Distributed Sand Monitoring Framework Using Wireless Sensor Networks

Abstract

Produced sand in oil pipelines poses a serious problem in many production situations, since a small amount of sand in the produced fluid can result in significant erosion in a very short time stage. Reliable sand detection is an important component of oil production systems; it would be valuable for assisting with optimizing well productivity and detecting sand as early as possible. In this work, a new reliable distributed sand-monitoring framework using a Wireless Sensor Network (WSN) is proposed. The framework combines three modules: Data Acquisition (DA) module, Wireless Receiving and Transmission (WRT) module, and Distributed Data Fusion module (DDF). The WRT module implementation is based on TinyOS and Crossbow MICAz mote. The DDF module is based on a resource-aware Distributed Kalman Filter (DKF). The data is collected from pipeline using acoustic sensor (SENACO AS100) and Flow Analyzer (MC-II) in real time. The experimental results validate the proposed framework. Sand production is considered as one of the major problems facing the petroleum industry. In the oil and gas industry, the sand production causes considerable erosion damage in the well tubing, piping, fittings, separators, valves, and other equipment. It can cause poor performance in injection wells, and can lead to lost production. It arises in the case of failure of sand control measures. Sand screening is also a critical part of the mining process (1). Every year, cleaning and repairing operations related to sand production and restricted production rates cost the industry millions of dollars. However, a reliable sand monitor system is essential to decide whether sand control measures need to be installed during well operations or not. Sand monitoring allows timely actions to be taken to handle sand production such as: (a) Increasing inspection to detect erosion; (b) Reducing the flow rate or stopping the production in some of the extreme cases; (c) Installation of sand handling systems (2). The sand problem is not new and affects the entire oil industry. In some cases several tones of sand can come from a reservoir in one day. Deciding when it is necessary to deal with sand is a difficult decision. Several fields have installed a sand detection system. Installation of a system to monitor and measure sand production would be helpful to detect sand as early as possible. Early sand detection would then lead to possible remedial action that could avoid the erosion and improve the system production. However, to be able to do this, we need repeatability, sensitivity, and also real time measurements (3). Oil and gas companies started to integrate their legacy sensors with Wireless Sensor Networks (WSNs) technology to help them to increase production, streamline operations, and reduce expenses. We introduced some solutions toward this direction. In (4), a reliable prototype of a remote measuring of flow rate using a MC-II Flow Analyzer was introduced. In (5), we developed a Wireless Sensor Data Acquisition (WSDA) prototyping. It composed of both legacy and modern sensors. Some other solutions introduced centralized and distributed fusion algorithms for sand detection using WSN (6-9). These fusion algorithms include Fuzzy Art (FA), Maximum Likelihood Estimator (MLE), Moving Average Filter (MAF), and Distributed Kalman Filter (DKF). The DKF showed promising results over the other algorithms. Unlike the other algorithms, Kalman filter does not lend itself for easy implementation in WSN; this is because WSN has limited computation capability, small memory, and limited energy. DKF involves seventeen matrix operations: ten matrix multiplications, two matrix inversions, four matrix additions, and one matrix subtraction. These operations are computationally intensive and could strain the energy resources of any single computational node in WSN. In order to share the information gathered by all the sensors, we encounter fundamental problems: communication congestion and scalability. The scalability issue and communication congestion are closely related in DKF algorithm. The more sensors we add to our system, the more communication we will require. Moreover, the scalability problem is not only related to communication issues, but also to computation problems. Higher dimensional measurement vectors mean higher computational demands. In this paper, a new distributed sand-monitoring framework using WSN is proposed. In order to meet with constraints of WSN, a resource-aware DKF is proposed. The proposed DKF is based on fast polynomial to increase the convergence rate. Fast convergence contributes to energy saving because it saves time. In addition, we reduce the communication traffic by exchanging the estimates between the nodes instead of exchanging the estimates and its variance. Each node uses its measurement, its associated variance, and its neighbors' estimates. This leads not only to significant energy saving, but also reduces the memory usage and avoids congestion. Moreover, we redesigned the DKF to reduce the number of arithmetic operations. Thus, the proposed optimizations make the DKF applicable to be implemented in WSN. The experimental results validate the proposed framework.