A novel design and implementation of a wireless sensor network aimed at monitoring the vibrations produced by oil and gas activities

Abstract

This paper describes a novel, battery-powered Wireless Sensor Network (WSN) architecture specifically targeted at detecting and measuring vibrations produced by Oil and Gas exploration and production activities (e.g. seismic acquisition campaigns, drilling of wells, large compression units, etc.) to prevent structural damage to equipment and infrastructure. The WSN is self configurable and optimized for very low power consumption. Some network nodes ("sensor nodes") are equipped with 3D-accelerometers to measure the vibrations while other nodes act as relays or routers. A gateway connects the WSN with a back-end server, where the data are consolidated and processed. Both types of nodes are equipped with the same microprocessor and a radio operating within the IEEE 802.15.4 specifications for low-rate wireless personal-area networks. Each "sensor node" is equipped with a Field Programmable Gate Array (FPGA) which provides local signal-processing capability to filter and integrate the acceleration signal, compute the frequency spectrum, identify the maximum vibration velocity and the corresponding dominant frequency as required by the DIN 4150-3 standard [1]. The velocity spectrum is compared with DIN 4150-3 thresholds for industrial buildings, dwellings and historical buildings. If such thresholds are exceeded, an alarm is fired and the full acceleration waveform is transmitted to the gateway. Otherwise, only the maximum absolute velocity and dominant frequency values are reported to the gateway once per second. Due to the demanding transmission requirements of the application, a novel network stack was developed which performs node discovery, routing, synchronization and scheduling. It minimizes power consumption so as to achieve a six-month battery life. Seamless integration with the back-end application environment is achieved with a publish-subscribe messaging system called Message Queue Telemetry Transport Protocol (MQTT-S) which was optimized for low-power sensor networks. The vibration sensing units received a calibration certificate for the most critical frequencies (1-40 Hertz) of the DIN 4150-3 standard. The field tests demonstrated a high reliability of the network with a packet delivery ratio of over 99.9 %.