Intelligent safety guarantee platform for crane

Abstract

The invention relates to the technical field of data analysis, and discloses an intelligent safety guarantee platform for a crane. The platform is characterized in that a data acquisition module acquires crane operation parameters through a sensing network and performs feature extraction; the state monitoring module monitors and evaluates the states of steel wire rope abrasion and brake clearance key components; the basic operation and maintenance module formulates a maintenance plan based on the health indexes and historical data; the operation guarantee module carries out safety assessment on hoisting operation and generates a control scheme; the system management module monitors a platform operation process; and the knowledge base module performs association analysis on various data and constructs an intelligent operation and maintenance knowledge base. The method is suitable for various cranes such as portal cranes, bridge cranes and gantry cranes specified in GB / T 20776, and intelligence and precision of the crane maintenance process are achieved.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to an intelligent safety assurance platform for cranes. Background Art

[0002] At present, the safe operation and maintenance of large cranes mainly rely on manual inspections and regular maintenance. Traditional maintenance methods mainly include regular inspections of mechanical parts wear, measurement of electrical system operating parameters, and detection of safety protection device action status. When equipment fails or is abnormal, maintenance personnel need to determine the cause of the failure based on experience and formulate maintenance plans. At the same time, the existing crane maintenance management adopts a preventive maintenance strategy, performs maintenance at fixed time intervals or operating hours, and records maintenance information through paper records or simple spreadsheets. This method ensures the safe operation of the equipment to a certain extent.

[0003] However, this traditional maintenance model has obvious shortcomings. First, the manual inspection method is highly subjective, and the detection accuracy and reliability are difficult to guarantee, which can easily miss potential faults. Secondly, the fixed-cycle preventive maintenance cannot dynamically adjust the maintenance plan according to the actual status of the equipment, resulting in waste of maintenance resources or untimely maintenance. Thirdly, the recording and management of maintenance information is relatively scattered, making it difficult to establish a health record for the entire life cycle of the equipment, and it is impossible to fully utilize historical maintenance experience to guide subsequent work. Finally, the lack of intelligent control over the maintenance process makes it difficult to ensure the consistency and traceability of maintenance quality. Summary of the invention

[0004] This application provides a crane intelligent safety assurance platform, which is used to realize the intelligence and precision of the crane maintenance process. Through the fusion analysis of multi-source data, an equipment health assessment model is established to realize scientific maintenance decision-making. At the same time, a unified knowledge management platform is built to improve maintenance efficiency and quality.

[0005] The present application provides a crane intelligent safety assurance platform, which includes: a data acquisition module, which is used to collect equipment operating parameters through a multi-dimensional sensor network arranged in important crane structures, hoisting mechanisms, luffing mechanisms and slewing mechanisms, and to extract and standardize the equipment operating parameters to obtain equipment feature data sets; a status monitoring module, which is used to analyze and evaluate the wear of crane wire ropes, brake clearance, reducer vibration and motor temperature according to the equipment feature data sets, and generate component health indicators; a basic operation and maintenance module, which is used to analyze and evaluate the component health indicators based on the component health indicators combined with the crane's historical load spectrum and working condition data. The module is used to carry out maintenance planning and resource scheduling according to the equipment feature data set and to form an intelligent maintenance task list; the operation guarantee module is used to carry out safety assessment of the crane's hoisting space, lifting height and working radius according to the equipment feature data set and the maintenance task list, and to generate a dynamic control plan; the system management module is used to monitor the maintenance process of the crane's limit protection device, anti-collision system and overload protection device according to the maintenance task list and the dynamic control plan, and output the quality assessment result; the knowledge base module is used to carry out multi-dimensional correlation analysis of the equipment feature data set, component health indicators, maintenance task lists, dynamic control plans and quality assessment results, and to build a knowledge base for intelligent operation and maintenance of cranes.

[0006] In the technical solution provided by this application, by deploying a multi-dimensional sensor network on important structural parts and mechanisms of the crane, real-time monitoring and data collection of the equipment's operating status are achieved, providing a reliable data basis for health status assessment; key parameters such as wire rope wear, brake clearance, reducer vibration and motor temperature are analyzed and evaluated, achieving accurate assessment of component status and quantitative expression of health indicators; maintenance planning and resource scheduling are carried out in combination with the crane's historical load spectrum and operating condition data, achieving reasonable scheduling of maintenance tasks and optimal allocation of resources; and the safety of maintenance operations is ensured by real-time safety assessment of hoisting space, lifting height and working radius. It effectively prevents the risks of collision and interference during the operation; it monitors and evaluates the maintenance process of the crane's limit protection device, anti-collision system and overload protection device throughout the process to ensure the standardization and reliability of the maintenance operation; by establishing a multi-dimensional correlation analysis model of equipment feature data sets, component health indicators, maintenance task orders, dynamic management and control plans and quality assessment results, it builds a complete crane intelligent operation and maintenance knowledge base, which provides effective knowledge support for subsequent maintenance decisions, thereby realizing the intelligence, precision and standardization of the crane maintenance process, significantly improving maintenance efficiency and reliability, reducing maintenance costs and extending equipment service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0008] Figure 1 This is a schematic diagram of an embodiment of the intelligent safety assurance platform for cranes in the embodiments of the present application;

[0009] Figure 2 This is a schematic diagram of another embodiment of the intelligent safety assurance platform for cranes in the embodiments of the present application. DETAILED DESCRIPTION

[0010] The embodiment of the present application provides a crane intelligent safety assurance platform. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0011] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 , an embodiment of the crane intelligent safety assurance platform in the embodiment of the present application includes:

[0012] The data acquisition module is used to collect equipment operating parameters through a multi-dimensional sensor network arranged on important structural parts, hoisting mechanism, luffing mechanism and slewing mechanism of the crane, and perform feature extraction and standardization on the equipment operating parameters to obtain an equipment feature data set;

[0013] A condition monitoring module is used to analyze and evaluate the wear of the crane wire rope, brake clearance, reducer vibration and motor temperature according to the equipment characteristic data set to generate component health indicators;

[0014] A basic operation and maintenance module is used to perform maintenance planning and resource scheduling based on the component health indicators combined with the crane's historical load spectrum and operating condition data to form an intelligent maintenance task list;

[0015] An operation guarantee module is used to perform safety assessment on the crane hoisting space, lifting height and working radius according to the equipment characteristic data set and the maintenance task list, and generate a dynamic control plan;

[0016] A system management module, used to monitor the maintenance process of the crane limit protection device, anti-collision system and overload protection device according to the maintenance task list and dynamic control plan, and output quality assessment results;

[0017] The knowledge base module is used to perform multi-dimensional correlation analysis on the equipment feature data set, component health indicators, maintenance task orders, dynamic control plans and quality assessment results to build a crane intelligent operation and maintenance knowledge base.

