An ESCO in the Philippines signed a performance contract with a 14-story commercial tower in Makati. The contract guaranteed 18% energy savings over a 7-year term, with payments tied to measured savings. The measurement and verification (M&V) plan required sub-metering of 127 end-uses — AHUs, chillers, lighting panels, elevator banks, and IT loads — with data granularity at 15-minute intervals and a measurement uncertainty below 5%.

The energy efficiency terminal is the device that makes contracts like this bankable. It is not a meter in the traditional sense — it is a data acquisition and control platform designed specifically for energy management applications where the goal is not billing, but insight and control.

Energy Consumption Analytics

The terminal's core function is multi-circuit energy data acquisition. A single unit monitors up to 60 single-phase circuits or 20 three-phase circuits through external CT inputs (5A secondary, with CT ratios programmable from 5/5A to 10,000/5A). The sampling architecture uses a 24-bit sigma-delta ADC per channel, synchronized across all channels with a jitter below 10 µs — critical for computing cross-circuit quantities like total building power factor or generator vs. grid power split.

The analytics engine runs onboard and in the cloud. Onboard, the terminal computes:

- Per-circuit active/reactive/apparent power and energy (import/export)

- Power factor per circuit and aggregated

- Total harmonic distortion (THD) per voltage and current channel

- Voltage/current unbalance ratios

- Demand (sliding window and fixed interval, with timestamp of maximum demand)

In the cloud, the analytics layer adds:

- Baseline modeling using change-point regression (for cooling-dominated buildings) or multi-variable linear regression (for mixed-load buildings)

- Savings quantification using IPMVP Option C (whole-facility M&V)

- Anomaly detection using statistical process control (SPC) charts

- Load disaggregation using non-intrusive load monitoring (NILM) algorithms for circuits where sub-metering is not economically justified

A hospital in Ho Chi Minh City installed 8 terminals covering 340 circuits. The analytics identified that the chilled water pumps were operating at constant speed despite 40-70% part-load conditions for 18 hours per day. Installing VFDs (variable frequency drives) on the pumps, with the terminal providing the load profile justification, yielded a 22% reduction in pump energy consumption — equivalent to USD 18,400 in annual savings.

Demand-side Management Integration

The terminal is not a read-only device. It includes 12 digital output channels (relay contacts, 5A resistive load) and 8 digital input channels (dry contact or 12-24V DC). This I/O capability enables direct demand-side management (DSM) without a separate building automation system.

The DSM logic runs in the terminal's firmware and can be configured through the cloud platform or locally via the web interface. Typical DSM strategies include:

Peak demand limiting. The terminal monitors total building demand in real time. When demand approaches a user-configurable threshold (e.g., 85% of the utility's demand charge threshold), the terminal cycles non-essential loads — water heaters, EV chargers, pool pumps — according to a priority list. A shopping mall in Jakarta reduced its monthly demand charge by 14% using this strategy, saving approximately IDR 42 million (USD 2,800) per month.

Load shedding on utility price signal. In markets with time-of-use (TOU) tariffs or real-time pricing, the terminal receives price signals via API and pre-cools or pre-heats the building, or shifts deferrable loads to off-peak periods. A cold storage facility in Durban reduced its energy cost by 19% by shifting 340 kWh of daily consumption from peak to off-peak periods, using the terminal's price-signal integration with the local utility's API.

Generator import/export management. For facilities with onsite generation (diesel gensets or solar PV), the terminal can control the import/export setpoints by modulating controllable loads or, where available, modulating the generator governor setpoint via a 4-20 mA output. This is particularly valuable in regions with unreliable grids, where generator fuel consumption optimization has a direct impact on operating costs.

The DSM configuration is saved to non-volatile memory and continues operating during communication outages. The terminal's real-time clock, maintained by a supercapacitor for 72 hours, ensures TOU schedules and demand windows remain synchronized.

Cost-saving Case Examples

The business case for energy efficiency terminals is best demonstrated through deployed examples:

Case 1: Textile factory, Bangladesh. 4 terminals monitoring 180 circuits (spinning, weaving, dyeing, finishing). Identified 23% of energy consumption occurring during non-production hours due to compressed air leaks and lighting left on in unoccupied areas. Installed automated shutoff schedules through the terminal's digital outputs. Annual savings: BDT 3.2 crore (USD 37,600). Payback period: 14 months.

Case 2: University campus, Kenya. 12 terminals across 8 buildings. The analytics identified that the central chiller plant was producing chilled water at 6.5°C while the building thermostats were set to 24°C — a supply-air temperature reset opportunity. Adjusting the chiller setpoint to 8.5°C reduced chiller energy by 11% with no comfort impact. Annual savings: KES 8.4 million (USD 58,800). Payback period: 9 months (energy savings only; did not include demand-side participation incentives).

Case 3: Data center, Malaysia. 6 terminals monitoring IT load, cooling, and power distribution losses. The terminal's per-circuit monitoring identified that 14% of the facility's power was consumed by cooling systems operating to compensate for hot-aisle/cold-aisle containment breaches. Sealing the breaches reduced PUE (Power Usage Effectiveness) from 1.72 to 1.58. Annual savings: MYR 1.14 million (USD 257,000).

Compatibility with Existing Infrastructure

Retrofitting energy monitoring into an existing facility typically faces two constraints: (1) the existing switchboard does not have space for additional meters, and (2) running new CT wiring to a central meter location is prohibitively expensive.

The energy efficiency terminal addresses both. The terminal is DIN-rail mountable in a separate enclosure adjacent to the switchboard — not inside it. The CTs are split-core type (available up to 600A primary current), which can be installed on existing cables without disconnecting the load. The CT secondary wiring (5A) runs to the terminal enclosure through conduit, with typical wire runs under 30 meters.

Communication with the existing BMS or SCADA uses protocols already present in the facility. The terminal supports Modbus RTU (RS-485), Modbus TCP (Ethernet), BACnet MS/TP, and BACnet/IP. For facilities without a BMS, the terminal pushes data to our cloud platform, which provides a RESTful API for integration with the facility's own energy management software.

The CTs are class 1.0 accuracy (IEC 61869-2), with a phase error below 1° across the 5% to 120% current range. The terminal's input circuitry adds less than 0.2° of additional phase error, resulting in a combined current measurement accuracy better than 1.5% — sufficient for energy management applications (IEC 61557-12 Class 1 accuracy for power measurement).

For new construction, the terminal can be specified with Rogowski coil inputs instead of CT inputs. Rogowski coils provide a voltage output proportional to di/dt, eliminating the burden and saturation limitations of traditional CTs. They are particularly useful for large facilities where CT secondary runs would exceed 30 meters, as the Rogowski integrator can be located up to 100 meters from the coil.