What happened
Dearborn Industrial Generation (DIG) runs a power and steam plant in Dearborn, Michigan. It asked EPA to approve CMC Solutions' SmartCEM-75 predictive emission monitoring system (PEMS) as the NOx monitor on its units. EPA describes the PEMS on boiler Unit BL1100 as neural-network software that turns boiler sensor readings into NOx values. The PEMS ran beside a certified analyzer for 720 operating hours in July and August 2004, including 98 hours of startup and shutdown. EPA found that its output passed all three statistical tests and approved it on 6 September 2006. EPA had given final approval to the PEMS on turbine Unit GTP1 on 5 April 2006. There a model trained on 40 hours of data was tested on the remaining 762 hours and passed EPA's own calculations. EPA estimated that monitoring GTP1 this way, with the extra audits it required, would cost less than half as much a year as an analyzer. The cost figures come from EPA's cost model. The letters report no change in emissions.
Sources
-
Regulatory document, by authors with no tie to the system.
U.S. Environmental Protection Agency, Clean Air Markets Division. Approval of the Predictive Emission Monitoring System Installed on Unit BL1100 at Dearborn Industrial Generation (Facility ID (ORISPL) 55088). Letter to Dearborn Industrial Generation, 6 September 2006.
The PEMS NOx lb/mmBtu output passed each of the three statistical tests for all unit operations.
-
Regulatory document, by authors with no tie to the system.
U.S. Environmental Protection Agency, Clean Air Markets Division. Final Approval of the Predictive Emission Monitoring System (PEMS) Installed on Unit GTP1 at Dearborn Industrial Generation (Facility ID (ORISPL) 55088). Letter with attachments, 5 April 2006 (scanned copy).
the estimated annual cost of operating and maintaining a PEMS at Unit GTP1 using trained in-house staff and a portable analyzer would be less than half the cost associated with CEMS
Details
AI system
- Workflow
-
DIG contracted CMC Solutions to collect the data for the 720-hour demonstration. The certified analyzer on the unit gave hourly reference data while the PEMS was trained and tested. The PEMS read boiler inputs such as fuel flows, air flow, temperatures and oxygen levels.
Money
- Kind of value
-
Estimated yearly saving of the PEMS on Unit GTP1 over an analyzer even if outside contractors run the extra audits
- Amount
-
Estimate: $12,950
- Recipients and payers
-
Dearborn Industrial Generation
Follow-up
- Conditions for replication
-
Each input must stay within the range of the training data, or DIG must report the unit's maximum potential NOx rate. The PEMS must keep passing daily checks, monthly audits in the ozone season and periodic test audits.
Entry details
- Sector
-
Environment and climate.
- Purpose
-
CMC Solutions built SmartCEM-75 to predict a unit's NOx emissions from its sensor readings so that the prediction can serve as the unit's emissions monitor.
Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, AI contribution, beneficiaries, reach, distribution and access, duration, every measurement question, currency and prices, period and unit, realization, gross or net, included costs, comparison, method, return and payback, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). SmartCEM-75 stands in for NOx analyzers at a Dearborn plant (AIBD-0843). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0843.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)