Title:  NALCO -- AVEVA System Platform integrated with AI/ML algorithm Enabling Specific fuel consumption, Failure prediction, automated RCA and enhance the combustion process.

The digital transformation of coal-fired boilers, with an emphasis on return on investment, role of powerful data infrastructure which can support key pillars of revenue optimization. This transformation focuses on enhancing the combustion process by reducing specific coal consumption, early fault detection, and providing automated root cause analysis for boiler failures. The AVEVA System Platform, formerly known as Wonderware, brings together utilities, industry experts, researchers, and technology providers to explore the latest advancements, case studies, and strategies. These efforts aim to leverage optimization techniques to enhance boiler efficiency, reduce boiler losses and failures, and improve overall boiler health score. Artificial intelligence and machine learning [AI/ML] are utilized to automate coal-fired boiler operations, optimizing boiler efficiency, minimizing boiler losses, reducing specific coal consumption, predicting failures early, and providing automated root cause analysis. This presentation will delve into the integration of AI/ML with the AVEVA System Platform, highlighting the impact on fuel savings, fault prediction, and the enhancement of the combustion process.

Improvement in combustion control to achieve reduction in specific consumption of coal using AI/ML algorithm. 

To achieve a reduction in specific coal consumption through improved combustion control and efficiency optimization, we have embarked on a detailed project focused on optimizing boiler efficiency and reducing boiler loss to improve boiler efficiency. Our approach leverages historical data analytics and AI/ML modelling to provide actionable insights and recommendations.

Detection/display of parameters deviating from norm to reduce unplanned outages. 

This involves continuous monitoring of critical operational metrics as shown in below dashboard, such as Air flow, temperature, pressure, and emissions. By Implementing advanced analytics and machine learning algorithms, anomalies can be identified in real-time such as showing prediction and deviation in real time with triggering alerts for immediate corrective actions. This proactive approach enables early fault detection, preventing minor issues from escalating into major failures.

Early failure detection and provide automated RCA [Root Cause Analysis] about impending failures. 

The automated Root Cause Analysis [RCA] Report for coal-fired boilers provides detailed insights into impending failures by identifying key operational anomalies. The RCA report, as displayed, lists various failure modes such as incomplete combustion, low boiler efficiency, Flame failure, low fuel flow, Drum level very low, low air flow, Master furnace trip along with their specific causes, including issues like increased fuel flow and unchanged secondary air flow respectively cause for incomplete combustion. By pinpointing these deviations, the system facilitates early failure detection and corrective actions. The AVEVA System Platform, integrated with AI/ML algorithms, continuously monitors boiler parameters, detects anomalies, and generates comprehensive RCA reports, thereby enhancing reliability, reducing unplanned outages, and optimizing overall boiler performance.

Identification of deviation from normal best operation and provide guidance/recommendation messages. 

Real-time monitoring of critical parameters for boilers and auxiliaries enables swift identification of deviations from optimal limits. Upon detecting such deviations, the system promptly notifies the client via notifications. These notifications include detailed information about the cause of the deviation and suggested corrective actions. This integrated approach ensures proactive management of operational issues, minimizing disruptions and maximizing efficiency.

Goals and Challenges:

Goals:

·        To optimize combustion process

·        To reduce fuel consumption

·        To reduce CO, SOx, NOx emission

·        To reduce unplanned breakdown

·        To provide automated root cause analysis of impending failures

·        To monitor heat balance in superheaters, Economiser, and drum.

·        To reduce troubleshooting time, provide guidance messages to the operator.

Challenges:

·        Availability of data across multiple platforms

·        Challenges in correlating disparate data sources

·        Difficulties in developing accurate models for boiler operations

·        Identification of key factors contributing to process deviations

·        Providing actionable corrective measures to operators

·        Compiling and analysing historical failure data

·        High coal consumption relative to production output

·        Occurrence of unplanned boiler shutdowns

·        Issues with improper combustion processes

·        Predominance of reactive maintenance strategies

·        Increased plant downtime due to prolonged fault resolution times

Results and benefits

·        Reduced unplanned shutdowns

·        Reduced boiler heat losses

·        Improve boiler efficiency

·        Reduced fuel consumption 

·        Improved combustion process

Author Profiles:

Santosh Kumar: 

Santosh is an experienced professional in industrial automation with over 17 years of industry experience. His expertise lies in dealing with automation systems for boilers, turbines, and their auxiliaries, as well as field instrumentation. Santosh has successfully upgraded and modified outdated systems and actively incorporates cutting-edge technologies, including Industry 4.0 solutions. His proficiency extends to industrial automation systems [DCS/PLC], Python, IoT, and field instrumentation.

Pankaj Chavan:  

Pankaj is an experienced professional in industrial automation with over 17 years of industry experience. His expertise lies in dealing with automation systems for boilers, turbines, and their auxiliaries, as well as field instrumentation. Santosh has successfully upgraded and modified outdated systems and actively incorporates cutting-edge technologies, including Industry 4.0 solutions. His proficiency extends to industrial automation systems [DCS/PLC], Python, IoT, and field instrumentation.

Jevin Makadia:  

Jevin Makadia serves as a Manager of Accounts and Projects at a Business Analytics Consulting Firm specializing in leveraging the Industrial Internet of Things [IIoT] to generate actionable insights. In this role, Jevin leads business opportunities and collaborates closely with clients to understand their challenges, propose tailored solutions, and ensure successful project execution.

His expertise spans various industries, including power generation, power transmission, oil and gas, food and beverages, fast-moving consumer goods [FMCG], and chemicals. An Electrical Engineer by training, Jevin is adept in the digitization of industries and the implementation of Industry 4.0 initiatives. His proficiency encompasses a range of Operational Technology [OT] and Information Technology [IT] solutions, including Industrial IoT, artificial intelligence, data visualization, and AVEVA system platforms.