End-cloud cooperative vehicle intelligent weighing monitoring system and intelligent processing method

Abstract

The invention discloses an end-cloud cooperative vehicle intelligent weighing monitoring system and an intelligent processing method. The system adopts a three-stage processing architecture, vehicle dynamic load signals are obtained through a piezoelectric film sensor array, and after the signals are processed through a signal conditioning and multi-channel AD conversion module, high-speed data acquisition and packaging are carried out through an FPGA processing unit. And then, the main control unit carries out dynamic load analysis and running speed calculation on the data packet uploaded by the FPGA, and carries out real-time environment compensation in combination with data of the temperature sensor, so that vehicle characteristic modeling is realized. And the system automatically snapshots vehicle information when detecting that the vehicle passes. And if the system judges that the vehicle is overloaded or overspeed, integrating the data to generate a complete violation data stream. According to the invention, through an end-cloud cooperative processing mechanism and an optimized hardware architecture, and by introducing an environment compensation algorithm, the detection precision and the real-time performance of data processing are significantly improved, and a more reliable technical solution is provided for the field of dynamic weighing.

Description

Technical Field

[0001] The present invention relates to the field of intelligent traffic monitoring technology, and in particular to an end-cloud collaborative vehicle intelligent weighing monitoring system and an intelligent processing method. Background Art

[0002] Existing vehicle detection systems mostly use a single piezoelectric film combined with a local processing unit to achieve vehicle weight detection. Their hardware architecture and data processing methods have significant limitations. In traditional solutions, the sensor signal processing link lacks a dynamic environmental compensation mechanism, resulting in long-term measurement errors caused by factors such as temperature drift and electromagnetic interference, and generally low weight detection accuracy. In addition, overweight judgment and image recognition functions are mostly independent modules that rely on manual triggering or timed polling, with high response delays, making it difficult to achieve accurate linkage between overweight events and license plate capture. At the data management level, existing systems mostly use local storage or one-way data upload modes, lacking cloud-based collaborative analysis capabilities, resulting in low historical data utilization, long equipment calibration and maintenance cycles, and inability to achieve remote real-time monitoring and dynamic parameter optimization.

[0003] With the development of intelligent technologies, particularly advances in sensor technology, image recognition technology, and cloud computing, existing vehicle detection systems are facing the need for technological upgrades. However, while existing vehicle detection systems based on image recognition or piezoelectric films exist, they often lack system integration, data processing capabilities, and real-time data transmission capabilities. Furthermore, they lack compensation for external factors such as ambient temperature, which can affect the accuracy of detection results.

[0004] In summary, the prior art has the following disadvantages

[0005] 1. Poor environmental adaptability and insufficient measurement accuracy: Most systems do not integrate temperature compensation modules, and the piezoelectric film signal is significantly affected by environmental temperature drift. In particular, the error is exacerbated under extreme temperature conditions, resulting in inaccurate weight detection.

[0006] 2. Low data processing efficiency and high response latency: Traditional main control units struggle to process multimodal data such as multi-lane pressure signals, temperature data, and image streams in real time. This results in delays in coordination between overweight determination and image recognition, impacting the efficiency of tracing illegal vehicles.

[0007] 3. Overweight detection and image recognition are disconnected: Existing technologies require manual intervention to initiate image acquisition after overweight determination, or use fixed-frequency capture. This results in redundant, invalid image data, accounting for over 40%, and can easily miss vehicles passing through highways.

[0008] 4. Low cloud integration and lack of dynamic management: The data upload protocol is simple, lacks layered encapsulation and verification mechanisms, and has low transmission reliability. In addition, the cloud platform only has basic storage functions, making it impossible to achieve remote optimization of equipment calibration parameters and analysis of the spatiotemporal patterns of overweight events.

[0009] Therefore, there is an urgent need for a vehicle detection system that integrates multi-sensor dynamic compensation, edge-side real-time decision-making and cloud-side intelligent analysis. Through hardware-level signal processing optimization, overweight event-triggered image acquisition and end-cloud collaborative management mechanism, it can solve the collaborative optimization problem of measurement accuracy, response speed and data utilization efficiency, and provide high-precision and highly reliable vehicle detection and violation tracing support for smart transportation. Summary of the Invention

[0010] The technical problem solved by this application is: how to overcome the problems existing in the prior art.

