4G acquisition instrument

Abstract

The 4G acquisition instrument comprises a processor, and a core scheduling module, a data processing layer, a configuration management system and five driving modules are implanted into the processor; the processor adopts a Harvard architecture; the processor is connected with the RS485 communication module and communicates with the upper computer through the RS485 communication module, and the upper computer selects the types of sensors to be collected and issues corresponding parameter instructions to the sensors; the processor is connected with the 4G communication module, and the 4G communication module communicates with a server side of the cloud platform. The data acquisition functions of various types of sensors are fused, the aspects of resisting strong magnetic field interference, vibration interference, noise interference and temperature interference are improved, multi-channel synchronous data acquisition is achieved through the core scheduling module, the data processing layer, the configuration management system and the five driving modules, and rapid, stable and cooperative transmission of data is reliably achieved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent sensing data acquisition and communication technology, and involves multiple industries such as industrial production process monitoring, water conservancy and hydropower facility monitoring, geological disaster early warning, building structure safety monitoring, energy and power system status perception, and smart city infrastructure management. Specifically, it involves a 4G data acquisition device that can comprehensively acquire data from bridge sensors, 485 interface water level sensors, tilt sensors, liquid level sensors, vibrating wire sensors, and current sensors. Background Technology

[0002] In monitoring and control systems across numerous fields, the accurate acquisition and efficient transmission of sensor data are crucial. With technological advancements, various sensors are widely used in industries such as industrial production, water conservancy projects, building monitoring, and energy management to obtain key physical parameter information. For example, bridge sensors are commonly used to measure mechanical quantities such as pressure and strain, playing a key role in monitoring the stress state of structures in fields like machinery manufacturing and bridge construction. Water level, tilt, and liquid level sensors with 485 interfaces play indispensable roles in water level monitoring of water conservancy and hydropower facilities, geological landslide monitoring, and liquid level monitoring of industrial storage tanks, respectively. Vibrating wire sensors, due to their high precision and stability, are widely used in geotechnical engineering and dam safety monitoring to measure parameters such as stress, strain, and pressure. Current sensors in power systems monitor current, which is fundamental to ensuring the normal operation of power equipment and the rational allocation of energy.

[0003] However, traditional data acquisition methods have many limitations. Early data acquisition devices were often single-function, and while the Harvard architecture significantly improved computer speed, throughput, and real-time processing capabilities through the physical separation of instructions and data and parallel access mechanisms, it also suffered from high hardware complexity, low flexibility, limited scalability, and difficult development and debugging. Furthermore, it could only acquire data from specific types of sensors. Although it has significant advantages in specific fields (such as embedded systems and DSPs), these shortcomings limit its widespread application in general computing and complex systems. It cannot meet the needs of multiple sensors working simultaneously in practical applications. This leads to the need to deploy multiple different types of acquisition devices in complex monitoring systems, increasing costs and system complexity, and making equipment installation, maintenance, and management extremely cumbersome.

[0004] In terms of data transmission, most traditional data acquisition devices rely on wired transmission methods, such as RS232, RS485 serial communication, or Ethernet connections. While these wired transmission methods offer certain advantages in stability, they are limited by cabling constraints in practical applications, resulting in poor flexibility. This is especially true in complex environments or long-distance transmission scenarios, where cabling costs are high and construction is difficult. For example, in the monitoring of water conservancy facilities in mountainous areas, transmitting widely distributed sensor data back to the monitoring center via wired connections requires laying a large amount of cabling. This not only consumes significant manpower, resources, and time, but the cables are also susceptible to damage from the natural environment (such as weathering, animal gnawing, and flooding), affecting the reliability of data transmission.

[0005] With the development of IoT technology, higher demands are being placed on the real-time performance and remote accessibility of data. Although some data acquisition and transmission solutions based on 2G and 3G networks have alleviated the limitations of wired transmission to some extent, the data transmission rates of 2G and 3G networks are relatively low, making it difficult to meet the needs of high-definition video data and high-frequency data transmission from a large number of sensors. Furthermore, they also have shortcomings in network coverage and stability. In some remote areas or industrial environments with strong signal interference, data transmission interruptions and excessive delays frequently occur, seriously affecting the normal operation of monitoring systems and the validity of data.

