Power utility monitoring and maintenance with AWS IoT and SAP integration

The rise in electricity consumption, particularly due to electric vehicles (EVs), presents several challenges for power utilities, such as significant costs and long planning timelines of infrastructure upgrades, grid stability issues caused by fluctuations in demand due to EV charging, and the intermittent nature of renewables.

In the first post (Improved utility asset management and maintenance using AWS IoT and generative AI technologies), we discusses how AWS IoT (Internet of Things) services and generative AI technologies can enhance utility asset management and maintenance. It highlights the benefits of integrating these technologies, such as improved data analysis, predictive maintenance, and efficient resource management. The post emphasizes real-world applications, showcasing how utilities can use artificial intelligence (AI) for better decision-making and operational efficiency. The combination of AWS IoT and generative AI aims to streamline processes, reduce costs, and enhance service reliability in the utility sector.

SAP (Systems, Applications, and products in Data Processing) is an enterprise resource planning (ERP) software suite. SAP plays a crucial role in streaming and enhancing power utility operations, such as supply chain management and financial management. SAP provides tools for asset management and maintenance, and helps with scheduling preventive maintenance, managing repair activities, tracking asset performance, etc. When certain asset issues such as transformer abnormalities are detected, it is important to notify SAP for proper maintenance, repair, or even replacement on time and automatically.

In this post, we detail the solution architecture regarding how to integrate AWS IoT solutions with SAP solutions for predictive maintenance.

SAP Plant Maintenance

SAP Plant Maintenance (PM) comprises a set of data and processes to maintain the high availability of technical systems. This helps users move away from outdated interval or breakdown-based maintenance cycles to a planned or preventative maintenance cycle.

The following diagram illustrates the business processes that effectively plan, inspect, record and act on information about various equipment’s devices and assets within an organization in a timely manner. More detailed information about each of the technical objects can be found in the SAP documentation.

Figure 1: SAP Plant Maintenance modules

Example Business Process

The following diagram depicts a business process flow of maintenance notification.

A lot of equipment readings are recorded in plant maintenance and user service processes. In case of a potential disruptive event in an assembly or sub-component, a service notification can be triggered.

Figure 2: Maintenance notification process flow

Some users would also use SAP Enterprise Asset Management.

SAP Enterprise Asset Management (EAM) helps organizations plan, optimize, execute, and track the necessary activities, priorities, skills, materials, tools, and information associated with an asset. Failure to manage and maintain enterprise assets can lead to unplanned downtime, suboptimal asset performance, and supply shortages. Users also rely on EAM systems to demonstrate compliance with regulatory bodies to preclude liability if a failure occurs. To learn more about the SAP EAM solution, refer to SAP Enterprise Asset Management.

Figure 3: SAP Enterprise Asset Management

IoT based predictive maintenance SAP integration

The implementation of IoT event processing requires a robust framework that can handle real-time data streams. The following architecture presents the event-to-business actions framework, demonstrating how IoT events are captured, processed, and converted into meaningful business actions.

Figure 4: AWS IOT Integration with SAP Plant Maintenance

In part 1, we covered data sensing and collection, transmission, data processing, storage, analysis, AI/machine learning (ML), and data presentation. In the following steps, we use one example of a transformer overload event to depict the information flow across systems for predictive maintenance SAP integration. In a regular residential subdivision where one transformer serves multiple users, more and more residences purchased and installed Electric Vehicle Supply Equipment (EVSE) chargers at home, while the service transformer was not upgraded. In the evening when residences came back and started to charge their EV cars, the transformer overload would happen.

IoT based heat sensors measure the transformer hotpot temperature and the data is sent to the AWS cloud as described in the first post.

(1) Transformer overloaded event is detected, the event is sent to Amazon Simple Notification Service (Amazon SNS) (highlighted in red circle in preceding diagram), and it triggers an AWS Lambda function

(2) Lambda is a serverless function that orchestrates the process of detecting whether a stream contains any alerts related to failures or warnings. Then, the inference result is passed to the SAP Advanced Event Mesh.

(3) AWS Secrets Manager is used to store credentials. These are used by the Lambda function to provide a payload to SAP Advanced Event Mesh.

(4) The Event-to-Business-Action framework (extension app) processor module’s endpoint is subscribed to SAP Advanced Event Mesh, and thus receives this event.

(5) The Event-to-Business-Action framework processor module uses the Decisions capability of SAP Build Process Automation to derive a business action. In this scenario, the Plant Maintenance Notification creation in SAP S/4HANA system is based on certain characteristics of the incoming event.

(6) The Event-to-Business-Action framework preprocessing module takes the entire message from Amazon SNS and sends it to Amazon Bedrock through the SAP Generative AI hub. The resulting summary report is passed on as a long text to the Maintenance notification.

(7), (8), (9), (10), and (11) The Event-to-Business-Action framework processor module triggers the defined action in the SAP S/4HANA system by using the SAP Destination Service and SAP Private Link Service. In this scenario, transformer overload is detected, actions are taken regarding whether to continue monitoring the situation, or to send a crew to repair/upgrade the transformer if it is in stock. If a new transformer is needed, then a purchase order could be issued. The overall process is event-driven and automated.

The integration architecture uses the SAP Events-to-Business Actions architecture, which is an event-driven framework on SAP Business Technology Platform (SAP BTP). It responds to and integrates with events generated from industrial production processes in plants, warehouses, and logistics into enterprise business systems. It triggers associated business processes to enhance enterprise operations and enable rapid decision-making. A sample integration flow with code samples is available in the SAP Discovery mission.

SAP integration benefits

Integrating SAP with IoT based predictive maintenance brings the following benefits:

• Event-driven integration architecture with SAP Advanced Event Mesh as a central hub.
• Seamless integration to SAP EAM and SAP S/4 HANA depending on the user’s landscape
• Network security-focused design with SAP Private Link specifically for RISE with SAP users between SAP BTP on AWS (any Region) and SAP S/4HANA on AWS.

Conclusion

The growing demand for electricity and the increasing complexity of the power grid present significant challenges for utility. AWS IoT services and generative AI help power utility predictive maintenance by enabling real-time monitoring of equipment and identifying potential problems before they occur. However, when critical events happen (such as defectiveness), you must take early actions to repair or replace defective devices as soon as possible to reduce the utility asset down time and improve user satisfaction.

The design and description outline in this post demonstrates how utility can use and integrate the SAP Events-to-Business Actions architecture to quickly take actions and automate the overall process from service notification and work order creation, maintenance order planning, and even logistic and supply chain management.

In future posts, we will provide the sample code to help users with a quickstart.

Many thanks to Praveen Kumar Padegal, PVN Pavan Kumar, Sivakumar N and Swati Maste for their collaboration on this blog.