[0011] In a first aspect, the present application provides a vehicle intelligent weighing monitoring system with end-cloud collaboration, comprising:

[0012] A plurality of vehicle passing detectors are respectively arranged in a plurality of lanes, and each of the vehicle passing detectors is used to sense the vehicle entering the weighing area of ​​the lane where it is located; a plurality of piezoelectric film sensors are respectively integrated in the area where the plurality of vehicle passing detectors are located or the front end area, and each of the piezoelectric film sensors is used to generate a piezoelectric analog signal corresponding to the load of the vehicle; a plurality of signal conditioning modules are respectively connected to the plurality of piezoelectric film sensors for signal connection, and each of the signal conditioning modules is used to amplify and filter the input piezoelectric analog signal to obtain a corresponding conditioning signal; a multi-channel AD conversion module is connected to the plurality of signal conditioning modules in a one-to-one correspondence, and the multi-channel AD conversion module is used to perform high-precision analog-to-digital conversion on each of the input conditioning signals to obtain a corresponding digital signal; a temperature sensor is used to collect ambient temperature and generate corresponding ambient temperature data; an FPGA processing unit is connected to the multi-channel AD conversion module and the temperature sensor signal The FPGA processing unit is connected to the main control unit for dynamic load analysis of the data packet, and for performing environmental compensation, overload determination and data visualization through intelligent decision-making; a plurality of digital image sensors are respectively arranged in a plurality of lanes and are all connected to the main control unit for signal connection, each of the digital image sensors is used to capture the vehicle in the lane, and the captured image is used for feature recognition and violation evidence collection of the vehicle, and the captured image is transmitted back to the main control unit for violation analysis and processing; a PHY communication chip is connected to the main control unit for signal connection, and is used to establish a data link between the main control unit and the cloud vehicle management platform, and transmit the violation analysis and processing results to the cloud vehicle management platform; the cloud vehicle management platform is used to record the information of the vehicle that violates the law.

[0013] Furthermore, each of the signal conditioning modules includes a Wheatstone bridge, an RC low-pass filter, an instrumentation amplifier and a zeroing circuit; the Wheatstone bridge is used to obtain the input piezoelectric analog signal, and to perform preliminary adjustment and balancing on the piezoelectric analog signal to improve the measurement accuracy, thereby obtaining a first adjustment signal; the RC low-pass filter is connected to the Wheatstone bridge, and is used to filter out high-frequency noise in the first adjustment signal to ensure the stability and accuracy of the signal, thereby obtaining a second adjustment signal; the instrumentation amplifier is connected to the RC low-pass filter, and is used to amplify the second adjustment signal through high input impedance and high common-mode rejection ratio, thereby obtaining a third adjustment signal; the zeroing circuit is connected to the instrumentation amplifier, and is used to perform zero-point correction on the third adjustment signal to ensure the accuracy and linearity of the signal, thereby obtaining a corresponding conditioning signal.

[0014] Furthermore, the FPGA processing unit is configured to: perform preliminary data acquisition on the digital signals corresponding to the multiple lanes output by the multi-channel AD conversion module, and package the pressure data, ambient temperature data and calibration status parameters of the vehicle to obtain the data packet.

[0015] Furthermore, the data encapsulation format of the FPGA processing unit includes a start character, multiple lane data fields, an ambient temperature floating point value, a CRC-16 check code and an end character.

[0016] Furthermore, the main control unit is configured to: disassemble the data packet, perform integral calculation of pressure data of multiple lanes, dynamically correct the preset piezoelectric film characteristic curve based on ambient temperature data, and calculate the vehicle's driving speed based on the signal time difference of adjacent piezoelectric films; and control the triggering of each digital image sensor to perform snapshots, and generate an overweight warning signal based on the pressure value and the standard load threshold of the vehicle model, and generate an overspeed warning signal based on the driving speed and the road section speed limit threshold.

[0017] In a second aspect, the present application provides an intelligent processing method for dynamic weighing, which is applied to the vehicle intelligent weighing monitoring system described in the first aspect above, and the intelligent processing method includes:

