Energy Auditing, Green House Gas and Carbon Footprint reduction are strategic issues challenging every CEO. Moreover are in many countries mandatory elements to be reported in the financial statements of listed companies. The activity of ‘reporting & disclosure’ provides indicators of environmentally sustainable development of an enterprise and communicate its organizational capacity in addressing climate change. Participation in the CDP questionnaire means visibility to institutional investors  that are paying more and more attention to the level of awareness and the mode of management of risks related to the emissions of a company. Quantum is here to help in gathering the needed data and filling up wisely the information requested.

We help our customers to participate in the activities of international benchmarking and define a concrete emission reduction. We offer strategic advice in the identification, analysis, measurement and reporting of emissions in accordance with the guidelines of the Green House Gas Protocol (scope 1, 2 and 3) preparing the drafting of questionnaires developed by GRI (Global Reporting Initiative) and CDP (Carbon Disclosure Project).

The EARP (Enhanced Automated Reporting Platform) software embeds IPMVP (International Performance Measurement Validation Protocol) protocol for the automated baseline adjustment and energy pattern recognition through Artificial Intelligence enabling ECMs behavior prediction and Smart Alerts.
The EARP is not just an energy-monitoring dashboard. To the contrary it is a full turn-key solution that collects and analyzes energy data and any other factors providing correlations and predictions about the behavior of the systems. The EARP is as well an analytics environment where IPMVP industry standard protocol is used to assess and track adjustments to the baseline to ensure that savings are always benchmarked to the current situation which dynamically evolves due to non-energy directly related factors. The EARP platform makes use of a big data architecture that enables the analysis of massive data sets for instant replies to complex queries supporting the energy managers in their task of optimizing the ESMs performances over time.

The application reduces the costs and efforts to perform the mandatory M&V (Measurement and Verification’s) in an EPC (Energy Performance Contract) and conforms to the requirements of the ISO 50001 FOR THE ON LINE ENERGY DATA GATHERING.

Below is a synthetic description on the platform architecture and data workflow:

Platform Architecture

Before the data is presented to the client in the energy dashboard it goes through several steps as detailed below:

  1. Quantum Data Collection Engine: where the data will be collected from the different units at site and sent to a common gateway from which it will be sent to the Data warehouse to be analyzed.
  2. Data Warehouse: the data collected from the site will be stored in the cloud where it shall be analyzed.
  3. Energy Analytics: it’s the dashboard where the analyzed data will be represented to the client and the M&V engineers.

Data Collection

  • The cloud based data center is connected to the client site using secure VPN communications.
  • Data is collected real time from the customer’s premises. This includes indicators like electricity consumed and generated, water, gas, etc.
    Additional customers’ data can be collected and uploaded to the cloud. Examples are building occupancy, working shifts, production capacity, manufacturing output per day/per hour, etc.
  • Environmental data like temperature, humidity, solar irradiation as well as the weather conditions can also be pulled into the platform.
  • The platform is meter agnostic and can work with our turn-key physical metering solutions or solutions from any vendor.

Data Processing & Reporting

1. Pre-Processing

  • Data is normalized and cleaned in the preprocessing phase.
  • Meter raw data is always preserved for transparency and available to the customer.
  • Data is preprocessed to compare measurements to base line values to obtain energy saving metrics.

2. Base Line Adjustment

  • The EARP platform implements IPMVP protocol for baseline adjustment. A dedicated IPMVP data entry menu enables the registration of events that have an impact on the agreed baseline, like additional loads, change of occupancy or working hours, etc. The customer is notified of the baseline change and has to approve the baseline adjustment introduced.

3. ESM behavior prediction

  • The EARP platform performs Artificial Intelligence features based on BDL (Bidirectional Link) and a “knob-less” data selection approach. The platform, according to the selected ESM, is fed with data coming from meters and sensors and self-select the data input which proves to provide the best accuracy of its embedded prediction algorithm and model. References from different projects implemented by QuantumEsco are used to benchmark the accuracy of the prediction.

4. Reporting

  • The reporting engine delivers rich data visualization allowing clients and energy experts to quickly build customized visualization thanks to the widget editor, which simplifies the way of analyzing and correlating complex data.
  • Different reports are made available to the client depending on client’s custom requirements as well as automated M&V reporting
  • Dedicated Widgets helps energy engineers to building meaningful visualizations without having any specific IT competencies.

5. Smart Alerting

  • The alerts engine monitors the expected behavior of the ESM and based on the severity of the deviation, reacts with a predetermined, user defined, risk assessment and escalation protocol.

6. AI features

  • The platform has a dedicated AI menu interface showing the accuracy of the predictions and offering different settings from the full AI automated features selection to manual settings in the training menu where a new system without historical data could be “trained” by suggested human inputs;
    The Platform has a “knob less approach” where the features and parameters selection is automated and performed by the system itself based on the automatic training on historical data.