[0006] In conclusion, to meet the growing demand from various industries for data from multiple types of sensors, adaptability to large external temperature variations and high noise and vibration conditions, and the ability to ensure efficient, accurate, and ultra-long-range data acquisition and transmission even in such harsh environments, it is of significant practical importance and market value to develop a comprehensive data acquisition instrument based on a 4G network that is compatible with various sensors such as bridge sensors, 485 sensors (water level, tilt angle, liquid level), vibrating wire sensors, and current sensors. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of existing technologies by providing a multifunctional, efficient, and long-range data acquisition instrument based on 4G transmission that can guarantee high efficiency and accuracy even in harsh environments. This instrument can acquire data from bridge sensors, current sensors, water level, tilt, and liquid level sensors based on the 485 interface, as well as vibrating wire sensors. The data is then quickly and stably transmitted to a cloud server via a 4G communication module. This meets the needs of various fields such as industrial production, water conservancy and hydropower, construction engineering, and energy management for real-time monitoring and remote data acquisition of multiple physical quantities. It solves the problems of limited functionality, poor compatibility, and restricted data transmission in existing data acquisition equipment, thereby improving the overall performance and intelligence level of the monitoring system.

[0008] The technical solution adopted in this invention is as follows:

[0009] This invention's 4G data acquisition device includes a processor, which embeds a core scheduling module, a data processing layer, a configuration management system, and five driver modules. The processor adopts a Harvard architecture. The processor connects to an RS485 communication module, enabling communication with a host computer. The processor also connects to a 4G communication module, which communicates with the server side of the cloud platform. The following improvements have been made to address interference from magnetic fields, vibration, and noise:

[0010] ① To counteract magnetic field interference, a Faraday cage is provided to house the various sensors.

[0011] ② To address vibration interference, the hardware includes an isolation bracket with silicone or rubber shock-absorbing pads to house the Faraday cage, processor, and RS485 communication module. For signal processing, the system simultaneously acquires raw signals from each sensor and vibration acceleration data from the built-in accelerometer connected to the processor. A fast Fourier transform is used to identify vibration frequency bands from the acceleration data, and a band-stop filter is used to eliminate interference from these frequency bands in the raw signals from each sensor.

[0012] ③ To mitigate noise interference, in terms of hardware, sound-absorbing materials are provided for each sensor. In terms of signal processing, wavelet thresholding is used for noise reduction.

[0013] Preferably, the processor is an ARM Cortex-M series processor.

[0014] Preferably, the specific process of identifying vibration frequency bands from vibration acceleration data using the Fast Fourier Transform is as follows: s1, perform low-pass filtering on the vibration acceleration data with a cutoff frequency of 500Hz; s2, divide the vibration acceleration data processed in step s1 into segments according to a preset number of segments, then process them through a window function, and then perform Fast Fourier Transform to obtain the spectrum; s3, divide the spectrum into sub-frequency bands with a preset bandwidth; s4, calculate the energy of each sub-frequency band, and mark the frequency bands whose energy exceeds 1.5 times the average energy as vibration frequency bands.

[0015] Preferably, the specific process of wavelet threshold denoising is as follows: S1, select a wavelet basis, determine the number of decomposition levels, and perform wavelet decomposition; S2, use heuristic rules to automatically calculate the threshold, and then remove high-frequency wavelet coefficients above the threshold to obtain low-frequency wavelet coefficients; S3, reconstruct the processed coefficients into a denoised signal through inverse wavelet transform.

[0016] Preferably, to address temperature interference, the processor is placed inside a housing made of thermally conductive material, and heat sinks are made on both the inner and outer walls of the housing, while heat insulation structures are configured for each sensor.