[0018] The data packet receiving step includes receiving a data packet, wherein the data packet includes pressure data of multiple lanes, as well as ambient temperature data and calibration state parameters; the protocol decoding step includes parsing the data packet to extract the pressure data, calibration state parameters and ambient temperature data of multiple lanes; the calibration mode determination step includes executing the system calibration flow step when the calibration state parameter is an activated value, otherwise executing the comprehensive calculation step; the system calibration step includes matching the pressure data of multiple lanes with the test pressure value, and correcting the piezoelectric film characteristic curve in combination with the ambient temperature data; the comprehensive calculation step includes integrating the pressure data of multiple lanes to obtain a total pressure value, and calculating the pressure value of each lane based on the corrected piezoelectric film characteristic curve. The weight of the vehicle in the road is calculated based on the peak time difference of adjacent piezoelectric films; the image processing step includes obtaining a captured image of the vehicle in any lane, and identifying the vehicle's license plate, model and corresponding standard load threshold through a preset neural network model; the violation judgment step includes comparing the weight of the vehicle in any lane with the standard load threshold corresponding to the model, and comparing the driving speed with the speed limit threshold of the road section to generate an overload alarm signal and / or a speeding alarm signal; the data packaging step includes structured packaging of the weight data, speed data, captured images and model information of the overloaded and / or speeding vehicles to generate data to be reported; the cloud reporting step includes uploading the data to be reported to the cloud database.

[0019] Furthermore, the data encapsulation step specifically includes: packaging the weight data, speed data, captured image and vehicle model information of the illegal vehicle in a preset field order, and appending a CRC-16 check code; generating the data to be reported after verifying the data integrity through a check mechanism.

[0020] Furthermore, the dynamic calibration compensation specifically includes: in the system calibration state, inputting the pressure signal of a test vehicle of known weight into the system; collecting the output value of the piezoelectric film at the current ambient temperature, fitting the temperature-pressure characteristic curve; and storing the fitting parameters to compensate for measurement errors in real time.

[0021] Furthermore, the image processing step also includes: after identifying the vehicle license plate and model through a preset neural network model, associating the standard load threshold corresponding to the model; if the model cannot be identified, matching the load threshold of an approximate model from historical data and using it as the corresponding standard load threshold.

[0022] Furthermore, the comprehensive calculation step specifically includes: a weight integral calculation sub-step, including time domain integration of the pressure data of each lane, and the calculation formula is expressed as

[0023]

[0024] Where W is the weight of the vehicle, k is the calibration coefficient of the piezoelectric film characteristic curve, P(t) is the pressure signal amplitude, t1 and t2 are the start and end times of the vehicle axle pressure signal respectively;

[0025] The sub-step of calculating the driving speed includes calculating the vehicle speed based on the peak time difference of adjacent piezoelectric films, using the formula

[0026] v=L / Δt

[0027] Where v is the driving speed, L is the installation distance between adjacent piezoelectric films, and Δt is the peak time difference of the same axle pressure signal on the two piezoelectric films.

[0028] The beneficial effects of the technical solution of this application are:

[0029] 1. By integrating a temperature sensor and executing an environmental compensation algorithm in the main control unit, the characteristic curve of the piezoelectric film is dynamically corrected, effectively overcoming measurement errors caused by environmental factors such as temperature drift, significantly improving the accuracy of vehicle weighing and the system's adaptability in different environments.

[0030] 2. A three-level architecture employs collaborative processing between an FPGA processing unit and a main control unit. The FPGA leverages its hardware parallel processing capabilities to rapidly collect, verify, and package multi-lane data, while the main control unit focuses on dynamic load analysis and intelligent decision-making, significantly improving data processing efficiency and ensuring real-time system response.

[0031] 3. Precise synchronization and data association between vehicle detection and image capture are achieved. The moment the system senses a vehicle passing through, it triggers the digital image sensor in the corresponding lane to capture the image, ensuring complete vehicle information is captured. The system then binds the weight and speed data analyzed in real time to the captured image. Once a vehicle is determined to have violated a traffic violation, a complete chain of evidence, including timestamp, weight, speed, and vehicle image, is packaged and reported. This ensures timely and accurate evidence collection and effectively avoids missed evidence or disconnected information due to delayed capture.

[0032] 4. Based on a device-cloud collaboration mechanism, a stable data link is established via the PHY communication chip, reliably uploading violation data to the cloud-based vehicle management platform. The cloud not only stores and records data, but also conducts in-depth data analysis and supports remote calibration and parameter optimization of equipment in the local area or front-end areas, enabling dynamic and intelligent management of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 This is a schematic diagram of the overall architecture of a vehicle intelligent weighing and monitoring system with end-cloud collaboration in one embodiment of the present application;

[0034] Figure 2This is a schematic diagram of the architecture of a signal conditioning module in one embodiment of the present application;

[0035] Figure 3 This is a schematic diagram of the processing flow of the cloud module in one embodiment of the present application.