[0018] It is understandable that the execution subject of the present application may be a crane intelligent safety assurance system, or a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0019] Specifically, a multi-dimensional sensor network is deployed at key parts of the crane. Specifically, stress sensors for measuring stress changes are installed at key positions of important structural parts, acceleration sensors for monitoring abnormal vibration of the transmission system are installed at the lifting mechanism, angle sensors for measuring pitch angles are installed at the luffing mechanism, and temperature sensors for measuring temperature are installed at the slewing mechanism. The data acquisition module collects raw data at sampling frequencies of 1Hz for strain sensors, 1000HZ for acceleration sensors, 1Hz for angle sensors, and 1Hz for temperature sensors. Then, zero drift correction is performed on the strain signal, low-pass filtering is performed on the angle signal, and mean filtering is performed on the speed signal. The processed data is synchronized under a unified time base, and statistical features such as peak coefficient, standard deviation, and root mean square value are extracted. The frequency domain features are calculated through a 512-point fast Fourier transform. The state monitoring module evaluates the component state of the standardized equipment feature data set, analyzes the wire breakage rate and surface wear of the wire rope, measures the brake pad gap value and braking torque change, calculates the amplitude and frequency characteristics of the reducer, and monitors the motor temperature rise rate and thermal load level. The time series variation law of component status is obtained, and statistical features are calculated within the sliding time window. The status parameters are compared with the threshold range, abnormal points and abnormal intervals are marked, and the missing data are repaired using the cubic spline interpolation method. A statistical analysis with a 95% confidence level is performed, and the health indicators of each component are obtained through weighted fusion.

[0020] Combine the component health index with the equipment's historical load spectrum to conduct time series correlation analysis, count the state distribution law under different load levels, divide the time period based on the working condition data, calculate the demand degree and difficulty coefficient of each maintenance project, and determine the maintenance priority. Sort the maintenance projects according to the priority, estimate the maintenance time and working space requirements, count the number of required maintenance tools and maintenance personnel configuration, match the tool inventory status and personnel skill level, and generate a maintenance task list with a detailed time schedule. During the maintenance process, combine the equipment feature data set with the maintenance task requirements, extract the obstacle contour points of the hoisting space, the limit value of the lifting height, and the boundary value of the working radius, and establish a safety boundary through three-dimensional coordinate mapping. Calculate the minimum envelope to analyze the accessibility, height margin, and radius redundancy of the working area, divide the danger level and calculate the collision risk coefficient, generate a risk area map, and plan the optimal maintenance path and obstacle avoidance points based on this, forming a dynamic safety management and control plan.

[0021] The quality of the maintenance execution process is monitored, and the displacement sensor signals of the limit protection device, the distance sensor signals of the anti-collision system, and the weighing sensor signals of the overload protection device are collected respectively. The monitoring data is synchronously sampled and denoised, and the abnormal points and mutation intervals are counted. The performance indicators such as the action accuracy of the limit protection, the anti-collision response time, and the overload protection trigger accuracy are calculated by segmented analysis to evaluate the standardization and completion quality of the maintenance work. Multi-dimensional correlation analysis is performed on various types of data, and the equipment feature data, health indicators, maintenance tasks, management and control plans, and quality assessment results are aligned according to the timestamp. The numerical parameters are normalized in the interval of [-1, 1], and the text descriptions are structured and encoded. A correlation matrix between parameters is established, the correlation coefficient is calculated, and strongly correlated parameter pairs with a correlation degree higher than 0.8 are extracted. The correlation link of fault-symptom-cause is analyzed to form a systematic knowledge base.

[0022] For example, when maintaining the crane wire rope, the strain sensor monitors the periodic fluctuation of the wire rope surface strain value. After filtering and spectrum analysis, it is found that there is abnormal vibration in the 30Hz frequency band. Combined with the broken wire detection data, it is determined that the wire rope has fatigue damage locally. The system automatically generates a maintenance task order and arranges maintenance personnel with matching technical levels to replace the damaged wire rope segment within a specified time period. During the maintenance process, the distribution of obstacles in the hoisting space is dynamically monitored to plan a safe disassembly and assembly path. During acceptance, the load-bearing capacity of the newly replaced wire rope segment is tested by a weighing sensor to confirm the quality of the maintenance.

[0023] In the embodiment of the present application, the present invention realizes real-time monitoring and data collection of the equipment operation status by deploying a multi-dimensional sensor network on the important structural parts and various mechanisms of the crane, providing a reliable data basis for health status assessment; key parameters such as wire rope wear, brake clearance, reducer vibration and motor temperature are analyzed and evaluated to achieve accurate assessment of component status and quantitative expression of health indicators; the basic operation and maintenance module combines the historical load spectrum and working condition data of the crane for maintenance planning and resource scheduling, realizing the reasonable arrangement of maintenance tasks and optimal allocation of resources; the operation guarantee module ensures the safety of maintenance operations by performing real-time safety assessment of the hoisting space, lifting height and working radius. The system management module monitors and evaluates the maintenance process of the crane limit protection device, anti-collision system and overload protection device throughout the entire process, ensuring the standardization and reliability of the maintenance operation. The knowledge base module builds a complete crane intelligent operation and maintenance knowledge base by establishing a multi-dimensional correlation analysis model of equipment feature data sets, component health indicators, maintenance task orders, dynamic management and control plans, and quality assessment results, providing effective knowledge support for subsequent maintenance decisions, thereby realizing the intelligence, precision and standardization of the crane maintenance process, significantly improving maintenance efficiency and reliability, reducing maintenance costs, and extending equipment service life.

