A task flow driving and closed-loop control method for bridge cable detection

Abstract

The present application belongs to the technical field of bridge structure health monitoring, and relates to a task flow driving and closed-loop control method for bridge cable detection, comprising the following steps: generating a structured data quality contract; integrating a main detection sensor, an independent micro-environment perception unit and an active micro-posture compensation mechanism in a bearing platform; collecting and outputting real-time physical state data streams representing the micro-motion posture of the main detection sensor and the actual distance from the bridge cable by the independent micro-environment perception unit at a high frequency; generating posture compensation control instructions; completing physical constraints on the data acquisition process; encapsulating bridge cable damage data, physical state data and performance status flags into atomized quality stamp data packets; marking the bridge cable area corresponding to the default data to trigger a re-sampling instruction, and forming a certified bridge cable health data set. The present application solves the problem that data reports cannot prove their innocence and cannot trace whether the problem is caused by the bridge cable itself or the failure of the acquisition process.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of bridge structure health monitoring, and relates to a task flow driving and closed-loop control method for bridge cable detection. BACKGROUND

[0002] Bridge cables, as key load-bearing components, are directly related to the safety of the entire bridge structure. However, when detecting bridge cables automatically or robotically, the quality of data acquisition is severely challenged by many uncontrollable environmental factors. For example, high-altitude wind load, vibration of the bearing platform itself, and swing of the bridge cable will all cause the distance and posture between the detection sensor and the surface of the bridge cable to deviate instantaneously. This deviation will seriously affect the accuracy and effectiveness of the detection data. The current core problem is how to ensure that every piece of detection data collected in a dynamic and complex field environment is collected under strict physical conditions, and how to provide credible proof for the quality of these data.

[0003] The commonly used solution in the industry mainly relies on macro-motion control of the bearing platform and post-processing of the data. The detection robot or device usually moves along the bridge cable according to the preset trajectory, and its control system mainly focuses on completing the overall path planning and travel task, lacking effective real-time compensation ability for high-speed and small disturbances. After data acquisition is completed, engineers often need to filter, denoise and screen massive data through complex algorithms, or manually identify suspicious data based on experience.

[0004] Based on the above problems, the data report formed cannot prove its innocence, and when the detection result does not meet the expectation, it cannot be traced back to whether it is a problem of the bridge cable itself or a mistake in the collection process, which weakens the credibility of the detection conclusion, and even expensive secondary on-site re-inspection may be required. SUMMARY

[0005] To solve the above problems, the application provides a task flow driving and closed-loop control method for bridge cable detection.

[0006] A task flow driving and closed-loop control method for bridge cable detection, comprising the following steps:

[0007] S1, generating a structured data quality contract based on the received original task requirements and the preset quantifiable physical indicators;

[0008] S2, integrating the main detection sensor, the independent micro-environment perception unit and the active micro-posture compensation mechanism into the bearing platform according to the structured data quality contract;

[0009] S3, drive the detector actuator, and the independent micro-environment perception unit high-frequency collection and output real-time physical state data stream representing the main detection sensor micro-movement posture and the actual distance from the bridge cable;

[0010] S4, the real-time physical state data stream is compared with the quantifiable physical index in the structured data quality contract, the reverse motion control parameter for offsetting the deviation is calculated, and the posture compensation control instruction is generated;

[0011] S5, the posture compensation control instruction is sent to the active micro-posture compensation mechanism, the high-speed physical displacement and posture adjustment of the main detection sensor are driven, and the physical constraint on the data acquisition process is completed;

[0012] S6, the bridge cable damage data collected by the main detection sensor generates a performance state flag according to whether the physical state data meets the structured data quality contract, and the bridge cable damage data, the physical state data and the performance state flag are packaged into an atomized quality stamp data package;

[0013] S7, a series of atomized quality stamp data packages are gathered, the bridge cable damage data is filtered and integrated according to the performance state flag, the bridge cable region corresponding to the data in breach is marked to trigger a supplementary sampling instruction, and a certified bridge cable health data set is formed.

[0014] Further schemes of the present application, generating a structured data quality contract, include the following steps:

[0015] Receiving original task requirements containing the bridge cable number and specific parts to be detected;

[0016] Defining quantifiable physical indexes for judging data validity based on the original task requirements, the quantifiable physical indexes clearly indicate the distance, posture angle and maximum motion jitter amplitude that the detection sensor and the bridge cable surface need to maintain;

[0017] The bridge cable part information to be detected is combined with the quantifiable physical indexes to form a structured data quality contract.

[0018] Further schemes of the present application, the main detection sensor, the independent micro-environment perception unit and the active micro-posture compensation mechanism are integrated on the bearing platform, including the following steps:

[0019] Selecting the main detection sensor for collecting bridge cable damage information according to the structured data quality contract;

[0020] The independent micro-environment perception unit composed of a micro inertial measurement device and a high-frequency laser range finder is integrated on the main detection sensor;

[0021] The main detection sensor integrated with the independent micro-environment perception unit is loaded to the bearing platform through the active micro-posture compensation mechanism.

[0022] The further scheme of the present application outputs a real-time physical state data stream representing the posture of the main detection sensor and the actual distance from the bridge cable, and comprises the following steps:

[0023] The driving detection actuator is driven to approach the bridge cable detection position specified in the structured data quality contract;

[0024] The independent micro-environment perception unit carried by the main detection sensor is activated at the same time as the main detection sensor collects bridge cable damage information;

[0025] The independent micro-environment perception unit continuously outputs a real-time physical state data stream containing time stamps, posture values and distance values in a high-frequency sampling manner.