[0036] Figure 4 This is a schematic diagram of the field architecture of an end-cloud collaborative vehicle intelligent weighing and monitoring system in one embodiment of the present application. DETAILED DESCRIPTION

[0037] The following is a detailed description of this application through specific implementation methods combined with the accompanying drawings. Figure 1 The present application discloses an end-cloud collaborative vehicle intelligent weighing monitoring system, including multiple vehicle passing detectors 101, multiple piezoelectric film sensors 102, multiple signal conditioning modules 106, temperature sensors 104, multi-channel AD conversion modules 107, FPGA processing units 108, main control units 110, multiple digital image sensors 105, PHY communication chips 109 and cloud-based vehicle management platforms 111. Each functional component will be described in detail below.

[0038] Multiple vehicle-passing detectors 101 are deployed in multiple lanes, each detecting when a vehicle enters the weighing area of ​​its lane. For example, a ground-sensing coil sensor can be used. When a metal vehicle chassis passes by, the coil inductance changes, generating a trigger signal indicating vehicle entry.

[0039] Multiple piezoelectric film sensors 102 are integrated into the area where multiple vehicle-passing detectors 101 are located or in front of them. Each piezoelectric film sensor 102 generates a piezoelectric analog signal corresponding to the vehicle's load. For example, when a vehicle tire rolls over a sensor, the piezoelectric film is compressed, generating an electric charge, which forms a voltage or current signal proportional to the pressure.

[0040] Multiple signal conditioning modules 106 are connected to the multiple piezoelectric film sensors 102. Each signal conditioning module 106 is used to amplify and filter the input piezoelectric analog signal to generate a corresponding conditioned signal. For example, the weak electrical signal generated by the piezoelectric film sensor 102 is input into a Wheatstone bridge circuit for initial signal conditioning and balancing. The signal then passes through an RC low-pass filter to remove high-frequency noise interference. The processed signal is then fed into an instrumentation amplifier for high-fidelity amplification. Finally, a zero-point correction circuit is used to obtain a clear, stable, and conditioned signal suitable for subsequent processing.

[0041] The multi-channel AD conversion module 107 is connected to multiple signal conditioning modules in a one-to-one correspondence. It performs high-precision analog-to-digital conversion on each conditioned input signal to generate a corresponding digital signal. For example, the module receives high-quality analog conditioned signals from multiple lanes and synchronously samples the signals from each lane through a high-speed polling mechanism, converting them into high-precision digital values ​​to ensure that no detail in the weight signal is lost.

[0042] The temperature sensor 104 is used to collect the ambient temperature and generate corresponding ambient temperature data. Based on this temperature data, the main control unit will call a preset compensation algorithm to dynamically correct the characteristic curve of the piezoelectric film to compensate for the measurement error caused by temperature changes.

[0043] FPGA processing unit 108, connected to the multi-channel A / D conversion module and temperature sensor signals, is used to acquire data and ambient temperature data from multiple lanes via a built-in multi-channel polling mechanism. It also verifies and packages the acquired data to generate data packets. For example, it leverages hardware parallel processing to acquire pressure and ambient temperature data from multiple lanes at high speeds through the built-in multi-channel polling acquisition mechanism. After acquisition, it verifies and packages the data, ultimately generating a standard data packet that is sent to the main control unit via a high-speed bus.

[0044] The main control unit 110, signal-connected to the FPGA processing unit, is responsible for performing dynamic load analysis on data packets and implementing environmental compensation, overload determination, and data visualization through intelligent decision-making. For example, it is responsible for receiving and parsing data packets sent by the FPGA processing unit 108, breaking them down, integrating pressure data from multiple lanes, dynamically modifying the preset piezoelectric film characteristic curve based on ambient temperature data, and calculating the vehicle's speed based on the signal time difference between adjacent piezoelectric films. Furthermore, it controls the triggering of each digital image sensor to capture images, generates an overweight warning signal based on the pressure value and the vehicle's standard load threshold, and generates an overspeed warning signal based on the driving speed and the road section speed limit threshold.

[0045] Multiple digital image sensors 105 are respectively arranged in multiple lanes and are all connected to the main control unit signal. For example, each digital image sensor is used to capture the vehicles in the lane where it is located. The captured images are used for vehicle feature recognition and violation evidence collection. The captured images are sent back to the main control unit 110 for violation analysis and processing.