[0017] More preferably, the heat insulation structure is an aerogel disposed on the sensor housing.

[0018] Preferably, the core scheduling module, data processing layer, configuration management system, and five driver modules realize multi-channel synchronous data acquisition, as detailed below:

[0019] Driver module one processes current sensor signals. After the ADC acquires the voltage value, it converts the acquired voltage value into a current value using calibration coefficients and a linear calibration formula. Driver module two processes vibrating wire sensor signals. It captures the vibration period using timer one, calculates the frequency, and then converts it into a measurable physical quantity using the calibration formula. Driver module three processes bridge sensor signals. After configuring a Wheatstone bridge circuit, it measures the voltage difference using a differential ADC, eliminates zero-point drift using a polynomial fitting formula, and then converts it into a measurable physical quantity. Driver module four processes RS485 interface sensor signals. It parses the register data of each sensor using the Modbus RTU protocol and accesses each RS485 interface sensor in an orderly manner through a polling communication mechanism. For signals returned by RS485 interface sensors, driver module four is responsible for performing physical quantity conversion operations. Driver module five processes built-in accelerometer signals. It triggers a sampling task through timer two cycles to acquire vibration acceleration data from the built-in accelerometer.

[0020] Each driver module is encapsulated as an independent task. Through the FreeRTOS task priority allocation mechanism, priority is assigned to driver module 2 (1), driver module 5 (2), driver module 1 (3), driver module 4 (4), driver module 3 (5), and data storage task (6).

[0021] The core scheduling module achieves multi-threaded synchronization through mutex locks and event flag groups; mutex locks are used to control the writing of data to the independent buffer, RS485 communication and parameter updates. The vibrating wire sensor is triggered to collect data synchronously through the interrupt of Timer 3. Timer 3 interrupt sets the event flag and directly triggers the second driver module to drive the vibrating wire sensor to collect data. After collection, the data enters the data processing layer.

[0022] The data processing layer adopts a five-stage pipeline architecture. Stage 1 checks the CRC checksum and measurement range of the raw data. If the check fails, an alarm is triggered and the data is temporarily stored in an independent buffer. Stage 2 encapsulates the physical quantities uploaded by each drive module in a unified format. Stage 3 performs digital filtering. Stage 4 detects outliers: based on threshold comparison, if an outlier is found, the abnormal data is marked, an alarm is triggered, and the data is temporarily stored in an independent buffer. Stage 5 performs FFT spectrum analysis on the vibration acceleration data from the built-in accelerometer and outputs the frequency components to an independent buffer.

[0023] The configuration management system communicates with the core scheduling module via API. The host computer sends parameter update commands to the configuration management system via API. The configuration management system encapsulates the commands as task events and passes them to the core scheduling module through FreeRTOS queues. The core scheduling module then notifies the corresponding driver module to update the task parameters.

[0024] Preferably, a cloud platform is designed, and the cloud platform has the following functions:

[0025] (1) The processor packages the collected sensor data into a data packet according to the preset data format and sends it to the server through the 4G communication module. After receiving the sensor data sent by the 4G data acquisition device, the server of the cloud platform stores and processes it.

[0026] (2) The front-end interface of the cloud platform displays the data change graph of each sensor, the remaining power of the 4G data acquisition device, and whether data is being collected in real time.

[0027] (3) Control commands are issued through the front-end interface of the cloud platform. The control commands include the acquisition time interval, acquisition time, and sensor type. The control commands are transmitted to the processor for execution through the 4G communication module.

