Knowing where to start with using AI to solve real value-adding business problems can be challenging. The Ekkono Software Development Kit is a generalized edge machine learning tool, used for making connected things smart. Enjoy a selection of smart use cases that our SDK enables.
We like to illustrate product smartness using a cognitive staircase, where the steps represent the complexity level of each use case. As every step benefits from the one below, this highlights the importance of getting the basics right when enhancing your product portfolio with machine learning capabilities.
Virtual Sensor [Current State]
Physical sensors are expensive, prone to fail or to drift. With Ekkono’s SDK, you can create a virtual sensor based on related physical sensors. The virtual sensor can then replace a physical sensor, provide support on sensor failure or act as a reference when monitoring drift and anomalies.
Health Monitoring [Current State]
Ekkono’s SDK can analyze the data stream from your device under normal conditions and determine how the device behaves when it is healthy. Given this knowledge, we detect even the smallest deviations as an indication that something has changed. Early warnings of decreasing system health will give you an opportunity to act in advance.
Predictive Maintenance [Predictive]
Moving from preventive to predictive maintenance opens up the possibility to base maintenance operations on a future state of the machinery, instead of the current state. Knowing the future state, maintenance operations can be planned in an even more precise way, increasing asset lifespan and reducing environmental impact.
Decision Support [Prescriptive]
By letting a predictive model represent the behavior of a system and using that model to simulate different operational settings, Ekkono provide a tool for an operator to try out complex scenarios before making selections. When knowing the impact of different settings the operator can select the ones resulting in preferred performance.
Create a smart self-tuning system by using Ekkono’s machine learning models that continuously predict target values based on operational parameters and ambient factors. Use the predictions to optimize the machinery settings and let the system constantly improve the performance to prolong machinery life and lower the energy consumption.
Predictive Control [Cognitive]
When working with systems having a built-in inertia, for example heating and cooling systems, there is a need for step by step optimization. This can be performed by model predictive control. With Ekkono SDK’s predictive models, each part of a system can adapt to its environment (for example rooms in a building) and thereby increase overall performance such as equipment utilization and energy efficiency.