[0046] The PHY communication chip 109, connected to the main control unit (MCU), establishes a data link between the MCU 110 and the cloud-based vehicle management platform 111, serving as the physical layer network interface. For example, it is responsible for establishing a stable, high-speed data link between the MCU and the cloud-based vehicle management platform, and reliably transmits violation results processed by the MCU, including complete information such as weight, speed, image, and timestamp, to the cloud.

[0047] The cloud-based vehicle management platform 111 is used to record information about vehicles violating traffic rules. For example, it receives and stores data uploaded by all devices in the local area or front-end area, and archives information about vehicles violating traffic rules. Managers can use the platform to query historical data, conduct statistical analysis, and generate reports. Furthermore, the cloud-based platform supports remote monitoring and parameter configuration of devices in the local area or front-end area.

[0048] Together, these modules form an efficient vehicle load and speed monitoring system that can collect, process and analyze a variety of environmental and equipment data in real time.

[0049] In one embodiment, see Figure 1 and Figure 2 The signal conditioning module 106 includes a Wheatstone bridge 202, an RC low-pass filter 203, an instrumentation amplifier 204, and a zero adjustment circuit 205, which are described below.

[0050] The piezoelectric film 102 is used to detect external pressure changes and convert them into piezoelectric analog signals. The electrical signal is initially conditioned and balanced by the Wheatstone bridge 202 to obtain a first conditioned signal to improve measurement accuracy.

[0051] The output signal of the Wheatstone bridge 202 enters the RC low-pass filter 203. The RC low-pass filter 203 is used to filter out high-frequency noise in the first adjustment signal to ensure signal stability and accuracy, and obtain a second adjustment signal.

[0052] The filtered signal is sent to the instrumentation amplifier 204. The instrumentation amplifier 204 amplifies the second regulated signal through high input impedance and high common mode rejection ratio to obtain a third regulated signal.

[0053] The zero adjustment circuit 205 is used to perform zero point correction on the third adjustment signal to ensure the accuracy and linearity of the signal, and obtain a corresponding conditioned signal.

[0054] Finally, the conditioned signal after zero adjustment is sent to the corresponding channel in the multi-channel AD conversion module 107. The multi-channel AD conversion module 107 is responsible for converting the analog signal into a digital signal for further digital processing and analysis.

[0055] See also Figure 4 , provides a schematic diagram of the field architecture of the end-cloud collaborative vehicle intelligent weighing monitoring system, where multiple vehicle passing detectors are arranged on each lane, so that each vehicle passing detector can sense the vehicle entering the weighing area of ​​the lane where it is located. Multiple piezoelectric film sensors are respectively integrated in the area where the multiple vehicle passing detectors are located or the front-end area, so that each piezoelectric film sensor can generate a piezoelectric analog signal corresponding to the load of the vehicle. The temperature sensor is arranged in the road surface area of ​​the lane, and can collect the ambient temperature and generate corresponding ambient temperature data. Multiple digital image sensors are respectively arranged in multiple lanes, so that each digital image sensor is used to capture the vehicles in the lane where it is located, and the captured images are used for feature recognition and violation evidence collection of the vehicles. An FPGA processing unit, a PHY communication chip and a main control unit are also provided, each of which realizes corresponding functions, which will not be repeated here.

[0056] Based on the vehicle intelligent weighing monitoring system mentioned above, an intelligent processing method for dynamic weighing is also provided here. Figure 3 Give a detailed introduction.

[0057] Step S101 is a data packet receiving step, specifically including receiving a data packet containing pressure data for multiple lanes, as well as ambient temperature data and calibration state parameters. For example, the main control unit 110 receives data packets collected and packaged by the FPGA processing unit 108 and uploaded via the PHY communication chip 109 in real time through its internal communication interface.

[0058] Step S102 is the protocol decoding step, which involves parsing the data packet to extract pressure data, calibration status parameters, and ambient temperature data for multiple lanes. For example, the main control unit 110 parses the received data packet according to a preset data frame protocol and extracts all valid information, including pressure data, calibration status, and temperature data for multiple lanes.

[0059] Step S106 is the calibration mode determination step. For example, the main control unit 110 checks the decoded calibration status parameters. If the decoded calibration status is "1," comprehensive calculation is performed in step S108. If the decoded calibration status is "0," system calibration is performed in step S107.

[0060] Step S107 is the system calibration step, which involves matching the pressure data for multiple lanes with the test pressure values ​​and correcting the piezoelectric film characteristic curve in combination with ambient temperature data. For example, the obtained temperature and multiple lane pressures are compared with the standard pressure value of the test vehicle actively input by the main control unit to correct the piezoelectric film characteristic curve.

