVAC loads during EV charging peaks, directly contributes to improved building energy efficiency performance and supports compliance with ordinance requirements.
Hong Kong Climate Action Plan 2050
The project simultaneously addresses two of the four primary decarbonization pillars outlined in the Hong Kong Climate Action Plan 2050: building energy efficiency and transport electrification. By enabling highcapacity EV charging within existing buildings without grid upgrades, the system bridges these policy objectives and contributes to the city's carbon neutrality target.
Hong Kong Smart City Blueprint 2.0
The Hong Kong Smart City Blueprint 2.0 emphasizes IoT-enabled building management, data-driven urban infrastructure, and expansion of EV charging networks—all core features of this project. The system's integration of machine learning, real-time monitoring, and automated load management positions it as a demonstration of smart city principles in practice.
Hong Kong Policy Address 2023
The 2023 Policy Address commits to deploying 3,000 e-taxis, a target that depends on the availability of charging infrastructure where taxis park and charge. By enabling supercharging inside existing buildings, this project directly supports the practical realization of the e-taxi deployment commitment.
Others
At the regional level, the project contributes to the APEC ESCI Low Energy Building Network goals by demonstrating how buildings can absorb major new electrical loads such as EV charging without increasing peak grid demand, and by sharing knowledge and best practices in low-energy building operations with APEC member economies.
The system also earns credits under the BEAM Plus green building standard, contributing to categories including Energy Use, Innovations and Additions, and Site Aspects through sustainable transport provisions, thereby enhancing the green building certification outcomes for participating properties.
10.2 Long-Term Operational Sustainability
The project is designed for sustained performance over a minimum ten-year operational horizon, with technical and commercial structures in place to ensure ongoing reliability, adaptability, and value delivery.
Long-Term Operational Design Life
The battery system is rated for a minimum of 4,000 charge-discharge cycles, while power electronics are specified with a mean time between failures of fifteen years. The cascadable modular design allows individual battery modules to be replaced at end of life without requiring system-wide replacement, extending overall system longevity and reducing life-cycle costs. The cloud platform is continuously updated to maintain security, functionality, and compatibility with evolving standards.
Continuous Machine Learning Model Improvement
The machine learning model is retrained quarterly using updated building and charging data, enabling continuous improvement in prediction accuracy as the system accumulates operational patterns specific to each building. This adaptive capability ensures that load forecasting and energy optimization become more refined and effective over time.
Building Portfolio Expansion
For property groups with multiple buildings, the cloud platform supports centralized management of distributed systems from a single dashboard, enabling portfolio-wide deployment with consistent oversight, reporting, and optimization. This scalability enhances the commercial viability of the solution for large property owners and managers.
Project Continuous Implementation
The implementation organization of this project will be responsible for maintaining the dedicated hardware maintenance, software operations, and customer support, with long-term service agreements offered to building owners to ensure sustained system performance and peace of mind throughout the operational life of the installation.
11 Project Results and Achievement
The project has achieved all key technical milestones necessary to demonstrate the feasibility and readiness of the integrated battery storage and EV charging system.
A 500 kWh battery storage system has been assembled and tested at its rated voltage of 480 VDC, with its compact footprint of 2.4 m x 1.4 m x 1.9 m validated for installation within building plant rooms. The bi-directional power conversion modules have been demonstrated at 360 kW, confirming their capacity to support high-power EV charging applications.
Five operating modes have been validated, providing comprehensive operational flexibility. It includes (1) hybrid mode with grid and battery operation for heavy-duty vehicle charging; (2) grid-only operation for medium-rate charging; (3) battery-only operation enabling supercharging during building peak hours; (4) battery recharging during building off-peak hours; and (5) combined mode with grid plus battery operation during building off-peak hours. This range of modes ensures that the system can adapt to varying building load conditions, grid constraints, and charging demand patterns.
The machine learning model has been trained on 12 months of energy data from 10 Hong Kong commercial buildings, supplemented by EV charging demand data. The cloud platform has been deployed with full support for OCPP 2.0.1 and includes comprehensive building energy monitoring dashboards. A mobile application has been completed, offering users payment processing, reservation functionality, real-time building energy visualization, and tracking of carbon savings associated with their charging sessions.
