Embedded processing unit while drilling based on CompactRIO

The Challenge:
The embedded processing unit while drilling is the core of the surface system while drilling. It is mainly responsible for two major tasks: first, real-time acquisition, processing and decoding of mud pulse telemetry signals; second, accurate real-time measurement of drilling depth. Mud pulse remote signal and depth signal are mixed with a lot of environmental interference. The development of a processing algorithm that can effectively filter out these environmental interferences and restore the original signal is the biggest challenge for the entire processing unit, which has considerable complexity. Therefore, a real-time embedded system with powerful performance, stable and reliable work, simple and fast development is needed as the core of the unit to reduce the investment of development engineers in hardware-related aspects and concentrate limited resources on the development of processing algorithms.
　
The Solution:
Using NI LabVIEW and related modules, combined with NI CompactRIO and related data acquisition modules, developed a set of embedded processing unit while drilling to collect the signals of mud pressure sensors, depth coding discs, deadline sensors, hook load sensors, and pumping sensors installed on the drilling site. After processing, the mud pulse remote transmission data and drilling depth data are obtained, and then sent to the upper computer for post-processing.
　
Author(s):
Di’nan Jiang – China Oilfield Services Limited
Songwei Zhang – China Oilfield Services Limited
Huatao Lu – China Oilfield Services Limited
　
For many years, downhole data transmission has been a bottleneck restricting the development of logging while drilling technology. At present, the more mature data transmission method is mud pulse remote transmission. One of the core tasks of the surface-while-drilling system is to collect mud pulse signals for real-time acquisition and processing, and decode them into raw data; at the same time, another core task is to drill through the depth coding disk with the cooperation of hook-loaded or dead-line sensors. Real-time depth measurement. Using NI CompactRIO as the core of the embedded processing unit while drilling, with its powerful processing capabilities, real-time performance, rapid development features, and robustness and reliability, engineers can focus more on mud pulse signal processing, decoding algorithms, and depth measurement Algorithm development. At present, the MWD embedded processing unit has passed the ground cycle test and the downhole cycle test for two hundreds of hours of operation. It works stably and functions normally, and fully meets the requirements of the LWD surface system.
　
Development background
Logging while drilling technology is one of the key technologies for oil and gas field exploration and development, and is currently monopolized by several large international companies. In the domestic market for logging while drilling and related directional drilling services, foreign technology occupies a dominant position, and the domestic offshore market is completely monopolized by foreign technology. In recent years, it is estimated that the overseas logging-while-drilling equipment purchased by domestic companies will cost hundreds of millions of RMB each year. The logging-while-drilling equipment including gamma/resistivity/neutron density/MWD (downhole tools are divided into 2 Support configuration) The price exceeds 40 million yuan, not only the high purchase and maintenance costs, but also the dual restrictions of technology and the market, to a considerable extent, restrict the exploration and development of my country’s oil and gas resources, especially offshore oil and gas fields, and also restrict domestic related companies. development of. With the development of my country’s deepwater oil and gas resources exploration and development, this problem will become more serious.
　　
The successful implementation of the independent research and development while drilling project will form a relatively complete logging while drilling technology and equipment with independent intellectual property rights in my country, breaking the monopoly of foreign technology, and greatly reducing its production, maintenance and use costs compared with imported equipment. The exploration and development of offshore, especially deep-water oil and gas resources, and the development and growth of my country’s oilfield service companies provide effective technical support.
　
Mud Positive Pulse Coding Technology
The positive pulse signal is generated by temporarily blocking the passage of the mud column in the drill string and forcing the pressure to rise. Therefore, when the upward part of the pulser poppet valve blocks the mud flow, the pressure rises, and when the poppet valve returns, the drill string pressure returns to the original state, as shown in Figure 1:

Figure 1 Positive pulse signal generator

Mud positive pulse encoding is an encoding method that realizes digital encoding by adjusting the pulse interval between mud pulses. The information or data encoded by the pulse position is implicit in the pulse interval between two pulses, and different pulse interval lengths represent different information, as shown in Figure 2.