[0026] The further scheme of the present application generates posture compensation control instructions, comprising the following steps:

[0027] The real-time posture and distance values in the real-time physical state data stream are continuously compared with the allowed range of posture and distance set in the structured data quality contract;

[0028] When the comparison result shows that the actual physical state exceeds the allowed range, the reverse motion control parameter is calculated based on the difference between the current actual physical state value and the target physical state value defined in the structured data quality contract, and combined with the preset control gain coefficient.

[0029] The reverse motion control parameter is encapsulated into a posture compensation control instruction.

[0030] The further scheme of the present application completes the physical constraint of the data acquisition process, comprising the following steps:

[0031] The active micro-posture compensation mechanism receives and analyzes the posture compensation control instruction, and extracts the reverse motion control parameter;

[0032] The driving signal is calculated according to the reverse motion control parameter, and the main detection sensor is driven to perform high-speed physical displacement and posture adjustment in the opposite direction of the deviation;

[0033] Through continuous active adjustment, the working state of the main detection sensor is dynamically constrained within the physical index boundary specified in the structured data quality contract.

[0034] The further scheme of the present application encapsulates the bridge cable damage data, physical state data and performance status flag into an atomized quality stamp data package, comprising the following steps:

[0035] Each frame of bridge cable damage data output by the main detection sensor is time-synchronized and matched with the physical state data measured by the independent micro-environment perception unit at the same collection instant;

[0036] The matched physical state data and the structured data quality contract are subjected to compliance checking, and a performance state flag is generated for the frame bridge cable damage data;

[0037] The frame bridge cable damage data, the synchronized physical state data and the performance state flag are packaged into an indivisible atomized quality stamp data package.

[0038] In a further aspect of the present application, the performance state flag is generated by the following steps:

[0039] When all the physical state indicators are within the allowable range specified by the structured data quality contract, a performance state flag of "full performance" is generated;

[0040] When at least one physical state indicator approaches the boundary of the allowable range, a performance state flag of "critical" is generated;

[0041] When at least one physical state indicator exceeds the allowable range, a performance state flag of "default" is generated.

[0042] In a further aspect of the present application, the certified bridge cable health data set is formed by the following steps:

[0043] All the converged atomized quality stamp data packages are traversed, and the performance state flag contained in each data package is read;

[0044] Only the bridge cable damage data in the data package with the performance state flag of "full performance" is extracted and integrated;

[0045] The bridge cable region corresponding to the data package with the performance state flag of "default" is recorded, and a re-sampling instruction is generated based on the record and sent to the task control system.

[0046] In a further aspect of the present application, the certified bridge cable health data set is formed by the following steps:

[0047] The re-sampling instruction is received and executed;

[0048] The detection executor is driven to re-collect data from the bridge cable region;

[0049] Until the performance state flag of the atomized quality stamp data package generated in the region is "full performance".

[0050] In summary, the present application has the following beneficial technical effects:

[0051] 1. By introducing an active physical constraint mechanism at the source of data sampling, the intrinsic quality of raw data is fundamentally improved. Instead of passively accepting data pollution caused by environmental disturbances, a dynamic closed-loop control system is formed through high-frequency perception and micro-gesture compensation to offset the effects of various disturbances on the main detection sensor in real time. This approach forces the working state of the sensor to be maintained within the boundaries of the pre-set "data quality contract", ensuring that most data meets the optimal physical conditions at the moment of collection, improving the effectiveness and reliability of the data.

[0052] 2. Each frame of detection data is assigned an unforgeable quality certification information, realizing data quality self-consistency and process traceability. Each frame of damaged data is encapsulated with its collection moment physical state data and quality contract-based compliance status flag, creating a data unit containing complete "birth certificate". This allows the data user to understand the collection background of each piece of data, providing strong support for data credibility, establishing a complete data certification system, and solving the technical problem of untraceable data source and quality in traditional detection data.

[0053] 3. Realize the automation of data screening and resampling decision, improve the overall efficiency of detection work. Based on the compliance status flag in the atomized quality stamp data packet, the system can quickly and accurately screen out data that meets the quality requirements, and automatically identify data gaps caused by substandard collection conditions. The automatically generated resampling instructions ensure the integrity of the detection range, avoiding the tedious and subjective manual data review process, forming an unmanned management closed loop from collection, certification to resampling, significantly shortening the data processing cycle. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiment or prior art description will be briefly introduced below, and the drawings are used to provide further understanding of the present application. For those skilled in the art, other drawings can be obtained without creative labor based on these drawings.

[0055] Fig. 1 The flowchart of the embodiment of the present application is disclosed.

[0056] Fig. 2 The structural schematic diagram of the embodiment of the present application is disclosed. DETAILED DESCRIPTION

[0057] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. All other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application.

[0058] The above and other aspects of the present application will become more apparent by describing in detail the embodiments thereof with reference to the attached drawings in which: Figs. 1-2 The preferred detailed description of the present application is made.