[0028] The present invention has the following beneficial effects:

[0029] This invention integrates the data acquisition functions of multiple sensor types, communicates with a host computer via RS485, and allows the host computer to select the sensor types to be acquired and issue corresponding parameter commands to these sensors. Leveraging a 4G network, this invention achieves fast, stable, and long-distance data transmission, sending sensor data to a cloud platform server for data storage and processing, data display and monitoring, remote control, and command issuance. This invention provides crucial technical support for equipment operation status monitoring, environmental parameter monitoring, resource management, and safety early warning in various industries, solving the problems of efficient acquisition, integration, and remote transmission of multi-type sensor data in different application scenarios. It meets the growing demand for intelligent and refined management, promoting the deep application and development of IoT technology in various fields. Furthermore, this invention improves resistance to strong magnetic field interference, vibration interference, noise interference, and temperature interference, significantly enhancing the high accuracy and low power consumption performance of the 4G data acquisition instrument under strong magnetic field, vibration, noise, and extreme temperature environments, meeting the needs of industrial scenarios. Even further, this invention achieves multi-channel synchronous data acquisition through a core scheduling module, data processing layer, configuration management system, and five drive modules, reliably realizing fast, stable, and collaborative data transmission. Attached Figure Description

[0030] Figure 1This is a schematic diagram showing the connection between the 4G data acquisition device of this invention and the host computer, cloud platform, and various sensors.

[0031] Figure 2 This is an architecture diagram of the data processing layer in this invention.

[0032] Figure 3 This is a flowchart illustrating the communication between the configuration management system and the core scheduling module in this invention, which drives the module to update parameters. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0034] The 4G data acquisition device of the present invention includes a processor, which incorporates a core scheduling module, a data processing layer, a configuration management system, and five driver modules. The processor can be an ARM Cortex-M series, preferably an ARM Cortex-M7.

[0035] The 4G data acquisition device of this invention achieves fast, stable, and long-distance data transmission through the following design:

[0036] 1. Adopting the Harvard architecture

[0037] The processor uses the Harvard architecture and supports data acquisition from various sensors, such as 4-20mA current sensors, vibrating wire sensors, bridge sensors, and 485-type sensors. The Harvard architecture features independent instruction and data buses, providing multiple peripherals for parallel instruction fetching and data access, enabling efficient and accurate acquisition and processing of sensor data.

[0038] 2. Use RS485 communication to communicate with the host computer.

[0039] The processor connects to an RS485 communication module, which communicates with a host computer. The host computer selects the types of sensors to be acquired (one or more can be selected) and sends corresponding parameter commands to these sensors. RS485 communication is a serial communication based on differential signals. It has strong anti-interference capabilities, long transmission distance, effectively utilizes channel resources, reduces channel collisions, and its data transmission mode is relatively simple and direct. Therefore, it has high real-time performance and good signal integrity. Its hardware design uses isolation measures to increase data reliability and effectively reduce power consumption, making it suitable for various application scenarios such as overnight automation, intelligent buildings, and transportation. The host computer uses a simple interface design, allowing easy control of data acquisition time and selection of sensor types, and providing a clear overview of sensor data.

[0040] 3. In order to remotely observe the collected sensor data and its changes, this invention is equipped with a 4G communication module to send the sensor data to the server of the cloud platform.

[0041] The processor connects to the 4G communication module and configures the SIM card communication parameters to ensure a successful connection to the 4G network. Simultaneously, the data transmission protocol and format are set, such as using TCP / IP for data transmission. The data transmission process is as follows: the 4G data acquisition device packages the collected sensor data into data packets according to the preset data format and sends them to the server via the 4G communication module. The 4G communication module leverages its high-speed data transmission rate, low latency, large capacity, and wide coverage to ensure fast and stable data transmission. 4G communication is a communication method based on technologies such as Orthogonal Frequency Division Multiplexing (OFDM), Multiple-Input Multiple-Output (MIMO), and Long Term Evolution (LTE), enabling efficient data transmission from multiple sensors simultaneously.

[0042] The cloud platform has the following functions:

[0043] (1) Data storage and processing: After receiving the sensor data sent by the 4G data acquisition device, the cloud platform's server stores and processes the data. The sensor data is stored in the database for easy subsequent data querying and analysis. At the same time, the server can process the sensor data, such as data cleaning, anomaly detection, and data fusion, to improve the accuracy and usability of the data.