[0024] In a specific embodiment, the data acquisition module is specifically used to:

[0025] (1) The strain sensors of the important structural parts of the crane are used to collect the stress data of the structural parts, and the acceleration sensors, angle sensors, and temperature sensors of the lifting mechanism, luffing mechanism, and slewing mechanism are used to collect the vibration data, angle data, and temperature data of the transmission system;

[0026] (2) Perform zero drift correction on the force data of the structural parts, perform 1000 Hz high-order filtering on the vibration data, perform low-pass filtering on the temperature data, and perform mean filtering on the angle data to obtain various parameters after filtering;

[0027] (3) extracting the peak coefficient, standard deviation, and root mean square value time domain characteristics of each parameter after filtering to form time series feature data;

[0028] (4) Performing a 512-point fast Fourier transform on the time series feature data, calculating the power spectrum density and the main frequency component, and obtaining the frequency domain feature data;

[0029] (5) Perform a 95% confidence interval test on the frequency domain feature data, remove outliers that exceed the limit, and perform linear interpolation on the missing data segments to obtain valid operating data;

[0030] (6) The effective operation data is normalized according to the maximum and minimum method, converted to the interval [-1, 1], and an equipment characteristic data set is generated, wherein the equipment characteristic data set includes wire rope wear characteristic data, brake clearance characteristic data, reducer vibration characteristic data, and motor temperature characteristic data.

[0031] Specifically, the data acquisition module collects data through a multi-dimensional sensor network deployed in key locations. Resistive surface strain gauges are installed at important structural parts. The range of this sensor is ±3000με, which is mainly used to collect deformation data of important structural parts in real time. A universal piezoelectric acceleration sensor is installed in the lifting mechanism. According to the transmission scenario of the mechanism, the range is determined to be ±5g, which is used to monitor abnormal vibration of the transmission system during the lifting process. An angle sensor for measuring the pitch angle is installed on the boom, with a range of 0° to 360°, which is used to collect changes in the boom angle during the lifting process. A temperature sensor is installed on the slewing mechanism with a measurement range of -50° to 200°, which is used to obtain temperature data of the slewing motor.

[0032] For important components, the sampling frequencies are 1Hz for resistive surface strain gauges, 1000HZ for general piezoelectric accelerometers, 1Hz for angle sensors, and 1Hz for temperature sensors. The deformation data is decomposed into the fourth order by the db4 wavelet basis function to obtain the detail coefficients of four scales and an approximate coefficient respectively, and the denoised deformation signal is obtained after reconstruction. There is a zero drift problem in the boom pitch data. The mean value of the signal in the static state is calculated and used as the zero offset for correction. All types of pre-processed signals are uniformly sampled and processed, and the time domain statistical features are calculated within a 60-second time window: the peak coefficient reflects the impact of the signal, which is obtained by dividing the maximum value of the signal by the root mean square value; the standard deviation represents the degree of fluctuation of the signal; and the root mean square value reflects the effective value of the signal. The step size of each time window is 10 seconds, that is, the characteristic value is updated every 10 seconds to form continuous time series characteristic data.

[0033] Before fast Fourier transforming the time series feature data, windowing is performed first, and the Hanning window is used to reduce spectrum leakage. 512 points are selected for FFT transformation to obtain a spectrum in the range of 0-50Hz. The power spectrum density is calculated to reflect the distribution of signal energy in the frequency domain. At the same time, the main frequency component, that is, the frequency point with the largest amplitude and its amplitude, is extracted. These frequency domain feature data can reflect the vibration characteristics of the equipment. For the obtained frequency domain feature data, a 95% confidence interval is calculated based on the normal distribution assumption. Data points outside the confidence interval are marked as outliers, which are often related to equipment failure or abnormal operation. For data missing due to communication interruption, sensor failure, etc., if the missing time is less than 1 second, a linear interpolation method is used to repair the data to ensure data continuity.

[0034] The last step is data normalization, using the maximum and minimum value normalization method to map all data to the interval [-1, 1]. In this way, data of different dimensions and orders of magnitude can be uniformly compared and analyzed.

[0035] For example, take a type of gantry crane as an example, such as the strain data of the main beam of a 600t gantry crane, the vibration data of the reducer of the lifting mechanism, and the inclination data of the rigid and flexible legs. The gantry crane performs a lifting operation of 600 tons of weight, which lasts about 30 minutes. The strain sensor in the middle of the main beam of the important structural part records a maximum value of 400 microstrain, and there is no power frequency interference. The strain value fluctuates periodically, and wavelet decomposition finds that there are abnormal fluctuations in the third-level detail coefficient, indicating that there may be local deformation. Time domain feature extraction is performed on these data, and the calculated peak coefficient is 3.2 and the standard deviation is 0.85. FFT analysis finds that there is an interference frequency at 28Hz, which corresponds to the natural vibration frequency of the equipment. Through confidence interval analysis, it is found that there are 3 abnormal points in the force value data, and a continuous data curve is generated after interpolation processing. All features are normalized to form a standardized feature data set.

[0036] In a specific embodiment, the status monitoring module is used to:

[0037] (1) Calculate the wire breakage rate and surface wear based on the wire rope wear characteristic data, calculate the brake pad clearance value and braking torque based on the brake clearance characteristic data, calculate the amplitude and frequency characteristics based on the reducer vibration characteristic data, and calculate the temperature rise rate and heat load based on the motor temperature characteristic data to obtain the component state quantity;

[0038] (2) Collect component status data at high frequency, calculate statistical features within the sliding time window, and form status time series data;

[0039] (3) Compare the state time series data with the preset threshold range, mark the limit-exceeding points and abnormal intervals, and obtain abnormal marking data;

[0040] (4) Data repair is performed on the missing segments in the abnormal labeled data through cubic spline interpolation, and statistical analysis is performed with a 95% confidence level to obtain a reliable state value;

[0041] (5) Perform weighted fusion and normalization calculations on the trusted status values ​​to generate component health indicators.