Figure 2 Positive pulse coding based on pulse interval

Data = (Interval – MIN_TIME) / BIT_TIME
Among them, Data is the coded data, and Interval is the interval time. From the above formula, the larger the coded data, the larger the pulse interval, and vice versa.
　
For an actual telemetry physical system, some parameters need to be defined. The minimum interval duration (MIN_TIME) refers to the length of time used to encode the corresponding value 0 in the data encoding. If an interval is equal to the minimum interval duration, then this interval encodes the data The information is 0. Due to the presence of noise, the pressure signal generated at the bottom of the well is different from the pressure signal on the ground. The bit width (BIT_WIDTH) is defined here to correct the transmitted value. As long as an interval pulse falls within the BW window, it is considered that no matter how much the offset is For a valid pulse, the actual pulse position is consistent with the theoretical pulse position value within the bit width.
　
System components
The embedded processing unit while drilling (hereinafter referred to as EPU) simultaneously undertakes multiple acquisition, measurement and processing tasks and system communication tasks, and is the core center of the entire surface-while-drilling system. On-site depth, mud pressure, hook load, dead rope, and pumping sensor signals enter the EPU. After isolation by the safety barrier and conditioning by the signal conditioning board, they are sent to the cRIO data acquisition card, and the data is collected by the FPGA. After the data is processed by the CPU, it is transmitted to the host computer via the LAN. At the same time, the upper computer communicates with the driller’s Display and the DBC central control through the RS485 bus through the cRIO relay, drives the Bypass solenoid valve, and downloads the instructions. The system architecture is shown in Figure 3.

Figure 3 The framework of the surface-while-drilling system

Mud pulse signal filtering and decoding
At the well site, the mud pulse signal is collected by a pressure sensor installed on the riser.
　
Under field conditions, when the mud pulse signal sent from the downhole is transmitted to the ground, it is mixed with very strong background noise. The main sources are the mud pump noise generated by the periodic reciprocating movement of the mud pump piston, and the sudden release of the drill string after the bit sticks. The drill string torque noise generated by the large arc swing and the drill bit noise generated by the vibration of the drill string during drilling.
　
In the EPU, the acquired mud pulse waveform data undergoes a series of filtering processing such as FIR primary filtering, adaptive filtering, wavelet analysis, and cross-correlation processing to obtain a relatively clean pulse signal waveform. After that, by accurately determining the pulse position and calculating the time interval between adjacent pulses, the corresponding communication data can be calculated.
　
Depth measurement
The measurement of the drilling depth is obtained indirectly by measuring the up and down movement of the rig hook. In addition to accurate and real-time measurement of the hook position, it is also necessary to accurately determine whether the drilling tool is hung on the hook, and then lift and lower it together with the hook.
　
To determine whether the drilling tool is hung on the hook, first measure the hook load and compare it with the set threshold. If the value is higher than the threshold, it is considered to be hung on the hook, otherwise it is considered that it is not hung on the hook and is in the clamped state. However, during the drilling process, various vibrations are generated, causing the measured load to fluctuate sharply. When the drilling depth is shallow and the weight of the drilling tool is light, the threshold is very close to the large hook and empty hook load, and severe fluctuations can easily cross the threshold and cause misjudgment.
　
In order to eliminate the impact of vibration, it is necessary to filter the measured hook load. The filtering algorithm requires on the one hand to effectively filter out fluctuations and on the other hand to have a very low delay. These two aspects are a pair of contradictory relationships. After integrating the characteristics of FIR, IIR, average filtering, median filtering and other algorithms, a set of effective filtering algorithms is designed using the relevant filtering module of the FPGA module, which can carry out accurate depth measurement.
　
Test situation
In the ground cycle test, downhole cycle test and actual drilling test that have been carried out, the embedded processing unit while drilling has worked for hundreds of hours, successfully verifying its reliability and real-time performance. The main performance indicators achieved are:
　
1. Transmission rate: It has reached 3.0bps, which is the highest level of similar technologies in China;
　
2. Bit error rate: During the test, the decoded bit error rate is below 1%;
　
3. Reliability: During the test work, the MWD embedded processing unit works stably, the algorithm runs normally, and there is no system crash.
　
Summarize
After testing and verification, the embedded processing unit while drilling based on CompactRIO has initially reached the level of practicality.
　
In the entire processing unit development process, the fast and convenient LabVIEW graphical programming tool greatly liberated the developer’s energy and can focus more on the development of mud pulse waveform filtering and decoding algorithms and depth measurement algorithms. At the same time, CompactRIO’s sturdy structure and extremely high reliability also left a deep impression.
　
Next Steps
Watch industrial control video demo
　
Understand NI LabVIEW graphical programming