[0044] (2) Data display and monitoring: The front-end interface of the cloud platform can display the data change graphs of each sensor in real time, such as the line graph of water level height and the line graph of tilt angle. It can also display the remaining power of the 4G data acquisition device, and observe whether the 4G data acquisition device is collecting data by whether the network symbol on the front-end interface is lit.

[0045] (3) Remote Control and Command Issuance: Users can also issue control commands to the 4G data acquisition device through the front-end interface of the cloud platform. For example, the data acquisition interval, acquisition time, and sensor type of the 4G data acquisition device can be adjusted. These commands are transmitted to the data 4G data acquisition device through the 4G communication module, and the processor executes the corresponding operations to realize remote control of the 4G data acquisition device.

[0046] 4. Data acquisition in industrial environments is often accompanied by strong magnetic fields, vibrations, noise, and temperature, which cause significant interference to the sensors, resulting in insufficient accuracy. To ensure efficient, accurate, and ultra-long-range data acquisition and transmission even in harsh environments, this invention addresses these issues by making the following improvements to the data acquisition device:

[0047] ① To counter interference from strong magnetic fields, a Faraday cage is provided to house each sensor. Placing each sensor inside the Faraday cage blocks external electromagnetic interference.

[0048] ② To address vibration interference, in terms of hardware, an isolation bracket with silicone or rubber shock-absorbing pads is provided to house the Faraday cage, processor, and RS485 communication module, reducing vibration transmission. In terms of signal processing, the raw signals (target signals) from each sensor and vibration acceleration data from the built-in accelerometer connected to the processor (sampling rate ≥1kHz) are simultaneously acquired. The vibration frequency band is identified from the vibration acceleration data using Fast Fourier Transform (FFT), and interference from the vibration frequency band in the raw signals from each sensor is eliminated using a band-stop filter (integrated into the processor design). The specific process of identifying vibration frequency bands from vibration acceleration data using Fast Fourier Transform (FFT) is as follows: s1. Perform low-pass filtering (i.e., use an IIR Butterworth filter) on the vibration acceleration data with a cutoff frequency of 500Hz to eliminate high-frequency noise; s2. Divide the vibration acceleration data processed in step s1 into segments according to a preset number of segments, then process it using a window function (e.g., a Hamming window with a window length of 256 points and 50% overlap), and then perform Fast Fourier Transform to obtain the spectrum; s3. Divide the spectrum into sub-bands with a preset bandwidth (e.g., 10Hz); s4. Calculate the energy (sum of squared spectral amplitudes) of each sub-band, and mark the frequency bands with energy exceeding 1.5 times the average energy as vibration frequency bands.

[0049] ③ Regarding noise interference, in terms of hardware, sound-absorbing materials (aluminum foam) are used to wrap the outer shell of each sensor to reduce noise interference. In terms of signal processing, wavelet threshold denoising is used to separate and eliminate noise components in the frequency domain. The specific process is as follows: S1, Select a wavelet basis (Daubechies is selected), determine the number of decomposition levels, and perform wavelet decomposition, in which the number of decomposition levels is determined by trial and error; S2, Use heuristic rules (such as the minimaxi threshold criterion) to automatically calculate the threshold, and then remove the high-frequency wavelet coefficients (noise-dominant regions) that are higher than the threshold to obtain low-frequency wavelet coefficients (signal-dominant regions); S3, Reconstruct the denoised signal from the processed coefficients through inverse wavelet transform.

[0050] ④ To address temperature interference, the processor is placed inside a casing made of thermally conductive material, such as aluminum-magnesium alloy with high thermal conductivity, and heat sinks are made on both the inner and outer walls of the casing. Simultaneously, thermal insulation structures are configured for each sensor; specifically, a thermal insulation layer (such as aerogel) can be placed on the sensor casing to isolate it from high-temperature areas and reduce heat conduction.