[0042] Specifically, after acquiring the equipment feature data set, the condition monitoring module performs data classification processing. The original data is divided into four categories according to the detection object: the wire rope wear feature data includes the surface photoelectric signal and the broken wire electromagnetic signal; the brake clearance feature data includes the displacement sensor data and the pressure sensor data; the reducer vibration feature data includes the acceleration and speed signals; the motor temperature feature data includes the stator temperature and the rotor temperature data.

[0043] For the evaluation of the wear state of the wire rope, the calculation formula of wire breakage rate and surface wear is introduced:

[0044]

[0045] Among them, W r Indicates the comprehensive wear rate of wire rope, α i is the number of broken wires at the i-th position, β i is the wear depth of the surface at the i-th location (mm), L t is the total length of the test (m), P s is the wire rope pitch (mm), k1 and k2 are the weight coefficients of wire breakage and wear, and n is the number of detection points.

[0046] The brake condition is evaluated using the following formula:

[0047] M b =γ p F n (1-δ g / δ0)R e

[0048] Among them, M b is the braking torque (N·m), γ p is the friction coefficient, F n is the brake positive pressure (N), δ g is the current gap value (mm), δ0 is the initial gap value (mm), R e is the equivalent friction radius (m).

[0049] The calculation formula of reducer vibration characteristics is:

[0050]

[0051] Among them, V c is the comprehensive vibration characteristic value, H k is the weighting coefficient of the kth frequency band, A k is the amplitude of the kth frequency band (mm), ω k is the angular frequency of the kth frequency band (rad / s), and m is the number of frequency bands.

[0052] Motor temperature characteristic calculation formula:

[0053] T l =ξ s (dT / dt0+η r Q h

[0054] Among them, T l is the heat load index, ξ s is the temperature rise rate weight, dT / dt is the temperature change rate (℃ / s), η r is the thermal resistance coefficient, Q h is the heating power (W).

[0055] A 60-second sliding time window is used for these state quantities, and the window sliding step is 10 seconds. Statistical features such as mean, standard deviation, and peak are calculated in each time window to form a continuous state time series data stream. For wire ropes, focus on the sudden change and cumulative trend of the number of broken wires; for brakes, monitor the gap change rate and torque fluctuations; for reducers, analyze the change law of vibration amplitude; for motors, track the temperature rise rate and thermal equilibrium state. Set the threshold range of each parameter: wire rope broken wire rate threshold 0.15, surface wear depth threshold 2mm; brake gap change threshold 0.5mm, torque fluctuation threshold 5%; reducer vibration speed threshold 5mm / s, acceleration threshold 10m / s 2 The motor temperature rise rate threshold is 2°C / min, and the maximum temperature threshold is 120°C. Data points that exceed the threshold range are marked as abnormal points, and continuous abnormal points constitute an abnormal interval.

[0056] The marked abnormal data segments are processed. If the data missing time is less than 1 second, the cubic spline interpolation method is used for repair. The interpolation uses the endpoint value and the derivative value as the boundary conditions to ensure that the repaired data curve is smooth and continuous. The repaired data is statistically analyzed with a confidence level of 95%, sporadic anomalies are eliminated, and the true state change characteristics are retained. Finally, data fusion is performed, and the wire rope state weight is set to 0.4, the brake state weight to 0.3, the reducer state weight to 0.2, and the motor state weight to 0.1, and the comprehensive health index is calculated. The maximum and minimum value normalization method is used to map the health index to the 0-1 interval. The closer the value is to 1, the better the state.

[0057] For example, by scanning the surface of the wire rope, different degrees of wear were found in many places. The photoelectric sensor detected that the surface wear depth was between 1.2-1.8mm, and the electromagnetic sensor detected that there were 2-3 broken wires locally. Substituting these data into the wear rate calculation formula, the comprehensive wear rate was 0.12. At the same time, the brake clearance detection found that the wear of the brake pad caused the clearance to increase by 0.3mm and the braking torque to decrease by 3%. The reducer vibration monitoring showed abnormalities in the 28Hz and 56Hz frequency bands, and the vibration acceleration reached 8m / s2 The motor temperature rises to 95°C when running at full load, with a temperature rise rate of 1.5°C / min. These data are analyzed through time series and threshold comparison, and abnormal intervals are marked. The state change trend is confirmed through interpolation and statistical analysis, and the component health index is calculated to be 0.85, indicating that the equipment is in a normal state but needs attention.

[0058] In a specific embodiment, the basic operation and maintenance module is used to:

[0059] (1) Correlate the component health index with the crane's historical load spectrum in time series, count the equipment status distribution under different load levels, and obtain historical status data;

[0060] (2) Divide the historical status data and working condition data into time periods, analyze the maintenance demand level and maintenance difficulty coefficient in each time period, and generate maintenance priority data;

[0061] (3) Classify and sort maintenance items according to maintenance priority data, calculate the maintenance time and work space required for each item, and form a maintenance work list;

[0062] (4) Perform maintenance resource statistics according to the maintenance work list, calculate the number of maintenance tools and maintenance personnel required, and obtain resource demand data;

[0063] (5) Match and screen the inventory status of maintenance tools and the skill level of maintenance personnel based on resource demand data to generate a resource allocation plan;

[0064] (6) The maintenance work list and resource allocation plan are combined and arranged in chronological order to form an intelligent maintenance task list.

[0065] Specifically, the basic operation and maintenance module performs time alignment and correlation analysis on component health indicators and historical load spectra. The load spectrum records the load status of the crane at different time points, including lifting weight, number of lifts, working hours and other information. The component health indicators are matched with the load spectrum data according to the timestamp, and a mapping relationship between the health status and the load level is established. Through statistical analysis, the health status distribution characteristics of each component under different load ranges (such as 0-20%, 20%-40%, 40%-60%, 60%-80%, 80%-100% rated load) are obtained.

