Symphony AzimaAI - Best IoT Analytics Solution
Company: Symphony AzimaAI, Woburn MA
Company Description: For 50+ years, SAAI have been innovators of industrial insight - from machine component health to plant performance optimization. This has been enabled by a talented and rapidly growing team of deep domain experts in process industries, discrete manufacturing, IIoT, and artificial intelligence. Our solutions span Predictive Maintenance and Process Health & Optimization.
Nomination Category: New Product & Service Categories - Business Technology
Nomination Sub Category: IoT Analytics Solution
Nomination Title: Symphony AzimaAI's APM 360™
New Product - Released, March 10, 2020
APM 360™ - A solution covering overall machine health and performance, real-time prescriptive monitoring, and predictive maintenance insights for critical plant assets like turbines, compressors, motors, pumps, gearboxes to furnaces and columns. It enables the building of a “digital twin”, leveraging IIOT, AI & FMEA that provide anomaly detection with automated cause analysis and advisories to account for complex, dynamic machinery patterns to ensure critical machinery operates at peak.
Differentiators for legacy systems:
From process & low frequency vibration data to process & high frequency vibration data - Integration with 10khz+ vibration data produces a more accurate AI model for prediction.
From high false positives rates to low false positive rates with AI & FMEA - Pattern recognition routines are known to have high rate of failures. APM 360’s AI based anomaly engine looks for anomalies. Then the Failure Model Effects & Analysis (FMEA) engine and SIAI’s propriety scoring know-how acts as a double filter to ensure no spurious anomalies.
From diagnostic flags to automated apparent cause analysis - Macro-level diagnostics flags to detailed micro-level diagnostics assessment that provides more actionable intelligence. APM 360 apparent cause engine goes down to the root of the problem, it will pinpoint to the specific valve that is constricting flow and therefore leading to cooling system issues.
From fault detection with pattern recognition to supervised and unsupervised AI - Systems look for known fault signatures and runs analytics to do pattern matching. In case of no match, these systems do not raise defects. This does not mean that a defect is not there. APM 360 AI models looks for deviations from normal operating behavior and uses both supervised and unsupervised learning models to detect and diagnose “unknown unknowns”.
Key features include:
Pre-built asset templates
Out-of-the-box KPIs and analytics
Real-time monitoring
Domain-informed AI
Anomaly detection and prediction
Apparent cause detection
Mitigative/diagnostic action recommendation
Automatic AI model re-training
Blend of process & high-frequency vibration datasets
Scenario (what-if) analysis
Connectors to industrial IT/OT systems
Sensor health alerting
Cloud based continuous integration, continuous deployment
Introduced earlier in 2020, APM 360 has already won three awards honors – with its unique approaches compared to legacy systems previously described.
Plant Engineering Magazine-Finalist in the Asset Management category –April 2021 -winners announced. https://www.plantengineering.com/articles/2020-product-of-the-year-finalists-2/
IOT Global Awards-Finalist in the Industry & Construction categoryThursday, 19th November – Winners announced. https://iotglobalawards.com/shortlist/
Oil & Gas Engineering Magazine-Gold in the Data & Analytics category –to be announced December 7th https://www.oilandgaseng.com/articles/vote-for-the-best-products-of-2020/?oly_enc_id=7354I2969623J9L
Sample customer use cases include:
Large Integrally geared Air Compressors- For a petrochemicals ammonia plant. A digital twin model based on unsupervised deep learning models was deployed that combined high-frequency vibrations with process data to detect incipient anomalies. System level approach identified leading pre-cursor process factors that could lead to mechanical failure issues. The APM identified intercooler performance as a direct cause behind high vibration surges in the compressor; Savings $1MM from production avoidance
High-pressure Turboexpander– For a large polymer plastics manufacturer. A digital twin model based on semi-supervised machine learning was deployed that combined process and vibration data to detect incipient anomalies. A system-level approach identified leading pre-cursor operational factors leading to anomalous vibrations compounded by dynamic wheel imbalance that led to sudden tip failures. Savings $1MM+ from unplanned wheel failure replacement maintenance.
LNG Sour Water Stripper– For a large natural gas supplier. A digital twin model based on unsupervised machine learning was deployed that used process data to detect incipient anomalies. A system-level approach identified leading pre-cursor process factors that could lead to fouling. APM360 identified inefficient feed of solvents as the primary cause behind anomalous thermal profiles leading to fouling in the stripper.Improve sulphur removal by 1-3%.
LNG Off-gas Compressor– Early detection & identification of incipient faults for a 2-stage motor-driven boil off-gas reciprocating compressor to prevent their escalation to an unplanned downtime situation. Major faults predicted with valves, pistons, bearings & lube system 3-4 weeks in advance; Savings $1-2MM in outage avoidance and repair costs.
APM 360 user interface is through the Predictive Portal. A 5 minute video will provide insight into why APM 360 is award winning and experiencing rapid adoption.
More information on APM 360 can be found at https://symphonyazimaai.com/solution/asset-performance-management/