Technology for processing heterogeneous information received via telemetry channels from control devices of technical objects of various types of basing

* different types of basing – assumes the use of different types of communication channels

Processing tasks:

analysis of various critical physical parameters of measuring equipment,
identification of functioning modes of subsystems;
prediction of normal and emergency conditions, such as moving of controlled parameters beyond the range of permissible values, an unexpected change in operating modes.

Purpose:

to improve the operational characteristics of monitoring systems for technical objects:
to provide recognition and classification of the state of subsystems and patterns of their behavior with incomplete and noisy input data
to ensure effective use of software and hardware for monitoring the state of objects and subsystems through rational planning of telemetry sessions.

Application area:

telemetry systems for vehicle monitoring and navigation,
robotic video surveillance systems,
monitoring systems for equipment of various basing via telemetry data.

Formulation of the processing task

Given:

Multidimensional time sensors series of technical objects subsystems (TO)
Description of the maintenance subsystems behavior in the form ща:

Set of parameters (sensors) of subsystems, with the maximum permissible deviations of parameter values
List of subsystems state descriptions (regular, irregular)
Simulation models of subsystems (describe their structure, parameters and behavior in time)

It is necessary:

to assess the state of the subsystem at a given time (regular, irregular)
to predict the state of the subsystem at the next point in time (regular, irregular)
to localize the place of the emergency situation occurrence

Components of technology

Formation and management of simulation models of onboard equipment and technical objects subsystems
Construction of mathematical models of monitored on-board equipment based on the analysis of sensor data and assessment of the behavior conformity of on-board equipment with the adopted model, predicting its behavior
Rapid state assessment of the on-board maintenance facilities based on parametric differential models
Identification of operation modes of maintenance subsystems and detection of abnormal situations
Monitoring the state and operation modes of onboard equipment
Diagnostics and tate monitoring of the on-board objects and maintenance subsystems based on joint analysis of data coming from maintenance subsystems and the planned behavior of their simulation models
Formation and management of a library of known contingencies and emergencies
Saving the received TMI for subsequent analysis of TM data in interactive or automatic mode

Scientific and technical groundwork.

The prototype of a neural network system for monitoring subsystems of small spacecraft based on telemetry data for the ground command and measurement complex of the Belarusian spacecraft has been developed.

The list of telemetric information contains the parameters generated by the sensors of the Belarusian spacecraft subsystems (434 sensors):

Corrective propulsion system (CPS) sensors:

temperature parameters
xenon feed unit pressure levels
electrical parameters of the flow controller, anode and cathode of motors

Power supply system (PSS) sensors:

power regulation and distribution module parameters
solar battery (SB) parameters
parameters of the lithium-ion accumulator battery (AB)

Target imaging equipment (TE) sensors:

temperature parameters of TE
electrical parameters of TE

1. Marushko, E. E. Ensembles of Neural Networks for Forecasting of Time Series of Spacecraft Telemetry / E. E. Marushko, A. A. Doudkin // Optical Memory and Neural Networks. – 2017. – Vol. 26, No. 1. – Allerton Press, Inc., 2017. – P. 47–54. – DOI: 10.3103/S1060992X17010064.

2. Spacecraft Telemetry Time Series Forecasting With Ensembles of Neural Networks / Alexandr Doudkin, Yauheni Marushko, Jan Owsiński, Tadeusz Pawlowski // Proc of the 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, 18-21 September, 2019, Metz, France