[0051] The 4G data acquisition device of this invention is connected to the various sensors as follows: The 4-20mA current sensor, vibrating wire sensor, bridge sensor, and 485 interface (water level, tilt angle, liquid level) sensors are respectively connected to the corresponding interfaces of the 4G data acquisition device. For example, the 4-20mA current sensor is connected to the analog input port, the vibrating wire sensor is connected to the dedicated vibrating wire signal processing port, the bridge sensor is connected to the bridge signal conditioning port, and the 485 type sensors are connected via the RS485 interface.

[0052] like Figure 1 As shown, after the 4G data acquisition device connects to each sensor, the processor core starts up and achieves multi-channel synchronous data acquisition through the core scheduling module, data processing layer, configuration management system, and five driver modules, as detailed below:

[0053] The driver module processes the 4-20mA current sensor signal. After the ADC acquires the voltage value, it converts the acquired voltage value into a current value by combining the calibration coefficient K and the linear calibration formula.

[0054] I = (V - V0) * K

[0055] Where V is the collected voltage value, V0 is the zero-point voltage offset, and I is the converted current value.

[0056] Drive module two processes the vibrating wire sensor signal, captures the vibration period T through timer one, calculates the frequency f = 1 / T, and then converts it into a measurable physical quantity (such as stress, strain, pressure, displacement) using calibration formulas. The calibration formulas are as follows:

[0057]

[0058] Where L is the physical quantity after frequency f is converted, C is the calibration coefficient, and f0 is the unloaded initial frequency (i.e., the natural vibration frequency of the vibrating wire sensor).

[0059] The third driver module processes the bridge sensor signal. After configuring the Wheatstone bridge circuit, it measures the voltage difference using a differential ADC and eliminates zero-point drift using a polynomial fitting formula. Then, it converts the signal into a measurable physical quantity (such as pressure or strain). The polynomial fitting formula is as follows:

[0060] Y = aX 2 +bX+c

[0061] Where X is the voltage value after ADC conversion, Y is the physical quantity after voltage value X conversion, and the values ​​of coefficients a, b, c are obtained by fitting multiple sets of standard X and Y values ​​during calibration.

[0062] Driver module four processes signals from RS485 interface sensors (water level, tilt angle, liquid level sensors), using the Modbus RTU protocol to parse the register data of each sensor. Since the RS485 bus supports multi-device serialization, this module implements a polling communication mechanism between the master station and multiple slave devices, accessing each sensor in an orderly manner based on the slave device address, ensuring bus communication stability and data timeliness. For signals returned from the RS485 interface sensors, driver module four is responsible for performing physical quantity conversion operations. The physical quantity conversion formulas are as follows:

[0063] Y1=Y0×t+h0

[0064] Where Y1 is the physical quantity after conversion of Y0, Y0 is the physical quantity before conversion (i.e., the signal returned by the sensor from the RS485 interface), t is the coefficient, and h0 is the offset.

[0065] It should be noted that this polling task is specifically for sensors with RS485 interfaces, because these sensors communicate via a shared bus and must obtain data through a master-slave polling mechanism; while other types of sensors are point-to-point independent connections and do not involve bus conflicts, so they do not require a polling mechanism.

[0066] In addition, the drive module five processes the built-in accelerometer signal. This module triggers a sampling task through a timer cycle of two to acquire vibration acceleration data from the built-in accelerometer.

[0067] Each driver module is encapsulated as an independent task. The core scheduling module uses the FreeRTOS task priority allocation mechanism to assign priority 1 (highest priority) to driver module 2 (for the vibrating wire sensor), priority 2 to driver module 5 (for the built-in accelerometer), priority 3 to driver module 1 (for the 4-20mA current sensor), priority 4 to driver module 4 (for the RS485 interface sensor), priority 5 to driver module 3 (for the bridge sensor), and priority 6 (lowest priority) to the data storage task. The Modbus RTU protocol polling task is a subtask of driver module 4, and its priority (priority 4) is lower than the vibrating wire sensor task, the built-in accelerometer task, and the current sensor task, but higher than the data storage task and the bridge sensor task, thus ensuring a balance between communication and storage.