[0059] The above and other aspects of the present application will become more apparent by describing in detail the embodiments thereof with reference to the attached drawings in which: Fig. 1 The present application proposes a task flow driving and closed-loop control method for bridge cable detection, comprising the following steps:

[0060] S1, generating a structured data quality contract based on the received original task demand and the preset quantifiable physical indicators;

[0061] S2, integrating the main detection sensor, the independent micro-environment perception unit and the active micro-posture compensation mechanism in the bearing platform according to the structured data quality contract;

[0062] S3, driving the detection executor to high-frequency collect and output the real-time physical state data stream representing the micro-motion posture of the main detection sensor and the actual distance from the bridge cable by the independent micro-environment perception unit;

[0063] S4, comparing the real-time physical state data stream with the quantifiable physical indicators in the structured data quality contract, calculating the reverse motion control parameters for offsetting the deviation, and generating the posture compensation control instruction;

[0064] S5, sending the posture compensation control instruction to the active micro-posture compensation mechanism to drive the high-speed physical displacement and posture adjustment of the main detection sensor, and completing the physical constraint on the data acquisition process;

[0065] S6, generating the performance state flag according to whether the bridge cable damage data collected by the main detection sensor meets the structured data quality contract, and encapsulating the bridge cable damage data, the physical state data and the performance state flag into the atomized quality stamp data package;

[0066] S7, converging a series of atomized quality stamp data packages, screening and integrating the bridge cable damage data according to the performance state flag, triggering the resampling instruction for the bridge cable region corresponding to the data marked as default, and forming the certified bridge cable health data set.

[0067] In one of the embodiments of the present application, step S1 comprises the following steps:

[0068] The original task requirement of the bridge cable detection is received, and the original task requirement includes a bridge cable number to be detected and a specific part; a quantifiable physical index for judging whether subsequent collected data is valid is defined based on the original task requirement, the quantifiable physical index clearly indicates a distance, an attitude angle and a maximum motion jitter amplitude that the detection sensor needs to maintain with a bridge cable surface when working; the bridge cable part information to be detected and the set of quantifiable physical indexes are combined to form a structured data quality contract.

[0069] Specifically, an operator inputs the original task requirement of the bridge cable detection through a dedicated task management software system, and the original task requirement is the starting point of the whole process and specifies the specific object and range of the detection work. The operator is guided to enter a parameter setting stage based on the original task requirement, and a set of quantifiable physical indexes are defined for the type of the detection sensor to be used and the detection accuracy requirement in this stage, which is the key to ensuring data validity. The operator will set physical boundary conditions to be followed by the detection sensor when working according to the diameter of the bridge cable, the surface condition and the best working range of the sensor. After completing the parameter setting, the system automatically integrates the input bridge cable part information to be detected with the set of quantifiable physical indexes just defined. The system encapsulates these information according to a preset data structure template to generate a structured data quality contract file containing all task information and quality constraints.

[0070] The original task requirement of the bridge cable detection is an initial instruction provided by a task issuer, and the data structure is a text or form record containing at least two fields, which are a bridge cable number to be detected and a specific part. The bridge cable number to be detected is a string for uniquely identifying a certain cable on a bridge, for example, “N3-S05” represents the 5th sling of the north main cable.

[0071] The specific part is used to describe the start and end positions of the detection on the bridge cable, for example, “from 10 meters to 20 meters away from the lower anchoring end”. The quantifiable physical index is a set of threshold values preset for measuring whether the data collection action is standardized, and determines whether the collected data is acceptable. The core data characteristics of the index include a distance, that is, a vertical distance between a detection sensor probe and a bridge cable outer surface that should be maintained, which is set according to the sensor specification and experience. An attitude angle, that is, a deviation angle of the attitude of the sensor relative to the normal direction of the bridge cable surface, which is usually decomposed into a pitch angle and a roll angle. A maximum motion jitter amplitude refers to an instantaneous change amount of the position or angle of the sensor within a unit time, which is used to limit the data blur caused by the shaking of the bearing platform or wind vibration.

[0072] The structured data quality contract is a machine-readable file generated by the system, usually in XML or JSON format, containing a data set of key-value pairs, which functions as an automated standard for subsequent hardware execution and data verification; the structured data quality contract contains task information and explicit numerical values of all quantifiable physical indicators.

[0073] For example, for the corrosion detection of the main cable of a certain bridge, the operator inputs the original task requirements of the bridge cable detection into the task system, the bridge cable number to be detected is "C-08", and the specific part is "5m range on both sides of the lowest point in the middle". Based on the task requirements, the operator defines the quantifiable physical indicators, sets the distance to be 15mm and the allowable deviation range to be ±3mm according to the characteristics of the eddy current sensor; sets the attitude angle, requires the pitch angle and roll angle to be maintained at 0°, and the allowable deviation is ±2°; sets the maximum motion jitter amplitude to be not more than 5mm / s. The system integrates these information to generate a structured data quality contract file.

[0074] In one embodiment of the present application, step S2 includes the following steps:

[0075] The input data quality contract is used to select and install the main detection sensor according to the detection task specified therein, which is used to collect the surface or internal damage information of the bridge cable; the main detection sensor is integrated with an independent micro-environment perception unit, which is composed of a micro-inertial measurement device and a high-frequency laser range finder, for independently measuring the real-time attitude of the main detection sensor and the instantaneous distance from the bridge cable surface; the main detection sensor integrated with the micro-environment perception unit is loaded onto the bearing platform through the active micro-attitude compensation mechanism to form a detection executor.

[0076] Specifically, the generated structured data quality contract file is input, and the description of the detection task type in it, such as corrosion detection or crack detection, is identified, and the main detection sensor matched therewith is recommended or automatically selected for the operator. After the operator obtains the specified main detection sensor, the next integration operation is performed. The operator needs to take an independent micro-environment perception unit and fix it on the shell of the main detection sensor through a standardized mechanical interface such as a screw or a buckle, ensuring that there is no relative displacement between the two. After completing the physical fixation, the data and power cables of the micro-environment perception unit are connected to the reserved interface or central controller of the main detection sensor, and the main detection sensor integrated with the micro-environment perception unit is installed on the movable end face of the active micro-attitude compensation mechanism. The installation process also needs to ensure the rigidity of the connection.

