Data acquisition explained - the ANNEA approach enabling predictive maintenance 4.0
We are living in an era defined by data. The importance of data is increasing day by day and covers an essential part of the core of any business. At ANNEA, we work with a huge amount of data every day, creating AI models combined with physical and normal behaviour modelling principles. However, prior to building and training the models, there is a crucial step: data acquisition. Depending on your IT architecture and data policy, data gathering might become a real challenge. Through this reading, you will get an understanding of the whole process of data acquisition, pre-processing and the perspective of ANNEA. First thing is first! Let’s start with the basics.
What is data acquisition?
Data acquisition stands for the process of transmitting a real-world phenomenon into a digital numeric value. The phenomenon in the physical domain, such as warmth and cold (which are observable); while the digital domain as computers that stores values as digital numeric values (temperature and vibration). In other words, in a simple way, it is a translation process.
What is data acquisition system?
The importance of data acquisition systems (DAQ) have been increased significantly in the last decade due to the rise of the importance of data and as these systems provide accuracy. Data acquisition systems are defined as any instrument or computer that acquires data from sensors through amplifiers, multiplexers, and any necessary analogue to digital converters or the internet. The system itself transmits data to a central location for processing.
How to acquire data?
Sensors
Sensor Calibration
SCADA System
Supervisory control and data acquisition (SCADA) systems are generally used in industrial plants. SCADA is an advanced computer system collecting and analysing real-time data. Implementing SCADA to your plants will provide you the following benefits:
• Control of industrial processes locally or remotely • Monitoring, gathering and processing real-time data • Direct interaction with devices, such as sensors • Human machine interface (HMI) software • Recording of the events
Meteorological Data
Meteorological data stands for the meteorological phenomena caused by the atmosphere, such as wind, temperature, air density, etc. that may have an impact on the condition of industrial plants. We basically analyse the real-time weather forecast in a more detailed way to make your assets more efficient.
Condition Monitoring System (CMS)
As the name suggests, a condition monitoring system gathers data from implemented sensors in order to monitor the condition of the asset(s). Upon collecting the data, predictive maintenance can be conducted. Condition monitoring helps to decrease OPEX and the total labour cost, to increase the productivity and to avoid unplanned maintenance. Hence, in the end, your plant(s) will be more productive.
Gateway
A gateway is a software or hardware solution giving a single central point of access to connect to the target device(s) upon connecting to multiple data sources.
Focusing on the direction of data flow, there are two types of gateways:
• Unidirectional Gateways: Allowing data to flow in only one direction. They can be used as archiving tools.
• Bidirectional Gateways: Allowing data to flow in both directions. They can be used as synchronization tools.
Focusing on their functionalities, there is a high variety of different gateways. We highlighted the types of gateways we use at ANNEA:
Network Gateway: The most common type of gateway. Providing interface between two dissimilar networks operating with different protocols. Whenever a gateway is mentioned without its type, it generally means a network gateway.
Cloud Storage Gateway: It is a network node or server translating storage requests with different cloud storage service APIs. It facilitates integration of private cloud storage into applications without the necessity of application transfer into any public cloud, thus, simplifying data communication.
IoT Gateway: IoT gateway assimilates sensor data from IoT devices in the field and translates between sensor protocols before sending it to the cloud network. They provide connection between IoT devices, cloud network and user applications.
Is the collected data trustworthy?
How to process (convert and transmit) the acquired data?
As we mentioned at the beginning, data acquisition stands for the process of transmitting a real-world phenomenon into a digital numeric value. This process is done using a variety of sensors and signal conditioners. The outputs are sampled by analogue-to-digital converters (ADC) and then written in a time-based stream to a digital memory media. These systems are known as the measurement systems.
Transmission and data conversion are the key steps needed to be focused. Data is encoded in different ways in a digital domain. Data conversion is the conversion process of digital domain data from one format to another. While data transmission reflects the transfer of data through a point-to-point or point-to-multipoint to the target. These specific processes, in a simple way, can be considered as translation and communication respectively.
ANNEA enables you to process your data with our end-to-end solutions. Your measurements are transmitted through gateway(s) to your target storage. As a part of our unique solutions for our customers, we process your data with the ANNEA digital twin technology, offering more precise results for each of your assets. We set-up digital twins (virtual replicas) of your assets which allow us to predict the future health condition of their components individually. Moreover, your data are always transmitted with precision, full security and privacy on the ANNEA Cloud.
What is data storage?
Data storage is the process of collecting data in a storage medium. Today, digital data storage systems have become more essential for businesses, as they are faster and more secure then analogue options. Data storage in a digital, machine-readable medium is often called digital data. Digitalisation is the major factor driving the trends of change in these applications.
Businesses, in general, prefer to store their data on software-defined storage (SDS), cloud storage, network-attached storage (NAS), object storage, file storage. The way the data is stored is subject to what kind of data businesses want to acquire and how to store the data. The solution we offer with the ANNEA Cloud is a secured and private transmission of your data to our servers.
What is data analysis?
Usually, businesses start to analyse the data they acquired upon storing. Data analysis stands for inspecting, transforming and modelling the acquired data with the goal of discovering useful information, making conclusions.
When analysing the acquired data, there are crucial steps usually remain in shadows such as data pre-processing. It sometimes covers up to 80% of the whole data analysis, even in AI applications. Data scientists work on this crucial step, in order to bring the data into useful formats.
During data pre-processing, we conduct primary data evaluation reflecting the original analysis. This also includes data cleaning, which is the process of detecting and correcting corrupt or inaccurate data and then modifying or deleting “dirty” data in order to make the data more accurate.
The approach of ANNEA to the data acquisiton process
For us at ANNEA, data is the core of our daily life. We acquire, process, analyse and handle the whole data life cycle with professionalism. As a result, we provide the next generation automated predictive maintenance platform, enabling machines to analyse and communicate their own health status.
Nowadays, most of the assets do have sensors. What makes the difference is how the data is evaluated. With ANNEA’s autonomously working IoT-platform, our customers are taking one step ahead on the state-of-the-art data acquisition and evaluation process.
We extract, transform, load (ETL) data from multiple sources to set up a database. You can control, monitor and analyse real-time data through the state-of-the-art SCADA system whether at your asset(s) or in online platforms.
We combine different data sources, including SCADA, CMS, and Meteorological data in order to create a benefit for our customers and to predict the future failure events (in which component and when and what the root-cause will be) in their assets. Especially for machines that operate outdoors, meteorological data plays a key role in the whole process. Therefore, we analyse the meteorological data combining all the possibilities and come up with the best possible solutions to maximise the efficiency of your plant(s).
We do not see data as just a value bringing efficiency to your business. Data is also a private property. It is an important part of the ID of your company. Data security is a top priority for ANNEA. We always respect the regulations, such as GDPR, whether if we process your data or protect your data. Also, we treat your data with the maximum level of discreteness. You can choose on how to store your data in order to ensure their privacy.
What is next?
Products
ANNEA Failure Prediction
ANNEA Underperformance Detection
ANNEA Energy Production Forecasting
ANNEA (Drone) Image Recognition
ANNEA Maintenance Planner
ANNEA Monitoring (upcoming)
© 2022 annea.ai GmbH, All Rights Reserved.
English 
Deutsch