How Can Manufacturers Use Advanced Analytics Applications to Improve Efficiency, Reduce Waste, and Minimize Environmental Impact?

Written by Nick Gigliotti on July 06, 2020

Many manufacturers today list sustainability as one of their main corporate objectives. These companies understand that “going green” is more than just a trendy catchphrase—they realize that optimizing their processes to reduce energy consumption and waste gives them a significant competitive advantage.

However, identifying opportunities for continuous improvement in this area is difficult, as the engineers on the ground often do not have the advanced analytics applications required to investigate their process in a rapid, in-depth way. It is extremely difficult for plant engineers to eliminate waste and improve efficiency when they spend most of their analytics time sifting through time series data in spreadsheets.

With Seeq’s quick connectivity to time series data stores and easy-to-use advanced analytics applications, engineers can make their processes more sustainable using time series data that was previously too difficult to deal with using spreadsheets.

Seeq’s Workbench Analysis and Organizer Topic applications empower engineers to rapidly build energy models and identify wasteful processes, along with dashboards to monitor operating conditions across hundreds of assets. And, as some Seeq employees have found, these applications can also be used to decrease their carbon footprint at home!

Rapid Development of Energy Models Identifies Inefficient Equipment

In a process plant, there are hundreds of items of equipment that contribute to the total energy use. Sensor data from this equipment—such as flow, level, pressure, temperature, and other parameters—is stored in time series data historians.

Subject matter experts (SMEs), typically process engineers, can use this data to build predictive models of total energy consumption using each individual item’s sensor data as inputs. With this method, they can determine if each piece of equipment is operating efficiently based on the coefficients in the model.

For example, if a steam block valve is designed to provide 1000 lb/hr of steam when open, but it has a coefficient of 500 lb/hr in the model, one can assume it’s leaking when 100% closed. A properly operating valve would have a coefficient of close to 1000 lb/hr, which would indicate that when the steam valve’s discrete signal equals one, the total facility steam goes up by 1000, and when the discrete signal equals 0, no steam is passing by according to design. In addition, the y-intercept of the energy model is valuable because it shows what the energy consumption of the facility is when all the other pieces of equipment are not operating (Figure 1).

Model statistics for a basic linear regression energy model.

Figure 1: Model statistics for a basic linear regression energy model.

Using traditional analytics tools like spreadsheets, energy models are difficult to build and maintain, and they quickly become outdated and obsolete. Companies like Allnex, however, use Seeq Workbench’s Prediction Tool to construct energy models that they use to decrease their total energy consumption with minimal capital expense. They also use the tool to monitor parameter drift so issues are discovered early.

Seeq’ Prediction Tool (Figure 2) can be used to create linear and non-linear multivariate regression models and generate a signal (in this case, Total Energy Use) based on other correlated signals (Steam Valve ON/OFF, Ambient Temperature, etc.). A Prediction Statistics Panel, which is an output generated by Seeq for every Prediction Tool model, gives engineers the information they need to study their process. Compared to the hundreds of hours required to build energy models with spreadsheets, the engineers at Allnex have used the Prediction Tool to build energy models in significantly less time.

User interface for Prediction Tool in Seeq Workbench.

Figure 2: User interface for Prediction Tool in Seeq Workbench.

Minimizing Environmental Impact Across Assets at Scale

While decreasing overall energy consumption is a critical component of industrial sustainability, minimizing a facility’s impact on the surrounding environment is crucial as well. Companies are often aware of operating conditions that affect their neighboring environmental ecosystems, but often find it difficult to scale and monitor these conditions across a fleet of assets.

Wind turbine sites at Duke Energy Renewables needed monitoring of operational conditions that could put Indiana bats, an endangered species, at risk. SMEs wanted to identify these conditions across all turbines at that wind farm so that if those conditions were met, they could implement Bat Curtailment measures to minimize harmful impact to the bats. With spreadsheets and traditional analytical methods, calculating the conditions and monitoring many turbines spread across a large site was extremely time-consuming and difficult.

Duke leveraged the asset hierarchy and analytics tools in Seeq to identify Bat Curtailment periods based on certain dates, times and weather conditions such as wind speed and ambient air temperature, and to monitor these conditions across all turbines using the Treemap view in Seeq Workbench Analysis. Treemap visualization shows each turbine as a rectangle, with the color of each indicating that some condition is present in that time range (Figure 3).

Treemap Visualization in Seeq Workbench

Figure 3: Treemap Visualization in Seeq Workbench. Assets with higher priority conditions are in red, medium priority conditions in yellow, and low priority conditions in green.

Duke, for example, created a yellow “Warning” condition if a turbine was approaching a state where Bat Curtailment was needed. They also created a red condition to show when full Bat Curtailment measures were needed.

While large manufacturing organizations have found Seeq useful for sustainability improvements, some or our employees have used Seeq’s easy connectivity to data sources to limit their personal carbon footprint as well.

Jenny Reinman, a Seeq account manager, pulls wattage data from her residence’s main power feed into an InfluxDB database, which is then connected to a Seeq server running in her home. She uses this data to quantify energy use while performing various household tasks, and also to determine if her greenhouse heater gets stuck in the on mode.

Marius Oancea, a Seeq principal software engineer, also collects electricity data from his main power feed, and uses tools like Profile Search to find which of his household appliances are operating and when.

The simple connectivity of Seeq to multiple data sources allows thus allows individuals as well as manufacturers plant to reduce their carbon footprint.

Seeq enables manufacturers organizations to drive sustainability improvements with advanced analytics applications, monitoring visualization for assets at scale, and simple connectivity to many different types of data sources. SMEs armed with Seeq can rapidly build analyses to investigate, monitor, and improve plant processes.