[0077] The combination of compensation mechanism and sensor is loaded to the carrying platform responsible for macro movement on the bridge cable, such as a climbing robot. All hardware components are assembled together to form a complete detection executor. After the assembly is completed, the detection executor is powered on and software initialized, and the system will identify and confirm that the main detection sensor, the micro-environment perception unit and the active micro-posture compensation mechanism are all connected normally and in standby state.

[0078] The main detection sensor is the core measuring device for directly obtaining the bridge cable health information. Its function is to collect the apparent or internal damage information of the bridge cable. According to different detection tasks, it can be an eddy current sensor for detecting surface cracks, or a magnetic flux leakage sensor for detecting internal wire breakage. The function of the independent micro-environment perception unit is to independently measure the real-time posture and instantaneous distance of the main detection sensor from the bridge cable surface, only measuring the motion state of the sensor. The micro-environment perception unit is composed of two core components, one is a micro-inertial measurement device, which integrates gyroscopes and accelerometers inside the electronic components, and can sense and output three-axis angle changes and three-axis acceleration data of the device itself; the other is a high-frequency laser range finder, which uses the laser time-of-flight principle to measure distance. The "high frequency" characteristic refers to its ability to measure at a frequency of hundreds or even thousands of times per second.

[0079] The active micro-posture compensation mechanism is a precise motion platform that can respond quickly, usually driven by piezoelectric ceramics or micro-motors. Its function is to drive the object loaded on it to adjust the position and posture in a small range and at a high speed according to external instructions. The carrying platform is the basic carrier responsible for carrying the entire detection device and moving along the bridge cable, such as a tracked robot or a wheeled crawler. The detection executor is the general term for the integrated device formed in this step, which includes the main detection sensor, the micro-environment perception unit, the active micro-posture compensation mechanism, and the carrying platform.

[0080] For example, first input and parse the structured data quality contract for "C-08" main cable. Because the task is corrosion detection, the operator selects an eddy current sensor model EC-200 as the main detection sensor. The operator fastens the independent micro-environment perception unit model MPU-LD-V2 to the top shell of the EC-200 sensor through four M3 screws. The unit integrates a six-axis micro-inertial measurement device and a high-frequency laser range finder with a sampling frequency of 1000Hz. The main detection sensor integrated with the micro-environment perception unit is installed on the piezoelectric ceramic active micro-posture compensation mechanism model PZT-5D.

[0081] The combination loading model is a magnetic adsorption type cable climbing robot, which is a carrying platform. After completing all hardware connections and line checks, the system is started, and all the above devices are successfully identified. The assembled whole constitutes the detection executor of this task.

[0082] In one embodiment of the present application, step S3 includes the following steps:

[0083] The driving detection executor is driven to approach the bridge cable detection position specified by the data quality contract; the main detection sensor collects bridge cable damage information, and at the same time, the micro-environment perception unit carried thereby is activated; the micro-environment perception unit continuously outputs real-time physical state data stream representing the current micro-motion posture of the main detection sensor and the actual distance from the bridge cable in a high-frequency sampling manner.

[0084] Specifically, a moving instruction is issued to the detection executor, and after reaching the specified position, the data acquisition function is started synchronously, and finally continuous data stream output is obtained. The implementation process is as follows: the control system sends an instruction to the carrying platform in the detection executor, drives the detection executor to move along the bridge cable until the front end of the main detection sensor approaches the bridge cable detection position specified by the data quality contract. After reaching the starting point of the predetermined detection area, the system issues a synchronous activation instruction, which is sent to the main detection sensor and the micro-environment perception unit carried thereby at the same time. After receiving the instruction, the main detection sensor begins to perform its own work, such as emitting electromagnetic waves to collect bridge cable damage information. At the same time, the micro-environment perception unit is also activated, and the micro-inertial measurement device and high-frequency laser range finder inside it begin to work.

[0085] The micro-environment perception unit continuously measures and packages the measurement results into data frames in real time in a high-frequency sampling manner. Each data frame records the three-dimensional posture angle of the main detection sensor at that instant and the actual distance from the bridge cable surface. These data frames are continuously sent outward through the data interface to form a real-time physical state data stream representing the current micro-motion posture of the main detection sensor and the actual distance from the bridge cable.

[0086] Among them, the real-time physical state data stream is a time series data set, and the data structure is a series of data packets arranged in time sequence, each data packet contains an accurate time stamp, a value representing the posture, such as pitch angle and roll angle, and a value representing the distance; for the subsequent analysis link, it provides a physical state snapshot of the main detection sensor at each very short time instant.

[0087] The "high-frequency sampling" feature of the real-time physical state data stream is based on the hardware performance of the micro-environment perception unit, and the sampling frequency is usually between 500 and 1000 times per second according to the setting, to ensure that the smallest jitter and position changes can be captured.

[0088] For example, for the detection task of "C-08" main cable, the operator issues an instruction to the CableClimber-M robot through the control software to move to the "midspan lowest point" position. After moving to the position, the system issues a collection instruction, and the main detection sensor, EC-200 eddy current sensor, starts working, and the MPU-LD-V2 micro-environment perception unit carried on it is activated. The MPU-LD-V2 unit starts measuring and outputting data at a frequency of 1000HZ to form a real-time physical state data stream. At the time stamp "1653033601.125", the data packet content in the real-time physical state data stream is .

[0089] At this time, the pitch angle of the main detection sensor is 0.8°, the roll angle is -1.5°, and the distance from the bridge cable surface is 17.2mm. This data stream will be continuously generated as the detection executor moves.

