Coal port transportation equipment safety diagnosis method and system

Abstract

This invention relates to the field of equipment safety diagnosis and data processing technology, specifically a method and system for safety diagnosis of transportation equipment in coal port areas. The method includes: acquiring port environment data and transportation equipment data to generate a disturbance behavior data sequence, and constructing an environmental directional entropy field; generating a disturbance phase vector set of transportation equipment in the environmental directional entropy field; using an information resonance identification model to perform combined calculations on disturbances in all directions of the disturbance phase vector set, performing back-folding inference on resonant disturbance nodes that satisfy the continuous offset resonance condition, and calculating the asymmetric stability index; constructing a disturbance network topology, determining the disturbance propagation path and information impedance, and diagnosing the safety of transportation equipment based on the information impedance of the disturbance propagation path. This invention achieves dynamic safety diagnosis of transportation equipment in coal port areas by performing information resonance identification between the coal port environment and transportation equipment operation data.

Description

Technical Field

[0001] This invention relates to the field of equipment safety diagnosis and data processing technology, specifically to a method and system for safety diagnosis of transportation equipment in coal port areas. Background Technology

[0002] In coal port transportation systems, large rail-mounted transport equipment such as ship loaders, stacker-reclaimers, and gantry cranes operate for extended periods in the high-salt, high-humidity, and high-wind-pressure coastal port environment. The operational safety of port transport equipment depends on the long-term stability of the rail-mounted mechanism, drive system, and structural connectors. However, existing safety monitoring systems are mostly based on single-dimensional signals such as stress sensing, vibration analysis, or current monitoring for status identification. These methods mainly focus on changes in structural mechanical properties or mechanical wear characteristics, and are difficult to reflect the coupling relationship between environmental disturbances and equipment response. In the complex environment of the port, abnormal equipment operating status is often caused by the long-term bias of environmental factors. The static threshold identification or fixed rule judgment of traditional models cannot capture the dynamic impact of directional environmental disturbances on the operational stability of equipment. In actual operation, coal port areas are constantly affected by prevailing winds and high salt spray environments, making it easy for uneven corrosion layers to form on the surface of metal components of track-mounted transport equipment. Due to the obvious directional exposure differences in the equipment structure, the windward side components are exposed to high salt ion concentration areas for a long time, while the leeward side is affected by the coupled influence of the coal port environment, resulting in poor corrosion and equipment safety issues. This asymmetric corrosion causes state parameters such as track wheel pressure, running current, and rotational angular velocity to exhibit slow drift and periodic imbalance in different directions. In the stage where directional corrosion instability gradually develops, the equipment exhibits slight uneven running and asymmetric wear characteristics, but no significant deformation has yet appeared at the structural level, making it impossible for traditional threshold-triggered safety diagnosis methods to provide early warning. Summary of the Invention

[0003] The purpose of this invention is to provide a safety diagnosis method and system for coal port transportation equipment, in order to solve the equipment safety problem mentioned in the background art, where the equipment structure has obvious directional exposure differences in space, the windward side components are in a high salt ion concentration area for a long time, while the leeward side is affected by the coupling effect of the coal port environment, resulting in poor corrosion.

[0004] To achieve the above objectives, the technical solution of the present invention is: a safety diagnosis method for coal port transportation equipment, comprising: S1. Obtain environmental data and transportation equipment data of the coal port area, perform time-series alignment according to the preset time window and generate a disturbance behavior data sequence, and construct an environmental directional entropy field based on the disturbance behavior data sequence; The environmental directional entropy field is used to describe the spatial distribution characteristics of port area environmental data. S2. Discretize the transportation equipment data according to a preset time window and perform corresponding mapping processing in the environmental directional entropy field to obtain a set of state disturbance trajectories. Then, perform joint calculations on the amplitude and direction of the set of state disturbance trajectories to generate a set of disturbance phase vectors of the transportation equipment in the environmental directional entropy field. The state disturbance trajectory set is a set of trajectories formed by mapping equipment operation data in the environmental directional entropy field. The disturbance phase vector set is generated based on the state disturbance trajectory set and is used to describe the temporal disturbance characteristics of the transportation equipment operation state under the influence of the directional environment. S3. Use the information resonance identification model to combine and calculate the disturbances in all directions in the disturbance phase vector set to obtain the resonant disturbance node set. Perform back-folding inference on the resonant disturbance nodes that satisfy the continuous offset resonance condition to calculate the asymmetric stability index. The information resonance identification model is used to identify resonant perturbation nodes that satisfy the continuous offset resonance condition. S4. Construct a disturbance network topology based on the asymmetric stability index, sort and simplify the disturbance network topology, determine the disturbance propagation path and the information impedance of the disturbance propagation path, and diagnose the safety of transportation equipment based on the information impedance of the disturbance propagation path. The disturbance network topology is a directed weighted network constructed with resonant disturbance nodes as nodes and the information transmission relationship between resonant disturbance nodes as edges. It is used to characterize the transmission structure and mutual influence relationship of different directional disturbance behaviors in the time and space dimensions.

[0005] Preferably, in S1, the disturbance behavior data sequence is a time-series multi-source data set formed by synchronizing port area environmental data and transportation equipment data within a preset time window, used to represent the impact of port area environmental disturbance behavior on the operating status of transportation equipment; the environmental directional entropy field is a spatial information entropy distribution field calculated based on the distribution probability of port area environmental data in the disturbance behavior data sequence, with spatial direction angle as the independent variable. The specific method for constructing an environmental directional entropy field based on disturbance behavior data sequences includes: dividing the port area environmental data within each preset time window into spatial directional partitions, calculating the information entropy density value corresponding to each directional partition, and using the directional interval corresponding to the prevailing wind direction as a reference axis, normalizing and interpolating the information entropy density values ​​of all directional intervals through the directional entropy density function to form a continuous environmental directional entropy field.