[0066] The historical status data and working condition data are processed in time segments, and divided into 8-hour work shifts. In each time period, the speed of component degradation, the frequency of failures and maintenance records are counted to calculate the degree of maintenance demand. At the same time, the maintenance difficulty coefficient is considered, including factors such as the complexity of the working environment, the degree of limited maintenance space, and the need for special tools. Through weighted calculation of these indicators, the maintenance priority scores for different time periods are obtained. Based on the maintenance priority data, each maintenance project is graded and sorted. For each maintenance project, the maintenance operation steps are recorded in detail, the standard working hours required for each step are evaluated, and the total maintenance time is calculated. At the same time, the maintenance operation space requirements are calculated, including tool placement space, maintenance personnel operation space, and spare parts storage space. This information is organized into a maintenance operation list, which contains specific parameters such as project priority, estimated working hours, and space requirements.

[0067] Perform resource demand analysis based on the maintenance work list, and count the number and duration of use of various maintenance tools. Maintenance tools include general tools (wrenches, screwdrivers, etc.) and special tools (wire rope flaw detectors, brake clearance gauges, etc.). At the same time, calculate the number of maintenance personnel required, including mechanical maintenance workers, electrical maintenance workers, inspection technicians and other different types of workers. Summarize and form a detailed resource demand list. Screen and match inventory tools and maintenance personnel based on resource demand data. Check the inventory status of tools to confirm the available quantity and integrity of various tools. Then evaluate the skill level of maintenance personnel and match personnel of different levels (senior workers, intermediate workers, junior workers) with the technical requirements of maintenance tasks. Through optimal configuration, a reasonable resource scheduling plan is formed.

[0068] Combine the maintenance work list and resource allocation plan in a time series. Consider the priority of maintenance projects, the availability of personnel and tools, the use conflicts of work sites and other constraints, and reasonably arrange the execution time and sequence of various maintenance tasks. Compile an intelligent maintenance task list with detailed time nodes.

[0069] For example, analysis of recent working condition data revealed that the crane mainly worked in the 40%-60% rated load range, with a cumulative working time of 180 hours. Combined with component health indicators, the wear of the wire rope increased, the brake clearance increased, and the vibration of the reducer increased, among which the wire rope condition deteriorated the fastest. Segmented analysis of historical data found that the morning shift had a high intensity of work and the maintenance environment was dim, which increased the difficulty of maintenance. Therefore, wire rope replacement was listed as the highest priority maintenance project, and it was estimated that 4 mechanical maintenance workers would be required to operate at the same time, with a working time of 6 hours, and special tools including wire rope flaw detectors and hydraulic cutting tools were required. Inspection of maintenance resources found that the current inventory tools met the requirements, but there were only 2 senior workers who were skilled in wire rope replacement operations, and technical personnel from other teams needed to be deployed for support. On the basis of the full preparation of tools and the presence of personnel, the wire rope replacement operation was arranged on the maintenance day next Tuesday, and the relevant information was entered into the maintenance task list.

[0070] In a specific embodiment, the operation guarantee module is used to:

[0071] (1) Decomposing the equipment feature data set into hoisting space feature data, lifting height feature data, and working radius feature data, and performing operation area analysis on the maintenance task order to obtain a space parameter group;

[0072] (2) Extract the obstacle contour points, lifting height limit value, and working radius boundary value of the hoisting space from the spatial parameter group, perform three-dimensional coordinate mapping, and form safety boundary data;

[0073] (3) Calculate the minimum envelope of the safety boundary data, analyze the accessibility of the lifting space, the lifting height margin, and the working radius redundancy, and obtain the safety margin data;

[0074] (4) Divide the maintenance operation area into dangerous levels according to the safety margin data, calculate the collision risk coefficient and interference probability of each area, and generate a risk area map;

[0075] (5) Calculate the optimal path and obstacle avoidance points for maintenance operations based on the risk area map, set the hoisting space access conditions, lifting height limit values, and working radius constraint values, and obtain the control parameter set;

[0076] (6) Combine control parameter sets in time sequence and associate their states to form a dynamic control scheme.

[0077] Specifically, the operation guarantee module classifies and processes the equipment feature data set. The data set is divided into three subsets: hoisting space feature data, including spatial point cloud data scanned by the laser radar; lifting height feature data, including the measurement values ​​of the height sensor and the angle sensor; working radius feature data, including the data of the arm length sensor and the horizontal distance sensor. At the same time, the operation area information in the maintenance task list is parsed to extract parameters such as the location of the operation point, the operation range, and the required space size.

[0078] When extracting spatial parameters, the following three-dimensional space safety boundary mapping formula is used:

[0079]

[0080] Among them, S b represents the spatial safety boundary index, x p ,y p 、z p are the three-dimensional coordinates of the obstacle points, λ1, λ2, λ3 are the weight coefficients in each direction, u i 、v i 、w i is the direction cosine of the obstacle point, N c is the total number of feature points, D s It is the safety distance reference value.

[0081] The minimum envelope of the extracted spatial parameters is calculated, and the discrete spatial point set is constructed into a closed geometric body through the convex hull algorithm. When calculating the accessibility of the lifting space, the positional relationship between the crane's range of motion and the obstacle is considered. The lifting height margin is determined by comparing the rated height with the actual required height, and the working radius redundancy is obtained by calculating the difference between the actual working radius and the maximum allowable radius. Based on the safety margin data, the operating area is divided into three levels: safe area, warning area, and prohibited area. The collision risk is evaluated by calculating the shortest distance between the lifting equipment and the obstacle, combined with the movement speed and direction of the equipment. At the same time, the relative movement between multiple moving parts is considered to calculate the interference probability.

[0082] Plan the maintenance operation path according to the risk area map, use the A* algorithm to calculate the optimal path from the starting point to the target point, and set key obstacle avoidance points on the path. Set access conditions for different areas, including personnel qualification requirements, operation time limits, etc. Set dynamic constraint values ​​for the lifting height and working radius, and adjust them in real time with the working status of the equipment. Organize all control parameters in chronological order and establish the relationship between parameters. For example, the lifting height limit value and the working radius are coupled and need to be adjusted in coordination. At the same time, consider the order of maintenance operations to ensure that the constraints of various parameters match each other.