[0090] In one embodiment of the present application, step S4 includes the following steps:

[0091] The input real-time physical state data stream is synchronized to retrieve the preset quantifiable physical indicators in the data quality contract; the real-time attitude and distance values in the data stream are continuously compared with the allowed range of attitude and distance set in the contract; once the comparison result shows that the actual physical state of the main detection sensor is about to exceed or has exceeded the allowed range, the reverse motion control parameters for offsetting the deviation are calculated, and the attitude compensation control instruction is generated.

[0092] This step aims to dynamically generate a control signal for correcting the sensor attitude by analyzing the data stream obtained in the previous step in real time and comparing it with the preset standard. The entire process is a continuous running closed-loop calculation process, which receives real-time data and compares it with the contract standard, and calculates the reverse adjustment amount according to the deviation.

[0093] Specifically, the data processing module of the system continuously receives real-time physical state data stream from the micro-environment perception unit, and synchronously retrieves the first step generated structured data quality contract file from the storage when receiving the first frame data, loads the quantifiable physical indicators therein into the memory for quick query. The system enters an uninterrupted loop comparison state, and for each newly arrived data packet in the data stream, the system extracts the real-time posture and distance values therein. The system mathematically compares these real-time values with the allowed range of posture and distance read from the data quality contract, which is jointly determined by the preset base value and the tolerance, for example, if the distance set in the contract is 15 mm and the tolerance is ±3 mm, the allowed range is 12 to 18 mm.

[0094] The comparison operation checks whether the real-time value falls within this interval, and once the comparison result shows that the actual physical state of the main detection sensor is about to exceed or has exceeded the allowed range, such as the real-time distance becoming 11.8 mm or the posture angle exceeding the preset threshold, the system triggers the deviation calculation program. The program calculates the difference between the current actual state and the central target value required by the contract, calculates the reverse motion control parameter for offsetting the deviation based on the difference, and the reverse motion control parameter includes the direction and amplitude that need to be adjusted. The parameter is packaged into a standard format instruction to generate a posture compensation control instruction. The posture compensation control instruction is formed after the reverse motion control parameter is packaged, and is an electronic signal or data packet conforming to the hardware interface protocol.

[0095] The calculation of the posture compensation control instruction can be simplified as the following linear relationship:

[0096]

[0097] wherein, represents the reverse motion control parameter, indicating the motion direction and size that the compensation mechanism needs to perform, for example, a positive value indicates moving forward, and a negative value indicates moving backward. The reverse motion control parameter is a data containing motion adjustment amplitude and direction, and its data structure can be a vector containing multiple components, for example , respectively corresponding to the small displacement or rotation adjustment amount that needs to be performed in three degrees of freedom. represents the target physical state value defined in the data quality contract, for example, the target distance is 15 mm. represents the current actual physical state value parsed from the real-time physical state data stream.

[0098] k is the control gain coefficient, used to adjust the sensitivity and strength of the compensation response, and is usually adjusted by step response test on the system, and the setting principle is to ensure fast response while avoiding overshoot and oscillation of the system, and is usually set between 0.5 and 1.5.

[0099] For example, after receiving the data packet "{timestamp: 1653033601.125, pitch: 0.8, roll: -1.5, distance_mm: 17.2}" of the real-time physical state data stream of the main cable detection of "C-08", the data quality contract is invoked. The contract requires the distance to be 15 ± 3 mm, i.e. [12, 18] mm; the attitude angle to be 0 ± 2°. The current real-time values 17.2 mm and 0.8°, -1.5° are all within the allowed range, so no instructions are generated. At a certain time "1653033601.128" later, a new data packet arrives and the content is "{... distance_mm: 18.3}". The distance value 18.3 mm exceeds the upper limit of 18 mm, and the system triggers the calculation.

[0100] Assuming that the control gain coefficient k is 0.8, the target distance is 15 mm, and the current actual distance is 18.3 mm. According to the formula , the inverse motion control parameter is calculated. The sensor needs to be moved backward by 2.64 mm, and then the displacement parameter "-2.64 mm" is packaged into an attitude compensation control instruction.

[0101] In one embodiment of the present application, step S5 includes the following steps:

[0102] The attitude compensation control instruction is sent to the active micro-attitude compensation mechanism of the detection executor. The active micro-attitude compensation mechanism drives the connected main detection sensor to perform high-speed physical displacement and attitude adjustment according to the instruction. In this way, the working state of the main detection sensor is constrained within the physical index boundary specified by the data quality contract, and the physical constraint on the data acquisition process is completed.

[0103] Specifically, the attitude compensation control instruction is sent to the controller of the active micro-attitude compensation mechanism installed on the detection executor through an internal data bus or a wireless communication method. After receiving the instruction, the microprocessor built in the active micro-attitude compensation mechanism will analyze the instruction content and extract the inverse motion control parameter contained therein. According to the analyzed inverse motion control parameter, the controller will calculate the voltage or current signal that needs to be applied to its driving elements, such as piezoelectric ceramics or micro-motors. The controller outputs these electrical signals to the driving elements, which produce deformation or rotation under the excitation of the electrical signals, driving the connected movable platform, i.e. the main detection sensor loaded thereon, to perform high-speed physical displacement and attitude adjustment. The direction and amplitude of this adjustment are exactly opposite to the previously detected deviation, but the amplitude is adjusted by the gain coefficient.

