Design and Implementation of an Embedded System for Pipeline Inclination Monitoring in Foundation Pits

Description

Research hypothesis: We hypothesized that a cable‑free, STM32‑based inclinometer with integrated hoisting and 180° mechanical rotation can achieve high‑precision, repeatable inclination measurements for foundation pit pipelines, while eliminating sensor zero‑drift without extra electronic circuits. What the data show: - Figure 10 Data.xlsx: Eleven consecutive inclination measurements (0–10 cycles) at the same depth point inside a fixed vertical inclinometer casing. Measured angles (°) demonstrate repeatability. - Figure 11 Data.xlsx: Three test groups at true angle 2.5°. Forward and reverse measurements are provided; compensated angle = (Forward – Reverse)/2, showing zero‑drift removal. - STM32‑CODE (Master Control Board & Probe): Complete firmware for STM32WLE5JC (master) and STM32F103C8T6 (probe). Implements FreeRTOS tasks, Bluetooth communication, finite state machine, moving average filter, and motor control. Notable findings: - Maximum deviation among 11 repeats is –0.300° (Figure 10), confirming excellent repeatability. - Compensated angles (e.g., 2.482°, 2.472°, 2.4305°) deviate <0.03° from true 2.5°, reducing zero‑drift by ≈10× (Figure 11). - The firmware enables fully automated “descend – measure – ascend – rotate – remeasure” cycles with wireless data transmission. How data were gathered: - Inclinometer tube was fixed vertically (Fig.10) or at 2.5° (Fig.11). Probe moved by hoisting mechanism; depth position monitored via microswitches. - For Fig.10: 10 automatic cycles; angle recorded at the same depth each cycle. - For Fig.11: Probe lowered to same depth → forward measurement → 180° mechanical rotation → reverse measurement. True angle verified by reference inclinometer. - All raw readings transmitted via Bluetooth to ground unit; no post‑processing filtering applied to these values. - Code developed in Keil MDK, using STM32 HAL libraries. Master board controls motors, reads sensors, runs state machine; probe board reads SCA103T sensor and responds to Bluetooth commands. How to interpret and use the data: - Figure 10: Calculate standard deviation to assess repeatability. Small variation validates mechanical and wireless stability. - Figure 11: Apply (Forward – Reverse)/2 to any opposite‑direction pair to obtain drift‑free angle. This mechanical compensation works without temperature calibration. - Firmware: Can be reused or modified for other automated inclinometer or motor‑controlled sensor platforms. The state machine and FreeRTOS task architecture are extensible. Corresponding author: Shuai Zhang (zhangshuai@nynu.edu.cn)

Steps to reproduce

We developed a cable‑free, fully automated inclinometer system based on an STM32 embedded platform for monitoring pipeline inclination in foundation pits. Data were collected in a controlled laboratory environment using a PVC inclinometer tube fixed at a known angle (2.5° for zero‑drift test). Measurement procedure for zero‑drift test (data in Figure 11): - The system performed a complete automated cycle: descent → forward measurement → ascent → 180° rotation → descent → reverse measurement → data upload. - For each test group, the probe was lowered to the same depth and the following values were recorded: * Forward measurement (θ_forward): angle read before rotation. * Reverse measurement (θ_backward): angle read after 180° rotation. - Compensated angle θ_comp was calculated as (θ_forward – θ_backward)/2. - Each test was repeated 3 times; raw values are provided. Hardware and software: - Master control board: STM32WLE5JC microcontroller (Cortex‑M4 core). - Sensor board: STM32F103C8T6 with SCA103T‑D05 MEMS inclinometer (SPI interface). - Communication: Bluetooth between master and probe. - Motor control: TB6612 driver for DC geared motors (hoisting + 180° rotation). - Software: FreeRTOS real‑time OS for task scheduling; moving average filter (window size N=8) for noise suppression. The raw data in this dataset are directly read from the sensor without additional post‑processing filtering. The compensated value is calculated as (Forward – Reverse)/2 to eliminate zero‑drift.

Institutions

• Nanyang Normal University

  Henan, Nanyang

Categories

Licence