[0083] For example, the three-dimensional point cloud data of the working area is obtained through the laser radar to identify the positions of fixed obstacles such as overhead lines and pipelines. At the same time, the spatial coordinates of the maintenance points and the required working space are extracted from the maintenance task list. After three-dimensional coordinate mapping, the spatial distribution characteristics of the obstacles are obtained. The discrete point cloud is converted into a regular geometric body through the minimum envelope algorithm, and the minimum distance from the obstacle during the rotation of the boom is calculated. When dividing the danger level, the area within 2 meters from the obstacle is marked as a restricted area, and the range of 2-5 meters is marked as a warning area. According to the spatial distribution characteristics, an optimal maintenance path avoiding high-voltage lines is planned, and 5 obstacle avoidance points are set at key locations. In the generated dynamic control plan, it is stipulated that the maximum lifting height of the boom in this area shall not exceed 80% of the rated height, the working radius shall not exceed 60% of the maximum radius, and the rotation speed shall be reduced to half of the normal speed.

[0084] In a specific embodiment, the system management module is used to:

[0085] (1) Perform task analysis on the maintenance task list and dynamic control plan, extract the maintenance requirements of the crane limit protection device, the maintenance requirements of the anti-collision system, and the maintenance requirements of the overload protection device, and obtain the monitoring parameter table;

[0086] (2) According to the monitoring parameter table, the displacement sensor signal of the limit protection device, the distance sensor signal of the anti-collision system, and the weighing sensor signal of the overload protection device are collected to form original monitoring data;

[0087] (3) Perform noise reduction and data synchronization on the original monitoring data, count the signal mutation points and abnormal intervals, and obtain monitoring feature data;

[0088] (4) Perform segmented statistical analysis on the monitoring characteristic data, calculate the action accuracy of the limit protection device, the response time of the anti-collision system, and the triggering accuracy of the overload protection device, and generate performance index data;

[0089] (5) Evaluate the standardization of maintenance operations according to performance indicator data, mark the completion status and operation quality of key maintenance nodes, and form evaluation statistical data;

[0090] (6) Perform weighted calculation and comprehensive scoring on the evaluation statistical data and output the quality evaluation results.

[0091] Specifically, the system management module extracts specific monitoring requirements from the maintenance task list and dynamic control plan. The maintenance requirements of the limit protection device include the action position, action time and reset state of the limit switch in each direction; the maintenance requirements of the anti-collision system include the obstacle recognition distance, warning trigger time and braking response time; the maintenance requirements of the overload protection device include weighing accuracy, alarm threshold and action reliability. These requirements are organized into a structured monitoring parameter table to clarify the detection method and judgment criteria of each parameter. According to the monitoring parameter table, the data acquisition plan is set to synchronously collect the sensor signals of the three major protection devices. The limit protection device uses an LVDT displacement sensor with a measurement range of 0-500mm and a resolution of 0.1mm; the anti-collision system uses an ultrasonic distance sensor with a measurement range of 0-10m and a response time of 20ms; the overload protection device uses a strain gauge weighing sensor with a range of 0-50t and an accuracy level of 0.1. The raw data collected by these sensors contains signal values ​​and timestamps.

[0092] The collected raw data are uniformly converted to a sampling frequency of 100Hz, and a Butterworth low-pass filter is used for noise reduction, with the cutoff frequency set to 40Hz. The three signals are time-aligned, with the earliest timestamp as the reference point, and the synchronization of the data is ensured by linear interpolation. The first-order derivative of the signal is calculated, and a dynamic threshold is set to identify the signal mutation point. When the signal change rate exceeds the set threshold, it is marked as a mutation point. Continuous mutation points constitute an abnormal interval. The monitoring feature data after synchronization processing is segmented and statistically analyzed. The action accuracy of the limit protection device is obtained by calculating the deviation between the actual action position and the set position; the response time of the anti-collision system is determined from the time interval from the detection of the obstacle to the triggering of the brake; the triggering accuracy of the overload protection device is calculated by multiple loading tests, and the ratio of the number of accurate triggers to the total number of tests is statistically calculated. These performance indicators reflect the working status of each protection device.

[0093] Evaluate the standardization of maintenance operations according to various performance indicators. Mark key maintenance nodes, including steps such as device debugging, functional testing, and parameter setting. Record the completion status of each node, including whether the operation process is standardized, whether the test data meets the standards, whether the fault is eliminated, etc. For unqualified items, record the specific problems and treatment methods in detail. All evaluation data form a structured statistical table. Comprehensively process the evaluation statistical data and set weight coefficients for different evaluation items. The weights of the three systems of limit protection, anti-collision, and overload protection are 0.4, 0.3, and 0.3 respectively. The weights of different evaluation items are further subdivided within each system. The quality assessment score is calculated by weighted average.

[0094] For example, check the maintenance requirements of each protective device: the lifting limit action position error should be less than 5mm, and the boom amplitude limit action time should be less than 0.5s; the anti-collision system must trigger an alarm within 1s after detecting an obstacle and complete braking within 2s; the overload protection must accurately alarm when the load exceeds 110% of the rated value. Through sensor data collection, it was found that the lifting limit action position deviation at the upper limit reached 7mm, which was reduced to 4mm after adjusting the limit switch installation position; the anti-collision system detection distance was normal, but the alarm delay reached 1.3s, which was shortened to 0.8s after the controller parameters were adjusted; the overload protection weighing system showed an error of 0.5%, and triggered the reliability test 10 times and all operated normally. After weighted calculation of these data, the maintenance quality score was 92 points, which met the acceptance criteria.

[0095] In a specific embodiment, the knowledge base module is used to:

[0096] (1) Align equipment feature data sets, component health indicators, maintenance task orders, dynamic control plans, and quality assessment results according to timestamps to form time-series associated data;

[0097] (2) Normalize the numerical parameters in the time series related data and perform structured encoding on the text parameters to obtain standardized related data;

[0098] (3) Establish a correlation matrix between parameters according to the standardized correlation data, calculate the correlation coefficient between each parameter pair, and generate a correlation index;

[0099] (4) Arrange the correlation indexes in descending order according to their numerical values, extract parameter pairs with correlations higher than 0.8, and form a strongly correlated parameter set;

[0100] (5) Perform causal link analysis based on the strongly correlated parameter set, establish a fault-symptom-cause association mapping table, and obtain a knowledge rule set;

[0101] (6) Verify and classify the knowledge rule set to form a knowledge base for intelligent crane operation and maintenance.