12 Data Collection and Verification
12.1 Data Collected
Comprehensive data collection has been undertaken to validate system performance, model accuracy, and economic outcomes across multiple dimensions. Building energy data is captured at one-minute intervals, providing continuous demand profiles that enable comparison between pre-installation baseline conditions and post-installation operations. This generates data points per building, forming a robust basis for performance analysis.
EV charging transaction data is logged for every session, including timestamp, duration, energy delivered, peak power, energy source ratio between grid and battery, connector type, and payment amount. Test transactions have been recorded, providing a substantial dataset for validating charging behaviour and revenue modelling.
Battery performance data is captured at one-second intervals, including state of charge, voltage, current, and temperature, with complete charge-discharge cycle logs enabling detailed analysis of battery efficiency, degradation, and operational patterns. Machine learning model performance logs record hourly predictions against actual demand, with rolling mean absolute percentage error calculations that track model accuracy over time and support continuous algorithm refinement.
Building demand profile comparisons are presented as time-series overlays illustrating three situations: building demand prior to EV charging installation, projected demand had EV charging been installed without battery integration, and actual demand with the battery-integrated system. These visualisations demonstrate that peak demand remains consistently within the building's existing supply limit despite the addition of high-power EV charging loads.
12.2 Independent Verification
All performance claims are supported by independent verification from multiple sources. Battery system capacity, cycle life, and thermal performance will be validated through laboratory testing. Building peak demand data is verified against utility billing records from electricity companies, using monthly maximum demand charges as the reference standard.
The machine learning model has undergone standard time-series cross-validation. The system complies with relevant international standards, including IEC 61851 for electric vehicle conductive charging systems, IEC 62196 for plugs and sockets, and IEC 62619 for industrial lithium-ion batteries.
13 Project Impact
13.1 Transformative Impact on Building Energy Management
This project introduces a fundamental shift in the paradigm of building energy management, with implications that extend well beyond the immediate application of EV charging.
Redefines Building Electrical Capacity
Traditionally, a building's ability to serve additional loads is capped by its fixed incoming electrical supply. This project demonstrates that energy storage with intelligence can dynamically expand effective capacity without physical infrastructure changes, which is a concept applicable far beyond EV charging.
Turns Buildings into Grid Assets
By shifting EV charging demand from peak to off-peak, the building reduces overall grid stress. At scale, hundreds of battery-equipped buildings collectively function as a distributed energy resource, deferring billions of dollars in utility infrastructure investment.
Accelerates Building Decarbonization
The integration of EV charging into building energy management creates a single, holistic platform for managing all building energy flows, including future integration of on-site solar PV, building-level battery storage for HVAC optimization, and demand response participation.
Eliminates Infrastructure Barriers to EV Adoption
For building owners and tenants, the inability to install fast EV chargers has been a major barrier to EV adoption. By removing the grid upgrade requirement, the system makes EV charging as simple to add as any other building amenity.
13.2 Impact Across Multiple Operational Areas
The project delivers measurable impact across eight distinct operational areas:
Building Operations
Whole-building energy optimization reduces peak demand charges and improves overall energy cost performance, delivering direct financial benefits to building owners and occupiers.
Property Management
Buildings certified as EV-ready command higher rental premiums and achieve improved tenant retention, as EV charging availability becomes an increasingly important consideration in commercial leasing decisions.
Transport Electrification
By demonstrating that buildings can serve as the primary locus of EV charging infrastructure deployment, the project unlocks a vastly larger addressable market for charging installation than roadside or purposebuilt facilities alone can provide.
Utility Infrastructure
Building-level energy storage defers or eliminates the need for distribution network upgrades, enabling more efficient allocation of utility capital expenditure and reducing strain on grid infrastructure.
Green Building Certification
The system introduces a new pathway for earning sustainability credits under green building rating schemes, enhancing the certification outcomes for participating properties.