[0061] Step S108 is a comprehensive calculation step, which includes integrating the pressure data of multiple lanes to obtain a total pressure value, calculating the weight of the vehicle in each lane based on the modified piezoelectric film characteristic curve, and calculating the vehicle's driving speed based on the time difference between the peak values ​​of adjacent piezoelectric films. For example, the main control unit 110 performs a time domain integration operation on the decoded pressure signals of each lane, and the calculation formula is:

[0062]

[0063] Where W is the weight of the vehicle, k is the calibration coefficient of the piezoelectric film characteristic curve, P(t) is the pressure signal amplitude, and t1 and t2 are the start and end times of the vehicle axle pressure signal, respectively.

[0064] At the same time, the real-time vehicle speed is calculated based on the time difference Δt between the pressure peak triggering of two adjacent piezoelectric films on the same axle and the preset installation spacing L of the piezoelectric films:

[0065] v=L / Δt

[0066] Step S109 is an image processing step, which involves capturing a snapshot of a vehicle in any lane and identifying the vehicle's license plate, vehicle model, and corresponding standard load threshold using a preset neural network model. For example, upon receiving a data packet, the main control sends a command to the digital image sensor 105 to capture a photo of the lane.

[0067] Step S110 is the violation determination step, which involves comparing the weight of vehicles in any lane with the standard load threshold corresponding to the vehicle type, and comparing the driving speed with the speed limit threshold for the road section, to generate an overload warning signal and / or an overspeed warning signal. For example, the vehicle weight in each lane calculated in step S108 is compared with the load requirement corresponding to the vehicle type determined by image recognition in step S110, and the speed calculated in step S108 is compared with the speed limit for the road section to determine whether the vehicle is speeding or overweight.

[0068] Step S111 is the data packaging step, which involves packaging the weight data, speed data, captured images, and vehicle model information of the overloaded and / or speeding vehicle into a structured package to generate the data to be reported. For example, if step S110 determines that the vehicle is overloaded and / or speeding, the main control unit packages the vehicle's weight data, speed data, captured images, and vehicle model information into a structured package in a pre-set field order and appends a CRC-16 checksum. After verifying the integrity of the data packet through an internal checksum mechanism, the data to be reported is generated.

[0069] Table 1

[0070] Field Order Field Name Data Type describe A Frame header character Fixed to # B Lane 1 data Value Group Coil status (0 / 1), channel 1, channel 2 values C Lane 2 data Value Group Coil status (0 / 1), channel 1, channel 2 values D Lane 3 data Value Group Coil status (0 / 1), channel 1, channel 2 values E Lane 4 data Value Group Coil status (0 / 1), channel 1, channel 2 values F Lane 5 data Value Group Coil status (0 / 1), channel 1, channel 2 values G Lane 6 data Value Group Coil status (0 / 1), channel 1, channel 2 values H Lane 7 data Value Group Coil status (0 / 1), channel 1, channel 2 values I Lane 8 data Value Group Coil status (0 / 1), channel 1, channel 2 values J temperature floating point numbers Current ambient temperature (unit: °C) K Calibration status Boolean values 0 (uncalibrated) or 1 (calibrated) L CRC-16 checksum String Verify the integrity of the data packet M Frame tail character Fixed to*

[0071] Step S112 is the cloud upload step, which involves uploading the data to be reported to the cloud database. For example, the system stores the illegal vehicle data packet obtained in step S111 in the database and transmits the relevant information to the monitoring center via the network for further analysis and recording.

[0072] It's important to note that this system utilizes a modular design. The vehicle-passing detector 101 and piezoelectric film sensor 102 are scalable to support different lane specifications, and the digital image sensor 105 supports interchangeable lenses with different focal lengths to accommodate diverse installation scenarios. The dynamic weighing algorithm is compatible with the heavy-load detection requirements of various travel mechanisms, such as wheeled vehicles and tracked engineering machinery. The data interface supports protocol-level integration with existing overload control systems and traffic management platforms.

[0073] In a specific embodiment, referring to Table 1, an embodiment of a data packet structure includes multiple fields, each of which is arranged in sequence to form a complete data packet for transmitting and processing multi-lane pressure data and calibration status information. The following is a detailed description with reference to the attached table:

[0074] Field A is the frame header, which is fixed to the character "#" and is used to identify the starting position of the data frame;

[0075] Fields B to I are lane data from lanes 1 to 8, respectively. Each field is a value group containing the coil status (0 for untriggered, 1 for triggered), channel 1 measurement value, and channel 2 measurement value.

