ASC 4421MF
Uniaxial, capacitive
Measurement range: ±2 to ±200 g
Noise density: 10 to 680 µg/√Hz
Frequency range (±5 %): DC to 2900 Hz
From sensor reading to feature vector: Machine Learning
The smart sensor processes measurement data by extracting features from the measurements and relaying them as feature vectors. This is possible because the smart sensor combines computational power with algorithms and is in bidirectional communication with the other components of the system. This allows adaptation of the analysis algorithms for extracting features during operation.
The smart sensor systems ASC AiSys® have computational power that can be used freely to adapt the algorithms for extracting features to individual requirements!
Artificial intelligence helps to understand complex contexts and to describe them in mathematical equations. AI makes it possible to analyze large data volumes and, for example, to automatically create rules for optimal control processes.
Smart sensors have the capability to analyze feature vectors and to make decisions independently. This results in decentralized systems in which single sensor nodes operate independently of each other, yet are linked together to optimize processes.
The smart sensor systems ASC AiSys® are optimally adapted to the requirements of artificial intelligence. They combine intelligent analysis with communication!
A digital twin is the virtual model of a real physical object. It connects the real and virtual world. Digital twins are brought alive by real sensor data, which exactly determine working conditions or machine positions, for example. Interconnection of the two worlds and analysis of the data collected by the sensors enables early detection and elimination of problems in product development and utilization. This makes it possible to prevent machine and system failures and to develop new products without expensive physical prototypes.
Smart sensor systems ASC AiSys® are converting the acquired raw data to information. Those information are accessible by Digital Twins though flexible interfaces. Due to the integration of functionalities and features regarding data analyses directly in the smart sensor systems ASC AiSys® the overall system is significantly simplified.
During creation of a prototype it is possible to implement and test different design alternatives. This significantly decreases the development cycle and ensures fulfillment of all requirements.
The digital twin helps to boost efficiency and improve quality in production.
Predictive maintenance increases availability
Above all, the networking of sensors provides substantial added value within the framework of digitization. Networking facilitates integration of sensors in additional systems, as well as the exchange of information between sensors. The communication can use a wireless or cable connection. Mesh networks achieve long ranges and high redundancy.
The smart sensor systems ASC AiSys® speak many languages – such as EtherCAT, CAN, Bluetooth or LTE!
The basic concept of predictive maintenance is quite simple: Early detection of critical changes can prevent dangerous situations before damage occurs. This requires knowledge of the relevant parameters, as well as continuous monitoring and comparison with target values. ASC sensors provide the high-precision and reliable measurements needed to achieve this.
The better the data quality, the more reliable the prediction!
Machine learning and artificial intelligence reveal the features that are necessary for early detection of damage. During normal operation the digital twin then provides the information needed to determine whether processes are within the optimal range and when maintenance has to be planned.
Benefits
Flexible programming and high data quality make the smart sensor system ASC AiSys® ideal for use in predictive maintenance.
The modernization of the infrastructure and the railway vehicles is a basic requirement in order to optimize track capacities and handle the increasing volume of traffic. However, the digitization of the rail systems requires reliable and intelligent sensor solutions. With its smart sensor systems ASC AiSys®, ASC offers the right answer to these challenges.
The smart sensor system ASC AiSys® combine the complete measuring task in one single system and are based on a modular concept. This includes the integrated hardware components – like reliable and proven sensor elements for determining an acceleration and/or angular velocity, the housing which is adapted to the harsh environmental conditions as well as bidirectional, wireless or wired data communication with a large number of standardized interfaces. Furthermore, the modular concept also includes the flexibly combinable software components and algorithms in order to be able to analyze and evaluate the preprocessed measurement data and feature vectors directly on the smart sensor.
Smart monitoring solutions that are working in real time not only identifying the actual conditions of railway vehicles and infrastructure. Using self-learning algorithms, predictive analyses are done to describe the progress of future material weaknesses. Thanks to the ASC’s smart sensors maintenance tasks can be scheduled and performed much more efficiently!
A smart factory uses numerous new technologies. However, this is relevant not only for the production process within a factory, but for the entire company. In addition to using a digital twin with predictive maintenance to boost system availability, there are many other applications, such as virtual initial operation: Virtual initial operation makes it possible to install and test a factory or system in a simulated environment. The system’s digital twin makes all the information available as if the system were operating in reality. All processes can therefore be designed and tested prior to actual initial operation. In addition, employees can be trained in advance and acquainted with operation of the system. This enables a seamless switch to operation of the actual system, without the need for a transition period. Smart sensors provide fundamental data for this process: In simulation mode the sensors can deliver predefined measurements with appropriate analyses in order to simulate diverse situations for operation of the system. Afterwards, the smart sensors measure real values again, without having to replace or modify the sensors.
Smart sensors from ASC provide system parameters in real time! The current condition of systems and of the entire factory is therefore transparent at all times.
Another example relates to automated guided vehicles.
Smart factories are increasingly using automated guided vehicles, which are responsible for ensuring smooth production processes through the on-time supply of components. Automated guided vehicles perform this task as if by magic.
The smart sensor systems ASC AiSys® are used here as well. They ensure safe and accident-free delivery!
Predictive maintenance also represents a technological step in shipping. Especially container ships are powered by giant engines, which – once they are started – cannot be shut down again. Very early damage detection is extremely important for the planning of repairs. Since smart sensors enable precise identification of damage, this facilitates organization of the necessary replacement parts.
But the sensors also measure and process operating parameters and ambient conditions such as temperature, wind and sea state with the help of artificial intelligence. The result is optimal long-term adaptation of operating parameters, which also significantly reduces fuel consumption.
Smart sensor systems ASC AiSys®, with their robust design, high measuring accuracy and flexibility, offer the ideal basis for increased efficiency and availability!
Uniaxial, capacitive
Measurement range: ±2 to ±200 g
Noise density: 10 to 680 µg/√Hz
Frequency range (±5 %): DC to 2900 Hz
Triaxial, capacitive
Measurement range: ±2 to ±400 g
Noise density: 7 to 400 µg/√Hz
Frequency range (±5 %): DC to 2000 Hz