Urban Planning
By enabling EV charging within existing buildings, the project eliminates the need for standalone charging facilities, reducing land use requirements and preserving scarce urban space for other purposes.
Environmental
The system reduces carbon dioxide emissions from both transport electrification and optimized building energy operations, contributing simultaneously to two critical decarbonization objectives.
Technology
The project advances the development of integrated building-transport energy management systems, creating technological capabilities that can be applied to a wide range of future smart building and smart city applications.
14 Social Responsibility
14.1 Women Empowerment and Equality
The project involve female as a key technical and management roles including project manager, software manager and mechanical engineer. Meanwhile, the project is designed to ensure charging services are available to all building occupants, tenants, and visitors on a non-discriminatory basis via mobile app reservation or walk-up access.
14.2 Just Transition and Community Inclusiveness
The project is designed to ensure that the benefits of EV charging infrastructure are distributed equitably across communities, supporting a just transition to electric mobility while delivering tangible social and economic co-benefits. The battery-boosted EV charging system is suitable to be deployed in any buildings to ensure all resident can access EV charging infrastructure.
The project's collaborative structure, involving government, property owner, property management companies, electricity company and charging solution providers, facilitates knowledge sharing, drives innovation, and builds capacity across participating entities. This multi-stakeholder partnership contributes to organisational resilience by fostering strategic alliances, advancing research and development, and deepening technical expertise in the renewable energy, energy storage, and smart grid sectors.
Estimated Savings
Savings Narrative
The project generates building-level savings across four dimensions: (1) capital cost avoidance from eliminated building electrical upgrades, (2) operational savings from energy arbitrage and peak demand charge reduction, (3) new revenue from EV charging services and demand re-sponse programs, and (4) societal savings from reduced carbon emissions.
Capital savings are calculated by comparing contractor and utility quotes for a conventional building transformer upgrade and utility service capacity increase (CLP Power / HK Electric) against the all-in cost of the battery-integrated system. For a typical installation with four 150 kW DC fast chargers, the traditional grid upgrade path costs HK$10,700,000–19,200,000, while the battery-integrated system costs HK$4,800,000—a saving of HK$5,900,000–14,400,000 (55–75%).
Operational savings derive from energy arbitrage and peak demand charge reduction. The 500 kWh battery charges during off-peak tariff periods and discharges to serve EV charging during peak tariff periods. The off-peak versus peak electricity tariff differential is applied to the daily energy shifted (approximately 500 kWh/day) over 365 days. Peak demand charge reduction is calculated using the building's utility tariff structure; CLP Power's maximum demand charge is HK$52.4/kW/month, and the system shaves up to 360 kW from the building's peak demand registration. Additional revenue comes from participation in utility demand response programs and net EV charging service fees.
Carbon savings are calculated using the Hong Kong grid carbon intensity (0.71 kg CO₂/kWh, EMSD 2023) applied to the peak-versus-off-peak carbon intensity differential multiplied by energy shifted, plus ICE vehicle displacement savings at 0.21 kg CO₂/km for LPG taxis. At 100-building deployment scale, the project achieves 4,030 tonnes CO₂/year in ongoing reductions plus 5,000 tonnes of one-time embodied carbon avoided from eliminated grid infrastructure upgrades.
Net Operating Cost Savings per year: $164,000 USD
(HK$1,280,000: comprising energy arbitrage HK$425,000 + building peak demand charge reduction HK$350,000 + demand response revenue HK$115,000 + maintenance savings vs. grid upgrade HK$390,000)
Kilowatt-hours Savings per year: 182,500 kWh
(Peak-to-off-peak energy shifting, reducing building peak generation requirements; additional ~18,250 kWh saved through ML-optimized building-charger load coordination vs. unmanaged charging)
Carbon Emissions Offset: 40.3 Metric tons CO₂ per year
(Per building installation. 21.9 tonnes from peak-shift carbon intensity differential + 18.4 tonnes from ICE vehicle displacement. At 100-building scale: 4,030 tonnes CO₂/year. Additional 5,000 tonnes one-time embodied carbon avoided from eliminated grid upgrades.)