[0076] Field J is the temperature field, which uses floating-point numbers to represent the current ambient temperature in degrees Celsius (°C);

[0077] Field K is the calibration status field, which uses a Boolean value to indicate the calibration status of the device, where 0 indicates uncalibrated and 1 indicates calibrated;

[0078] Field L is a CRC-16 checksum field, such as a character string or data in a specific format, used to verify the integrity of the data packet;

[0079] Field M is the frame tail, which is fixed to the character "*" and is used to mark the end position of the data frame.

[0080] This data frame structure, through specific field order and data type definitions, enables structured encapsulation of multi-channel pressure data, ambient temperature, and device status information, effectively supporting data parsing and integrated processing within the pressure sensing system. Data from each lane uses a unified value group format, facilitating batch processing and extended maintenance. Independent calibration status fields enhance system status monitoring capabilities. The floating-point representation of the temperature field ensures high-precision recording of environmental parameters, and the fixed character design at the frame header and footer enhances data frame recognition reliability.

[0081] In the description of the present invention, it should be understood that the described embodiments are only part of the embodiments of the present invention, not all the embodiments.

[0082] The above content is a further detailed description of the present application in conjunction with specific implementation methods, and the specific implementation of the present application cannot be considered to be limited to these descriptions. For ordinary technicians in the technical field to which the present application belongs, several simple deductions or substitutions can be made without departing from the inventive concept of the present application.

Claims

1. A vehicle intelligent weighing monitoring system with end-cloud collaboration, characterized in that: include: A plurality of vehicle passing detectors are respectively arranged in a plurality of lanes, each of the vehicle passing detectors is used to sense a vehicle entering a weighing area of ​​the lane; a plurality of piezoelectric film sensors, respectively integrated in the areas where the plurality of vehicle passing detectors are located or in the front areas thereof, each of the piezoelectric film sensors being configured to generate a piezoelectric analog signal corresponding to the load of the vehicle; A plurality of signal conditioning modules are respectively connected to the plurality of piezoelectric film sensors for signal processing, and each of the signal conditioning modules is used to amplify and filter the input piezoelectric analog signal to obtain a corresponding conditioning signal; A multi-channel AD conversion module is connected to the plurality of signal conditioning modules in a one-to-one correspondence, and is used to perform high-precision analog-to-digital conversion on each of the input conditioning signals to obtain corresponding digital signals; A temperature sensor is used to collect ambient temperature and generate corresponding ambient temperature data; an FPGA processing unit, connected to the multi-channel AD conversion module and the temperature sensor signal, for acquiring data of multiple lanes and ambient temperature data through a built-in multi-channel polling mechanism, and verifying and packaging the acquired data to obtain a data packet; A main control unit, signal-connected to the FPGA processing unit, for performing dynamic load analysis on the data packets, and performing environmental compensation, overload determination, and data visualization through intelligent decision-making; Multiple digital image sensors are respectively arranged in multiple lanes and are all connected to the main control unit by signal. Each digital image sensor is used to capture vehicles in its lane. The captured images are used for feature recognition and violation evidence collection of the vehicles. The captured images are then transmitted back to the main control unit for violation analysis and processing. A PHY communication chip, connected to the main control unit signal, used to establish a data link between the main control unit and the cloud vehicle management platform, and transmit the violation analysis and processing results to the cloud vehicle management platform; The cloud-based vehicle management platform is used to record information about vehicles that violate traffic rules.

2. The vehicle intelligent weighing monitoring system according to claim 1, characterized in that: Each of the signal conditioning modules includes a Wheatstone bridge, an RC low-pass filter, an instrumentation amplifier, and a zeroing circuit; The Wheatstone bridge is used to obtain an input piezoelectric analog signal, and perform preliminary adjustment and balancing on the piezoelectric analog signal to improve measurement accuracy, thereby obtaining a first adjustment signal; The RC low-pass filter is connected to the Wheatstone bridge, and is used to filter out high-frequency noise in the first adjustment signal to ensure signal stability and accuracy, thereby obtaining a second adjustment signal; The instrumentation amplifier is connected to the RC low-pass filter and is configured to amplify the second adjustment signal through high input impedance and high common-mode rejection ratio to obtain a third adjustment signal; The zero adjustment circuit is connected to the instrumentation amplifier and is used to perform zero point correction on the third adjustment signal to ensure the accuracy and linearity of the signal and obtain a corresponding conditioned signal.

