drill_riser

Data logger for drill riser subsea monitoring

We help clients by providing drill riser subsea monitoring embedded software solutions for oil and gas industries.

Clients requirement

  • To develop a drill riser subsea monitoring system that can continuously collect data from various sensor.
  • Perform a kalman filtering on 1024 readings per sensor
    Store the data in primary memory.
  • Perform Power management algorithm by keeping the CPU in sleep mode and wake mode at regular intervals.
  • Transfer data to the remote location for real time data monitoring via RS232, RS485 interface.

Challenge

  • To choose micro-controller that supports multi thread operation. while consumes less power during run time.
  • Supports multiple interfaces like i2c, spi, RS232, RS485.
  • Should have enough internal memory for performing kalman filtering algorithm.
  • Should interface with external memory for storing sensor data.
  • Should be capable of sending sensor data to remote device on request.

Our Solution

  • Atmel cortex M4 with RS232, RS485, SPI, i2C, RTC interface and with 256KB internal memory which supports FreeRTOS for multithreaded functionality.
  • Sensor PHY chipset that queues 512 readings and can interrupt the CPU on completion of 512 readings.
  • Host CPU sleeps in this duration and wakes up on interrupt reads the samples stores in the external memory.
  • On completion of 1024 readings per sensor perform kalman filter algorithm on the samples.
  • Send the samples to the remote machine on request. via ethernet interface.

Advantages

  • Data logger can be automatic so no human presence is necessary
  • Readings can be continually taken over long periods
  • There is no chance of somebody forgetting to take a reading
  • Readings will be more accurate if a computer takes them because there is no chance of human error
  • Sensors can be placed in locations where it is dangerous for humans to go
  • Readings can be taken very quickly [several readings every second from a number of different sensors]. Humans could not achieve this.