[0102] Specifically, the knowledge base module performs time alignment processing on various types of data. The equipment feature data set contains real-time collected sensor data, the component health index records the equipment status assessment results, the maintenance task list contains specific maintenance operation records, the dynamic control plan records the safety control parameters, and the quality assessment results contain maintenance effect verification data. These data use a unified timestamp format with millisecond accuracy. Through timestamp comparison, data from different sources are matched one by one according to the time sequence to establish a time index relationship. The aligned data is standardized, and the maximum and minimum value normalization method is used to map numerical parameters such as sensor measurement values, health indicator values, and control parameter values ​​to the [-1, 1] interval. For text parameters such as fault descriptions, maintenance measures, and evaluation opinions, a professional terminology dictionary is established to convert text information into fixed-format codes. For example, "wire rope broken wire" is encoded as "F001" and "brake clearance is too large" is encoded as "F002" to achieve structured expression of text information.

[0103] The correlation between parameters is calculated based on the standardized data. The Pearson correlation coefficient is used to calculate the linear correlation between numerical parameters, and the mutual information entropy is used to calculate the correlation between categorical parameters. An N×N correlation matrix is ​​constructed, where N is the total number of parameters. Each element in the matrix represents the correlation coefficient between the corresponding parameter pairs, and the value range is [-1, 1]. The closer the absolute value is to 1, the stronger the correlation. The elements in the correlation matrix are arranged in descending order according to the absolute value, and the parameter pairs with the absolute value of the correlation coefficient greater than 0.8 are extracted. These parameter pairs have a significant correlation. For example, the correlation between the vibration frequency of the wire rope and the wire breakage rate, the correlation between the brake temperature and the braking torque, and the correlation between the motor current and the bearing temperature. These strongly correlated parameters constitute the skeleton structure of the parameter network.

[0104] Analyze the causal relationship between parameters based on the strongly correlated parameter set and establish a fault evolution link. Determine the order of parameter changes through time series analysis, and judge the causality between parameters based on professional knowledge. Organize the analysis results into a three-layer structure: the fault phenomenon layer records the abnormal performance of the equipment, the symptom feature layer records the warning signal before the fault occurs, and the root cause layer records the essential cause of the fault. In this way, a fault knowledge graph is constructed. Verify and classify the constructed knowledge rules. Classify by component type, including mechanical transmission, electrical control, safety protection, etc. Then classify by fault level, including critical faults, major faults, general faults, etc. Finally, classify by maintenance type, including daily maintenance, regular inspection, fault repair, etc. The organized knowledge base is easy to retrieve and apply quickly.

[0105] For example, all brake-related data, including brake pad temperature, braking torque, friction coefficient, action time and other parameters, are collected and aligned in time. Numerical parameters such as temperature and torque are normalized, and text descriptions such as "brake slip" and "brake overheating" are encoded. Correlation analysis found that the correlation between brake pad temperature and braking torque is 0.85, indicating that the two are highly correlated. Further analysis found that the decrease in braking torque precedes the increase in temperature, and the increase in temperature leads to a decrease in the friction coefficient, forming a fault knowledge rule: abnormal braking torque (root cause) leads to an increase in brake pad temperature (fault sign), which in turn leads to a decrease in braking performance (fault phenomenon).

[0106] In a specific embodiment, if Figure 2 FIG. 2 is another schematic diagram of the intelligent safety assurance platform for cranes in an embodiment of the present application. The intelligent safety assurance platform for cranes further includes:

[0107] Network communication module, used to encrypt and compress the transmission of equipment feature data sets, establish data transmission channels, and realize multi-network protocol conversion among industrial real-time Ethernet, wireless communication networks, and mobile communication networks;

[0108] The message center module is used to classify and prioritize the component health indicators, the maintenance task list, the dynamic control plan and the quality assessment results, and realize data interaction and information sharing among various functional modules through the message subscription and distribution mechanism;

[0109] The interface module is used to provide a unified data access interface and application programming interface to support the call and integration of equipment feature data sets, maintenance task orders and quality assessment results by third-party systems.

[0110] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A crane intelligent safety assurance platform, characterized in that: The crane intelligent safety assurance platform includes: The data acquisition module is used to collect equipment operating parameters through a multi-dimensional sensor network arranged on important structural parts, hoisting mechanism, luffing mechanism and slewing mechanism of the crane, and perform feature extraction and standardization on the equipment operating parameters to obtain an equipment feature data set; A condition monitoring module is used to analyze and evaluate the wear of the crane wire rope, brake clearance, reducer vibration and motor temperature according to the equipment characteristic data set to generate component health indicators; A basic operation and maintenance module is used to perform maintenance planning and resource scheduling based on the component health indicators combined with the crane's historical load spectrum and operating condition data to form an intelligent maintenance task list; An operation guarantee module is used to perform safety assessment on the crane hoisting space, lifting height and working radius according to the equipment characteristic data set and the maintenance task list, and generate a dynamic control plan; A system management module, used to monitor the maintenance process of the crane limit protection device, anti-collision system and overload protection device according to the maintenance task list and dynamic control plan, and output quality assessment results; The knowledge base module is used to perform multi-dimensional correlation analysis on the equipment feature data set, component health indicators, maintenance task orders, dynamic control plans and quality assessment results to build a crane intelligent operation and maintenance knowledge base.