3. The vehicle intelligent weighing monitoring system according to claim 1, characterized in that: The FPGA processing unit is configured to: perform preliminary data acquisition on the digital signals corresponding to the multiple lanes output by the multi-channel AD conversion module, and package the pressure data, ambient temperature data and the calibration state parameters of the vehicle to obtain the data packet.

4. The vehicle intelligent weighing monitoring system according to claim 3, characterized in that: The data encapsulation format of the FPGA processing unit includes a start character, a plurality of lane data fields, an ambient temperature floating point value, a CRC-16 check code and an end character.

5. The dynamic vehicle weighing system according to claim 3, wherein: The main control unit is configured as follows: Decomposing the data packet, performing integral calculation of pressure data of multiple lanes, dynamically correcting a preset piezoelectric film characteristic curve based on ambient temperature data, and calculating a vehicle's travel speed based on a time difference in signals from adjacent piezoelectric films; Furthermore, the control triggers each of the digital image sensors to perform snapshots, and generates an overweight warning signal according to the pressure value and the standard load threshold of the vehicle model, and generates an overspeed warning signal according to the driving speed and the road speed limit threshold.

6. An intelligent processing method for dynamic weighing, applied to the vehicle intelligent weighing monitoring system according to any one of claims 1 to 5, characterized in that: The intelligent processing method includes: A data packet receiving step includes receiving a data packet, wherein the data packet includes pressure data of a plurality of lanes, ambient temperature data, and calibration state parameters; A protocol decoding step includes parsing the data packet to extract pressure data, calibration state parameters, and ambient temperature data for multiple lanes; a calibration mode determination step, comprising executing a system calibration flow step when the calibration state parameter is an activated value, and executing a comprehensive calculation step otherwise; a system calibration step, comprising matching the pressure data of the plurality of lanes with the test pressure value and correcting the characteristic curve of the piezoelectric film in combination with the ambient temperature data; a comprehensive calculation step, comprising integrating pressure data of multiple lanes to obtain a total pressure value, calculating the weight of vehicles in each lane based on a modified piezoelectric film characteristic curve, and calculating the vehicle's travel speed based on a time difference between peak values ​​of adjacent piezoelectric films; The image processing step includes obtaining a snapshot image of a vehicle in any lane and identifying the vehicle's license plate, model, and corresponding standard load threshold through a preset neural network model; The violation determination step includes comparing the weight of the vehicle in any lane with the standard load threshold corresponding to the vehicle type, and comparing the driving speed with the speed limit threshold of the road section, and generating an overload warning signal and / or a speeding warning signal; The data packaging step includes structurally packaging the weight data, speed data, captured images, and vehicle model information of overloaded and / or speeding vehicles to generate data to be reported; The cloud reporting step includes uploading the data to be reported to a cloud database.

7. The intelligent processing method according to claim 6, characterized in that: The data encapsulation step specifically includes: packaging the weight data, speed data, captured image and vehicle model information of the illegal vehicle in a preset field order, and adding a CRC-16 check code; generating the data to be reported after verifying the data integrity through a check mechanism.

8. The intelligent processing method according to claim 6, characterized in that: The system calibration step specifically includes: in the system calibration state, inputting the pressure signal of a test vehicle of known weight into the system; collecting the output value of the piezoelectric film at the current ambient temperature, fitting the temperature-pressure characteristic curve; and storing the fitting parameters to compensate for measurement errors in real time.

9. The intelligent processing method according to claim 6, wherein: The image processing step also includes: after identifying the vehicle's license plate and model through a preset neural network model, associating the standard load threshold corresponding to the model; if the model cannot be identified, matching the load threshold of an approximate model from historical data and using it as the corresponding standard load threshold.

10. The intelligent processing method according to claim 6, characterized in that: The comprehensive calculation step specifically includes: The calculation sub-step of weight integration includes time domain integration of pressure data of each lane, and the calculation formula is expressed as Where W is the weight of the vehicle, k is the calibration coefficient of the piezoelectric film characteristic curve, P(t) is the pressure signal amplitude, t1 and t2 are the start and end times of the vehicle axle pressure signal respectively; The sub-step of calculating the driving speed includes calculating the vehicle speed based on the peak time difference of adjacent piezoelectric films, which can be expressed as v=L / Δt Where v is the driving speed, L is the installation distance between adjacent piezoelectric films, and Δt is the peak time difference of the same axle pressure signal on the two piezoelectric films.

Images