Articles

Driving Value from IoT Energy and Environmental Sensors

IoT ThumbIntelligent devices and context-aware sensors will soon enable the Internet of Things (IoT) to monitor, measure, manage and protect personal and corporate assets; from people, vehicles, and buildings; to networks, smart grids, infrastructure and equipment.  For early adopters, IoT sensors provide quantifiable benefits including substantial environmental, safety, and energy savings.

As an introduction to the IoT it may help to illustrate how data collected from energy and environmental sensors provides value and insight. Green Econometrics has worked with several wireless sensor companies.  The following article reviews some of the features and benefits of IoT sensors in commercial applications.  Energy and environmental sensors and secure wireless network infrastructure enable remote monitoring, measuring, and managing of assets.  IoT sensors along with data collection and analytics provide value and benefits in terms of energy and costs reductions, risk mitigation, and life and asset protection.

We begin with defining objectives and selection of appropriate sensors.  Our objective is improving energy efficiency and indoor air quality.

  • Select sensors and capture data values
  • Compile the data in a useable format
  • Customize a dashboard using visuals that convey understanding and insight
  • Make the data and visual analysis accessible

The following illustrates the framework to collect sensor data and provide services such as analytics.

F1

Figure 1 IoT Sensor Framework

There are hundreds of sensors available today and certainly more to follow.   The trends we see are falling prices and ease of deployment.  The following table provides an overview of some of the sensors used in energy and environmental monitoring.

Sensor Selection

Sensor

Application Benefit
Carbon Dioxide CO2 Demand control ventilation Air quality/Energy savings
Carbon Monoxide CO Indoor monitoring Safety
Temperature/Humidity Indoor monitoring Comfort
Current transducers Energy usage Energy efficiency
Motion sensors Control lighting Energy savings
Light level meters Control lighting Energy savings

The IoT is comprised of connected devices, wireless sensors, and controllers that enable device control and data collection.  Battery operated wireless sensors are economical and easy to deploy.  To demonstrate, we have deployed group of sensors including motion detectors, light level meters, temperature, humidity, and current transducers (CTs) to measure current from a circuit on the power panel and battery level on solar PV charging system.  The small wireless ZigBee (IEEE 802.15.4) sensors are arranged in a mesh network and connect with an Internet Protocol (IP) gateway to the Internet.  Z-Wave sensors also employ a wireless protocol operating at a frequency 900 MHz range and are supported by a large number of vendors.

Our energy and environmental sensors are configured and calibrated to collect data every 15 minutes and relay the data, timestamp, value and ID, to a cloud-based storage. Data storage can consist of virtually any database.  We have used Rackspace with MySQL and MongoDB as well as Amazon Web Services.

Figure 2 Energy and Environmental Sensors in Facility

Figure 2 Energy and Environmental Sensors in Facility

The wireless sensors are small, easy to deploy, and in most cases self-honing to the data aggregation gateway.  The following figure 3 shows the size of the wireless sensors in comparison to a US quarter.

Figure 3 Typical Energy and Environmental IoT Sensors

Figure 3 Typical Energy and Environmental IoT Sensors

Data Management and Curating

Data needs curating.  Applications written in Python, R, and MATLAB can be used to refine and manage data extraction and compiling.  More complex data processing may require Extract, Transform and Load (ETL) systems to collect data from various sources; and then to process and transform the data into a useable format and load it into a database. We have found platforms such as Platfora, Pentaho, and Tableau offer great tools for seamless data extraction and visual analysis.

There are also applications designed to manage the data reporting and notifications. Data reporting systems can be configured to report alerts, alarms, and faults.  Alerts can be notifications sent via email or text messaging regarding sensor values above or below designated threshold levels.  Alarms and warnings can be established for life threatening events such as smoke, fire, carbon monoxide, or methane detection.

Sensor data services should include two primary functions: 1) reporting of alerts, alarms, and faults; and 2) data analysis.  The sensor data services should include dashboard rendering either as a cloud-based service or data directed towards an enterprise server.  Analytics should also be part of the service.

We are suggesting that the energy and environmental sensors supplement existing building alarm systems.  Redundancy is key.  In dealing with false positive and false negatives, it is important to devise support systems.  A false alarm for a fire is aggravating but a false negative such as a CO sensor not detecting a build up of carbon monoxide can be fatal.  Therefore, sensor calibration and periodic system checks are warranted together with inherit redundancies in sensors to verify precision and accuracy.

Visual Analysis

The rule is: keep analysis simple, intuitive, and visual.  For a dashboard to generate meaning, the larger the number of people that can share in it’s understanding, the greater the value. Creating dashboards and visualizations are an integral aspect of delivering understanding and insight.

For instance, to better address energy efficiency improvements, it helps to have a comprehensive perspective of energy consumption.  By installing CTs across major circuits in a power panel, a more granular understanding of where and when energy is consumed can be established.  This energy perspective provides deep insight as to what activities are responsible for energy usage.  By applying the lessons learned from the analytics, energy usage patterns can be correlated to building activities, process flow, operations, and human resource interaction.

Figure 4 IoT Sensor Analytics

Figure 4 IoT Sensor Analytics

Source:  Analytics 2 Insight, LLC

Visual analysis can be used to identify patterns, detect anomalies, and generate insight.  Value is derived from insight gained through a deeper knowledge and understanding of process flow, operations and conditions, from which strategies and corrective actions can be formulated.  Often it takes many inputs to provide a thorough understanding.  It is through the Mosaic Effect, where a picture can only be revealed through the confluence of multiple inputs, that sensor data creates value and insight.  Visual analysis creates value by revealing where efficiencies, only derived through analytical insight, can be gained and costs can be lowered.

The IoT and analytics will profoundly change our lives.  Intelligent devices and sensors enable the IoT to monitor, measure, manage and protect people, property, equipment and assets.  Data reporting and analytics provide substantial value in reducing costs and protecting assets.  Analytics also plays a role in deriving insight to optimize processes and objectives.


Authored by:
Michael S. Davies, CFA

Mike is the Chief Strategy Office for Arkados, an Integrated Software and Technology Solutions Provider for Internet of Things (IoT) Applications and Analytics for intelligent devices. Mike is developing strategies and analytics to create value for energy and resource conservation through context aware connected devices, data analysis, visual analytics and renewable energy solutions.

Mike began his career selling software into vertical markets leading to marketing strategy consulting engagements. After business school he advanced to Wall Street as a financial analyst providing research and analysis on technology including semiconductors, data and wireless communication companies. Mr. Davies has provided business strategy and technology consulting to Apple, IBM, the Port Authority of NY & NJ, and NJ Dept. of Transportation. Analytics is a common thread throughout his career and his book Analytics 2 Insight on Amazon is featured in Data Science Association bookstore. Mike has been featured on CNBC, CNN, and Bloomberg with citations in the Wall Street Journal and Investor’s Business Daily. Mike graduated from Columbia University with BA in Economics and a MBA from University of California, Los Angeles.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s