[0104] The adjustment action, which is one-time or continuous, counteracts the disturbance, such as vibration or attitude drift, transmitted by the bearing platform. This process is continuous as long as there is deviation and control instructions are generated, and the adjustment is executed by the compensation mechanism to dynamically constrain the working state of the main detection sensor within the physical index boundary, such as distance and attitude angle, specified by the data quality contract, and finally complete the physical constraint of the data acquisition process.

[0105] The physical constraint of the data acquisition process is achieved by actively intervening to ensure that the data acquisition hardware meets the predetermined specifications throughout the entire working period. The characteristics are real-time and closed-loop.

[0106] For example, a posture compensation control instruction containing a "-2.64mm" displacement parameter is sent to the PZT-5D active micro-posture compensation mechanism on the detection actuator. After receiving the instruction, the controller of the mechanism analyzes that the main detection sensor needs to be moved backward by 2.64mm in the direction perpendicular to the bridge cable surface. The controller applies a negative voltage calculated accurately to the piezoelectric ceramic driver, and the piezoelectric ceramic shrinks under the voltage, driving the EC-200 main detection sensor on the top to move backward by 2.64mm accurately. The response time of this adjustment action is extremely short, completed in ms level.

[0107] Through this adjustment, the distance between the main detection sensor and the bridge cable surface is corrected from 18.3mm, which exceeds the standard, to 15.66mm, back to the allowed range of [12, 18]mm. As long as there is disturbance causing distance or attitude deviation during the movement of the detection actuator, the active micro-posture compensation mechanism will perform high-speed physical displacement and attitude adjustment according to the newly generated instructions to achieve the goal of real-time constraining the working state of the main detection sensor within the physical index boundary specified by the data quality contract, and complete the physical constraint of the data acquisition process.

[0108] In one embodiment of the present application, step S6 includes the following steps:

[0109] While performing physical constraint, each frame of bridge cable damage data collected by the main detection sensor is time-synchronized with the physical state data measured by the micro-environment perception unit at the moment of collection; according to whether the physical state data completely meets the provisions of the data quality contract, a compliance status flag is generated for the frame of damage data, which is divided into three states: "complete compliance", "critical" or "breach"; the frame of damage data, the physical state data synchronized therewith, and the compliance status flag are packaged and encapsulated into an indivisible atomized quality stamp data package.

[0110] Specifically, while the system performs the physical constraint action, the data processing module processes two data streams in parallel, one is the bridge cable damage data collected by the main detection sensor, and the other is the physical state data measured by the micro-environment perception unit. Through the internal high-precision clock, it is ensured that each frame of bridge cable damage data output by the main detection sensor is time-synchronized and matched with the physical state data measured by the micro-environment perception unit at the same collection moment. This matching process is strictly one-to-one, ensuring that the damage data and the physical environment information at the time of its birth can be accurately corresponded. The system performs final compliance verification on the matched set of physical state data and the quantifiable physical indicators specified in the data quality contract.

[0111] According to the verification result, the system generates a performance status flag for this frame of bridge cable damage data. This flag has three explicit states: the first is "full performance", indicating that when collecting this frame of data, all physical state indicators such as distance and attitude angle are strictly within the allowed range specified in the data quality contract. The second is "critical", indicating that at least one physical state indicator is close to the boundary value, such as reaching more than 95% of the allowed range. The third is "breach of contract", indicating that at least one physical state indicator has exceeded the allowed range, meaning that although there is a physical constraint mechanism in place, the instantaneous state may still not meet the requirements due to excessive or rapid disturbance.

[0112] The system packages this frame of bridge cable damage data, the physical state data synchronized with it, and the performance status flag just generated into an indivisible atomic quality stamp data package.

[0113] Among them, the bridge cable damage data refers to the smallest data unit generated by the main detection sensor in a single sampling period, which can be a two-dimensional image representing the surface condition or a one-dimensional array representing the magnetic field or eddy current intensity at a certain point, depending on the sensor type. The performance status flag is a discrete label used to represent the data quality level, which is an enumeration type variable or a simple string, informing the data user whether the damage data collected complies with the preset quality specification.

[0114] The specific state setting basis of the performance status flag is that "full performance" corresponds to all indexes being in the allowed interval; the trigger threshold of the "critical" state is configured according to the fineness requirement of the task, for example, the deviation of any index value from the target value is more than 95% of the allowed deviation range, this setting aims to warn the area that may have potential problems; "default" is triggered when any index value exceeds the allowed interval. The atomized quality stamp data packet ensures the permanent binding of data and its quality proof information, preventing data from being tampered with or separated from its collection background information. The "non-divisible" characteristic is ensured by data structure design or hash check technology, ensuring that the data packet is stored, transmitted and processed as a whole.

[0115] For example, at a certain moment when the "C-08" main cable is detected, the main detection sensor collects a frame of eddy current detection data, showing that there is a slight anomaly at this point. At the same time, the physical state data measured by the micro-environment perception unit is "{pitch: 0.5, roll: -0.2, distance_mm: 15.1}". The system compares the physical state data with the data quality contract (requiring a distance of 15±3mm and an attitude of 0±2°), and finds that all indicators are well within the allowed range center area, so a performance status flag is generated for this frame of damage data, and the value is "full performance". The frame of eddy current data, this physical state data packet and the "full performance" flag are packaged together to form an atomized quality stamp data packet.

[0116] Due to the violent shaking of the bearing platform, although the compensation mechanism has tried its best to adjust, the measured physical state data is "{... distance_mm: 18.1}", which exceeds the upper limit of 18mm. The system generates a performance status flag with a value of "default" for the damage data collected synchronously with this state, and also packages another atomized quality stamp data packet.