[0068] The core scheduling module implements multi-threaded synchronization through mutexes and event flag groups. Mutexes coordinate task access to shared resources; they are used during data acquisition and buffer writing, RS485 interface communication, and parameter updates to prevent access conflicts. Event flag groups are used for synchronization notifications between tasks, especially for vibrating wire sensors requiring strict synchronization. Synchronization is triggered by a timer interrupt: the timer interrupt sets the event flag and directly triggers driver module two to drive the vibrating wire sensor to acquire data, ensuring a timestamp error of less than 1ms. After acquisition, the data immediately enters the data processing layer.

[0069] like Figure 2 As shown, the data processing layer adopts a five-stage pipeline architecture. Stage 1 is raw data verification: checking the CRC checksum and measurement range. If the check fails, an alarm is triggered and the data is temporarily stored in an independent buffer. Stage 2 is encapsulation: encapsulating the physical quantities uploaded by each driver module in a unified format. Stage 3 is digital filtering: eliminating noise through moving average. Stage 4 is outlier detection: based on threshold comparison, if an outlier is found, the abnormal data is marked, an alarm is triggered, and the data is temporarily stored in an independent buffer. Stage 5 is FFT spectrum analysis: performed only on the vibration acceleration data of the built-in accelerometer, outputting the frequency components to an independent buffer. The configuration management system then extracts data from the buffer for vibration feature identification and structural response analysis.

[0070] like Figure 3 As shown, the configuration management system resides at the application layer and communicates with the core scheduling module via API. The host computer sends parameter update commands (such as the calibration coefficient K of the current sensor in driver module one) to the configuration management system via API. The configuration management system encapsulates the commands as task events and passes them to the core scheduling module through the FreeRTOS queue. The core scheduling module then notifies the corresponding driver module to update the parameters.

Claims

1. 4G acquisition instrument comprising a processor, characterized in that: The core scheduling module, the data processing layer, the configuration management system and the five driving modules are implanted in the processor; the processor adopts the Harvard architecture; the processor is connected with the RS485 communication module, and communicates with the upper computer through the RS485 communication module; the processor is connected with the 4G communication module, and the 4G communication module communicates with the server end of the cloud platform; the following improvements are made for the magnetic field, vibration and noise interference: ①For the interference of the magnetic field, a Faraday cage for placing various sensors is provided; ②For the interference of the vibration, in the hardware aspect, an isolation support with silica gel or rubber shock-absorbing pad for placing the Faraday cage, the processor and the RS485 communication module is provided; in the signal processing aspect, the original signals of various sensors and the vibration acceleration data of the built-in accelerometer connected with the processor are synchronously collected, the vibration frequency band is identified from the vibration acceleration data through fast Fourier transform, and the interference of the vibration frequency band in the original signals of various sensors is eliminated through a band-stop filter; ③For the interference of the noise, in the hardware aspect, sound-absorbing materials for various sensors are provided; in the signal processing aspect, a wavelet threshold denoising is used. The model of the processor adopts the ARM Cortex-M series.

2. The 4G harvester of claim 1, wherein: The specific process that the fast Fourier transform identifies the vibration frequency band from the vibration acceleration data is as follows: s1, low-pass filtering is performed on the vibration acceleration data with a cutoff frequency of 500Hz; s2, the vibration acceleration data processed in step s1 is segmented according to a preset number of segments, and then is processed through a window function and subjected to fast Fourier transform to obtain a frequency spectrum; s3, the frequency spectrum is divided into sub-frequency bands with a preset bandwidth; 3. The 4G harvester of claim 1, wherein: s4, the energy of each sub-frequency band is calculated, and the frequency band with energy exceeding 1.5 times of the average energy is marked as a vibration frequency band. The specific process of the wavelet threshold denoising is as follows: S1, a wavelet base is selected, the number of decomposition layers is determined, and wavelet decomposition is performed; S2, a threshold value is automatically calculated using a heuristic rule, and then high-frequency wavelet coefficients higher than the threshold value are removed to obtain low-frequency wavelet coefficients; S3, the processed coefficients are reconstructed into a denoising signal through inverse wavelet transform.