2. The intelligent safety assurance platform for cranes according to claim 1 is characterized in that: The data acquisition module is specifically used for: The strain sensors of the important structural parts of the crane are used to collect the stress data of the structural parts, and the acceleration sensors, angle sensors and temperature sensors of the hoisting mechanism, luffing mechanism and slewing mechanism are used to collect the vibration data, angle data and temperature data of the transmission system; Perform zero drift correction on the force data of the structural component, perform 1000 Hz high-order filtering on the vibration data, perform low-pass filtering on the temperature data, and perform mean filtering on the angle data to obtain various parameters after filtering; Extracting peak coefficient, standard deviation, and root mean square value time domain characteristics of each parameter after filtering to form time series feature data; Performing a 512-point fast Fourier transform on the time series feature data, calculating the power spectrum density and the main frequency component, and obtaining frequency domain feature data; Perform a 95% confidence interval test on the frequency domain feature data, remove outliers that exceed the limit, and perform linear interpolation on the missing data segments to obtain valid operating data; The effective operation data is normalized according to the maximum and minimum method, converted to the interval [-1, 1], and the equipment characteristic data set is generated, wherein the equipment characteristic data set includes wire rope wear characteristic data, brake clearance characteristic data, reducer vibration characteristic data and motor temperature characteristic data.

3. The intelligent safety assurance platform for cranes according to claim 2 is characterized in that: The state monitoring module is used to: The wire breakage rate and surface wear degree are calculated according to the wire rope wear characteristic data, the brake pad clearance value and braking torque are calculated according to the brake clearance characteristic data, the amplitude and frequency characteristics are calculated according to the reducer vibration characteristic data, and the temperature rise rate and heat load are calculated according to the motor temperature characteristic data to obtain the component state quantity; High-frequency acquisition of the component state quantity is performed, and statistical features within a sliding time window are calculated to form state time series data; Compare the state time series data with a preset threshold range, mark the exceeding limit points and abnormal intervals, and obtain abnormal marking data; Repair the missing segments in the abnormal marked data by cubic spline interpolation, perform 95% confidence statistical analysis, and obtain a reliable state value; The trusted state values ​​are weighted and fused and normalized to generate the component health index.

4. The intelligent safety assurance platform for cranes according to claim 1 is characterized in that: The basic operation and maintenance module is used to: Performing time series correlation between the component health index and the historical load spectrum of the crane, and statistically analyzing the equipment status distribution under different load levels to obtain historical status data; Divide the historical status data and the operating condition data into time periods, analyze the maintenance demand degree and maintenance difficulty coefficient in each time period, and generate maintenance priority data; The maintenance items are sorted according to the maintenance priority data, the maintenance time and working space required for each item are calculated, and a maintenance work list is formed; Perform maintenance resource statistics according to the maintenance work list, calculate the required number of maintenance tools and maintenance personnel configuration, and obtain resource demand data; Matching and screening the inventory status of maintenance tools and the skill level of maintenance personnel according to the resource demand data to generate a resource allocation plan; The maintenance work list and the resource allocation plan are combined and arranged in chronological order to form the intelligent maintenance task list.

5. The intelligent safety assurance platform for cranes according to claim 1 is characterized in that: The operation guarantee module is used to: Decomposing the equipment characteristic data set into hoisting space characteristic data, lifting height characteristic data and working radius characteristic data, and performing operation area analysis on the maintenance task list to obtain a space parameter group; Extract the hoisting space obstacle contour points, lifting height limit value, and working radius boundary value from the spatial parameter group, perform three-dimensional coordinate mapping, and form safety boundary data; Performing minimum envelope calculation on the safety boundary data, analyzing the accessibility of the hoisting space, the lifting height margin, and the working radius redundancy, and obtaining safety margin data; According to the safety margin data, the maintenance operation area is divided into danger levels, the collision risk coefficient and interference probability of each area are calculated, and a risk area map is generated; Calculate the optimal path and obstacle avoidance points of the maintenance operation according to the risk area map, set the hoisting space access conditions, lifting height limit value, and working radius constraint value, and obtain a control parameter set; The control parameter sets are combined in time sequence and associated in state to form the dynamic control scheme.

6. The intelligent safety assurance platform for cranes according to claim 1, characterized in that: The system management module is used to: Perform task analysis on the maintenance task list and the dynamic control plan, extract the maintenance requirements of the crane limit protection device, the maintenance requirements of the anti-collision system and the maintenance requirements of the overload protection device, and obtain a monitoring parameter table; According to the monitoring parameter table, the displacement sensor signal of the limit protection device, the distance sensor signal of the anti-collision system and the weighing sensor signal of the overload protection device are collected to form original monitoring data; The original monitoring data is subjected to noise reduction processing and data synchronization, and signal mutation points and abnormal intervals are counted to obtain monitoring feature data; Performing segmented statistical analysis on the monitoring characteristic data, calculating the action accuracy of the limit protection device, the response time of the anti-collision system, and the triggering accuracy of the overload protection device, and generating performance index data; Evaluate the standardization of maintenance operations according to the performance indicator data, mark the completion status and operation quality of key maintenance nodes, and form evaluation statistical data; The evaluation statistical data are weighted and comprehensively scored, and the quality evaluation result is output.

7. The intelligent safety assurance platform for cranes according to claim 1, characterized in that: The knowledge base module is used to: Aligning the equipment feature data set, the component health index, the maintenance task list, the dynamic control plan, and the quality assessment result according to timestamps to form time-series associated data; Normalizing the numerical parameters in the time series associated data and performing structured coding on the text parameters to obtain standardized associated data; Establishing a correlation matrix between parameters according to the standardized correlation data, calculating the correlation coefficient between each parameter pair, and generating a correlation index; Arrange the correlation indexes in descending order according to their numerical values, extract parameter pairs with correlations higher than 0.8, and form a strongly correlated parameter set; Perform causal link analysis based on the strongly correlated parameter set, establish a fault-symptom-cause association mapping table, and obtain a knowledge rule set; The knowledge rule set is verified and classified to form the crane intelligent operation and maintenance knowledge base.

8. The intelligent safety assurance platform for cranes according to claim 1, characterized in that: The crane intelligent safety assurance platform also includes: A network communication module is used to encrypt and compress the device feature data set for transmission, establish a data transmission channel, and realize multi-network protocol conversion among industrial real-time Ethernet, wireless communication network, and mobile communication network; A message center module is used to classify and prioritize the component health indicators, maintenance task orders, dynamic control plans, and quality assessment results, and to achieve data interaction and information sharing among functional modules through a message subscription and distribution mechanism; The interface module is used to provide a unified data access interface and application programming interface to support the invocation and integration of the equipment feature data set, the maintenance task list and the quality assessment result by a third-party system.

Images