[0117] In one embodiment of the present application, step S7 comprises the following steps:

[0118] A series of atomized quality stamp data packets are received and aggregated; all data packets are traversed, and the data is automatically screened according to the performance status flag contained therein; only the bridge cable damage data in the data packet with the performance status flag of "full performance" is retained and integrated, and a resampling instruction is triggered for the bridge cable area corresponding to the data marked as "default", forming a verified bridge cable health data set that has removed invalid data and is self-attestable in quality.

[0119] Specifically, a series of atomized quality stamp data packets are received and aggregated, which are stored in a temporary data set, waiting for final collation and adjudication. Starting the traversal program will check each atomized quality stamp data packet in this set one by one, and the core task of the traversal program is to read and identify the fulfillment status flag contained in each data packet during the checking process. According to the value of this flag, the system executes the automated data screening logic. For all data packets with a "fully fulfilled" fulfillment status flag, the system will extract the bridge damage data contained therein and retain it, and these retained data will be integrated into a new data set. For data packets with a "critical" fulfillment status flag, the system will mark them as "to be reviewed" for further confirmation of their validity by human or according to the preset strategy as valid data and integrated. For all data packets with a "default" fulfillment status flag, the system will directly discard them and record the bridge location information corresponding to the data packet.

[0120] The system will summarize all bridge areas corresponding to the "default" data marked, and automatically generate one or more re-sampling instructions containing the exact position that needs to be re-collected and send it back to the task control system. Through the screening and integration process, the system will collect all the bridge damage data with guaranteed quality together to form a certified bridge health data set that has excluded invalid data.

[0121] Among them, the re-sampling instruction is an automated instruction sent to the task control system, and the data structure usually contains the bridge number and specific start and end positions that need to be re-detected to ensure the integrity of the detection coverage. For the data gap area caused by unqualified data quality, it can automatically trigger re-detection to form a closed-loop workflow. The certified bridge health data set is a collection that has been sorted and contains only valid bridge damage data.

[0122] The "certified" and "quality self-certified" attributes come from the fact that each piece of data in the data set is derived from an atomized quality stamp data packet with a "fully fulfilled" fulfillment status flag, meaning that each piece of data comes with its own "birth certificate" of compliance in the collection process, and users do not need to doubt the validity of the data itself for bridge health status assessment and analysis.

[0123] For example, assuming that 10,000 atomized quality stamp data packets generated in the main cable detection task of "C-08" are converged, the system starts to traverse these data packets. During the traversal process, it is found that 9,850 data packets have a performance status flag of "full performance", 100 have a flag of "critical", and 50 have a flag of "default". The system extracts all bridge cable damage data from the 9,850 "full performance" data packets, and for the 100 "critical" data packets, according to the preset strategy, the system also extracts the damage data therein and marks them as "attention required". For the 50 "default" data packets, the system directly discards them as a whole, and records the bridge cable positions corresponding to the data packets, such as "1.5m to 1.7m near the lowest point of the span".

[0124] After all the data packets are processed, the system integrates the position information corresponding to the 50 default data into a supplementary collection instruction, the instruction content is "task: C-08-supplementary collection, position: 1.5m to 1.7m near the lowest point of the span", and sends it to the task management system. The 9,850 "full performance" data and the 100 "critical" data extracted are arranged and combined according to the position sequence on the bridge cable, and finally form a set of certified bridge cable health data.

[0125] Referring to the accompanying drawings Fig. 2 The present application also proposes a task flow driving and closed-loop control system for bridge cable detection, comprising the following modules:

[0126] A contract configuration module generates a structured data quality contract based on the received original task requirements and the preset quantifiable physical indicators;

[0127] An executor assembly module integrates the main detection sensor, the independent micro-environment perception unit and the active micro-posture compensation mechanism on the bearing platform according to the structured data quality contract;

[0128] A state perception module drives the detection executor to collect and output real-time physical state data streams representing the micro-motion posture of the main detection sensor and the actual distance from the bridge cable at a high frequency by the independent micro-environment perception unit;

[0129] An instruction generation module compares the real-time physical state data stream with the quantifiable physical indicators in the structured data quality contract, calculates the reverse motion control parameters for offsetting the deviation, and generates posture compensation control instructions;

[0130] A posture constraint module sends the posture compensation control instructions to the active micro-posture compensation mechanism to drive the high-speed physical displacement and posture adjustment of the main detection sensor, and completes the physical constraint on the data collection process;

[0131] The quality encapsulation module encapsulates the bridge cable damage data collected by the main detection sensor, generates a compliance state flag according to whether the physical state data conforms to the structured data quality contract, and encapsulates the bridge cable damage data, the physical state data and the compliance state flag into an atomized quality stamp data package;

[0132] The data set output module gathers a series of atomized quality stamp data packages, filters and integrates the bridge cable damage data according to the compliance state flag, triggers a supplementary sampling instruction for the bridge cable region corresponding to the data marked as non-compliant, and forms a certified bridge cable health data set.

[0133] The various modules can be realized by software, hardware and a combination thereof in whole or in part, support hardware forms embedded in or independent of the processor in the computer device, and also support software forms stored in the memory in the computer device, so as to facilitate the processor to call and execute the operations of the various modules.

[0134] It should be noted that the human information (including but not limited to human device information and personal information) and data (including but not limited to data for analysis, stored data and displayed data) involved in the present application are information and data authorized by the human body or fully authorized by all parties, and the collection, use and processing of the relevant data require relevant legal standards.