4. The 4G harvester of claim 1, wherein: In addition, for the interference of the temperature, the processor is placed in a shell made of heat-conducting material, and heat sinks are made in the shell and on the outer wall, and a heat insulation structure for various sensors is configured.

5. The 4G harvester of claim 1, wherein: The heat insulation structure is aerogel arranged in the shell of the sensor.

6. The 4G harvester of claim 5, wherein: The core scheduling module, the data processing layer, the configuration management system and the five driving modules realize multi-channel synchronous data acquisition, and the specific process is as follows:

7. The 4G harvester of claim 1, wherein: The driving module one processes the current sensor signal, the voltage value is collected through the ADC, and then the collected voltage value is converted into a current value by combining a calibration coefficient and a linear calibration formula; the driving module two processes the vibrating string sensor signal, the vibration period is captured through the timer one, the frequency is calculated, and then the measured physical quantity is converted through a calibration formula; the driving module three processes the bridge sensor signal, a Wheatstone bridge circuit is configured, the voltage difference is measured through a differential ADC, and then the measured physical quantity is converted by using a polynomial fitting formula to eliminate zero drift; ​ The driving module four processes the sensor signal of the RS485 interface, adopts the Modbus RTU protocol to analyze the register data of each sensor, and sequentially accesses the sensors of the RS485 interface through a polling communication mechanism; for the signal returned by the sensor of the RS485 interface, the driving module four is responsible for performing a physical quantity conversion operation; the driving module five processes the built-in accelerometer signal, and periodically triggers a sampling task through the timer two to obtain the vibration acceleration data of the built-in accelerometer; Each driving module is packaged as an independent task, and through the task priority allocation mechanism of FreeRTOS, the driving module two is allocated priority 1, the driving module five is allocated priority 2, the driving module one is allocated priority 3, the driving module four is allocated priority 4, the driving module three is allocated priority 5, and the data storage task is allocated priority 6; The core scheduling module realizes multi-thread synchronization through a mutex and an event flag group; a mutex is added for control in the process of data writing into an independent buffer, RS485 communication and parameter updating, the vibrating string sensor is synchronously collected through the timer three interrupt trigger, the timer three interrupt is set, the event flag is set, and the driving module two is directly triggered to drive the vibrating string sensor to collect, and after collection, the data enters the data processing layer; The data processing layer adopts a five-stage pipeline architecture, stage one is to check the CRC check code and the range of the original data, if the check fails, an alarm is triggered and the data is temporarily stored in an independent buffer; stage two is to uniformly package the physical quantities uploaded by each driving module; stage three is to perform digital filtering; stage four is to detect abnormal values: based on threshold comparison, if an abnormal value appears, the abnormal data is marked, an alarm is triggered, and the data is temporarily stored in an independent buffer; stage five is to perform FFT spectrum analysis on the vibration acceleration data of the built-in accelerometer, and output the frequency components to an independent buffer; The configuration management system communicates with the core scheduling module through an API; The host computer sends a parameter update instruction to the configuration management system through an API, the configuration management system packages the instruction as a task event, delivers it to the core scheduling module through the queue of FreeRTOS, and the core scheduling module notifies the corresponding driving module task to update the parameters.

8. The 4G harvester of claim 1, wherein: A cloud platform is designed, which has the following functions: (1) The processor packages the collected sensor data into data packets according to a preset data format, and sends them to the server end through a 4G communication module; after receiving the sensor data sent by the 4G acquisition instrument, the server end of the cloud platform stores and processes the data; (2) The front-end interface of the cloud platform displays the data change graph of each sensor, the remaining power of the 4G acquisition instrument, and whether it is collecting data in real time; (3) Control instructions are issued through the front-end interface of the cloud platform, including collection time interval, collection time, and collected sensor type; the control instructions are transmitted to the processor through the 4G communication module for execution.

Images