[0135] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for task flow driving and closed-loop control oriented to bridge cable detection, characterized in that, The method comprises the following steps: S1, generating a structured data quality contract based on the received original task requirement and the preset quantifiable physical index; Receiving an original task requirement containing the bridge cable number and specific part to be detected; Defining a quantifiable physical index for judging data validity based on the original task requirement, which specifies the distance, attitude angle and maximum motion jitter amplitude that the detection sensor and the bridge cable surface need to maintain; Combining the bridge cable part information to be detected with the quantifiable physical index to form a structured data quality contract; S2, integrating the main detection sensor, independent micro-environment perception unit and active micro-attitude compensation mechanism on the bearing platform according to the structured data quality contract; S3, driving the detection executor to continuously collect and output real-time physical state data stream representing the micro-motion attitude of the main detection sensor and the actual distance from the bridge cable by the independent micro-environment perception unit; S4, comparing the real-time physical state data stream with the quantifiable physical index in the structured data quality contract to calculate the reverse motion control parameter for offsetting the deviation and generating the attitude compensation control instruction; S5, sending the attitude compensation control instruction to the active micro-attitude compensation mechanism to drive the high-speed physical displacement and attitude adjustment of the main detection sensor, and completing the physical constraint of the data acquisition process; S6, generating a compliance status flag according to whether the physical state data meets the structured data quality contract, and encapsulating the bridge cable damage data, physical state data and compliance status flag into an atomized quality stamp data package; S7, filtering and integrating the bridge cable damage data according to the compliance status flag, marking the bridge cable region corresponding to the data in breach to trigger a re-sampling instruction, and forming a certified bridge cable health data set.

2. The task flow driving and closed-loop control method oriented to bridge cable detection according to claim 1, characterized in that, The main detection sensor, independent micro-environment perception unit and active micro-attitude compensation mechanism are integrated on the bearing platform, comprising the following steps: Selecting a main detection sensor for collecting bridge cable damage information according to the structured data quality contract; Integrating a micro-inertial measurement device and a high-frequency laser range finder on the main detection sensor to form an independent micro-environment perception unit; Loading the main detection sensor integrated with the independent micro-environment perception unit to the bearing platform through the active micro-attitude compensation mechanism.

3. The task flow driving and closed-loop control method oriented to bridge cable detection according to claim 1, characterized in that, Outputting real-time physical state data stream representing the micro-motion attitude of the main detection sensor and the actual distance from the bridge cable, comprising the following steps: Driving the detection executor to approach the bridge detection part specified in the structured data quality contract; Activating the independent micro-environment perception unit carried by the main detection sensor to collect bridge cable damage information; Continuously outputting real-time physical state data stream containing time stamp, attitude value and distance value by the independent micro-environment perception unit in high-frequency sampling mode.

4. The task flow driven and closed loop control method for bridge cable detection according to claim 1, wherein, Generating the attitude compensation control instruction, comprising the following steps: Continuously comparing the attitude value and distance value in the real-time physical state data stream with the allowed range of attitude and distance set in the structured data quality contract; When the comparison result shows that the actual physical state is out of the allowed range, based on the difference between the current actual physical state value and the target physical state value defined in the structured data quality contract, and combined with the preset control gain coefficient, the reverse motion control parameter is calculated; The reverse motion control parameter is encapsulated into a posture compensation control instruction.

5. The task flow driven and closed loop control method for bridge cable detection according to claim 1, wherein, Complete the physical constraints of the data acquisition process, including the following steps: The active micro-posture compensation mechanism receives and analyzes the posture compensation control instruction, and extracts the reverse motion control parameter; According to the reverse motion control parameter, the driving signal is calculated, and the main detection sensor is driven to perform high-speed physical displacement and posture adjustment in the opposite direction of the deviation; Through continuous active adjustment, the working state of the main detection sensor is dynamically constrained within the physical index boundary specified by the structured data quality contract.

6. The task flow driven and closed loop control method for bridge cable detection according to claim 1, wherein, Bridge cable damage data, physical state data and performance status flag are encapsulated into atomized quality stamp data package, including the following steps: Each frame of bridge cable damage data output by the main detection sensor is time-synchronized and matched with the physical state data measured by the independent micro-environment perception unit at the same collection moment; The matched physical state data is checked for compliance with the structured data quality contract, and a performance status flag is generated for the frame of bridge cable damage data; The frame of bridge cable damage data, the synchronized physical state data and the performance status flag are packaged into an indivisible atomized quality stamp data package.

7. The task flow driven and closed loop control method for bridge cable detection according to claim 6, wherein, Generating a performance status flag, including the following steps: When all physical state indicators are within the allowed range specified by the structured data quality contract, generate a "fully performed" performance status flag; When at least one physical state indicator is within the allowed range but exceeds the preset percentage of the allowed range, generate a "critical" performance status flag; When at least one physical state indicator is out of the allowed range, generate a "default" performance status flag.

8. The task flow driven and closed loop control method for bridge cable detection according to claim 1, wherein, Forming a certified bridge cable health data set, including the following steps: Iterate through all the gathered atomized quality stamp data packages and read the performance status flag contained in each data package; Only extract and integrate the bridge cable damage data in the data package with a "fully performed" performance status flag; Record the bridge cable regions corresponding to the data packages with a "default" performance status flag, and generate a re-sampling instruction based on the record and send it to the task control system.

9. The task flow driven and closed loop control method for bridge cable detection according to claim 8, wherein, Forming a certified bridge cable health data set, also including the following steps: Receive and execute the re-sampling instruction; Drive the detection executor to re-sample the data of the bridge cable region; Until the performance status flag of the atomized quality stamp data package generated in this region is "fully performed".

Images