[0006] Preferably, in S2, the state disturbance trajectory set is a time-series trajectory set formed by discretizing the transportation equipment data within a preset time window and mapping it according to the directional partitioning result of the environmental directional entropy field. Each time-series trajectory in the time-series trajectory set consists of a time index, directional coordinates and corresponding equipment state value, and is used to record the discrete response process of the transportation equipment's operating state in different directional entropy density intervals. The method for discretizing transportation equipment data according to a preset time window and performing corresponding mapping processing in the environmental directional entropy field includes: within each preset time window, calculating the difference amplitude of adjacent sampling points based on the rate of change of equipment operating data; when the difference amplitude exceeds a dynamic threshold, determining the sampling point as an effective disturbance point; matching the equipment state parameters of the effective disturbance point with the directional interval under the same time window in the environmental directional entropy field; connecting the effective disturbance points in chronological order according to the matching results to form a directional disturbance trajectory line; and aggregating all directional disturbance trajectory lines according to the preset time window number and directional partition number to generate a state disturbance trajectory set.

[0007] Preferably, in S2, the disturbance phase vector set is a composite vector set generated based on the state disturbance trajectory set. Each disturbance phase vector consists of a disturbance amplitude component and a direction phase component, and is used to describe the temporal disturbance characteristics of the transportation equipment in the environmental directional entropy field. The method for performing joint amplitude and direction calculations on the set of state disturbance trajectories includes: standardizing the disturbance amplitude of each trajectory line in the set of state disturbance trajectories; performing vector projection operation on the amplitude sequence and the environmental direction angle data within the same direction interval to calculate the phase angle difference; forming a composite vector by combining the amplitude sequence and the phase angle difference within each preset time window according to the corresponding index; and performing weighted aggregation on the composite vector set of all preset time windows to form a disturbance phase vector set.

[0008] Preferably, in S3, the information resonance identification model is a statistical inference model based on the perturbation phase vector set. By establishing the resonance energy distribution matrix of the perturbation phase components in different directions, it is used to identify the resonance perturbation relationship that satisfies the continuous offset resonance condition in the time series. The method for combining and calculating perturbations in all directions of the perturbation phase vector set using the information resonance identification model includes: establishing a time-series vector set for each directional component of the perturbation phase vector set, calculating the cross-correlation coefficient between adjacent directional components; when the cross-correlation coefficient exceeds a preset correlation threshold, determining that the directional pair is a resonance directional pair; performing energy spectrum analysis on all resonance directional pairs, calculating the phase resonance energy value and forming a resonance energy distribution matrix, and performing eigenvalue decomposition on the resonance energy distribution matrix to extract the main eigenvectors as feature representations of the resonance center; and forming a resonance perturbation node set with the resonance energy peak position and its corresponding directional pair as nodes.

[0009] Preferably, in S3, the continuous offset resonance condition refers to the condition that the phase center of the resonant perturbation node continuously shifts in the directional angle dimension within adjacent time windows and the shift amplitude exceeds the preset minimum phase drift threshold. This condition is used to determine whether the resonant perturbation node is in a continuous phase shift state during the time evolution process. The back-reflection inference is a spatial back-tracking calculation process based on the resonant perturbation node that satisfies the continuous offset resonance condition. It is used to determine the back propagation path of the resonant energy in the environmental directional entropy field and its energy back-reflection center.

[0010] Preferably, in step S3, the asymmetric stability index is a scalar quantitative index calculated based on the combination of the reversion rate and the phase drift rate, used to characterize the stability state of the transportation equipment under directional disturbances; the value range of the asymmetric stability index is a positive real number, and the higher the asymmetric stability index, the greater the degree of imbalance in the equipment's operating state under directional disturbances.

[0011] Preferably, in step S4, the specific method for constructing the perturbation network topology based on the asymmetric stability index is as follows: extract the asymmetric stability index from the resonant perturbation nodes and use it as input; calculate the synchronicity of the changes in the asymmetric stability index between any two resonant perturbation nodes; when the synchronicity exceeds a preset association threshold, establish a directed edge between the corresponding resonant perturbation nodes, and determine the direction of the edge according to the phase difference of the changes in the asymmetric stability index of the resonant perturbation nodes; the weight of each directed edge is determined by the relative ratio of the rates of change of the asymmetric stability index between the two resonant perturbation nodes; and a perturbation network topology containing multiple layers of directed edges is formed.

[0012] Preferably, in S4, the disturbance propagation path refers to the shortest directed path sequence connecting the source node and the target node in the disturbance network, used to represent the propagation path of directional disturbances between different spatial nodes; the information impedance of the disturbance propagation path is a metric characterizing the difficulty of the disturbance behavior being transmitted on the disturbance propagation path; The method for calculating the information impedance of the disturbance propagation path is as follows: the system takes the reciprocal of the weight of each directed edge on the disturbance propagation path and accumulates them in the order of the path to obtain the information impedance value of the disturbance propagation path; when the path contains multiple intermediate nodes, the system normalizes the rate of change of the asymmetric stability index of each node in the path, and uses the normalization rate as the impedance correction factor to correct the information impedance value to obtain the information impedance of the disturbance propagation path.

[0013] On the other hand, the present invention provides a safety diagnostic system for coal port transportation equipment, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the aforementioned safety diagnostic method for coal port transportation equipment.

[0014] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects: 1. In this invention, a dynamic coupling relationship between directional environmental disturbances and equipment state response is constructed based on an information resonance identification model. This can identify the potential impact of environmental changes on equipment operational stability from multi-source port area data. By conducting information-based data analysis on the synchronicity and energy distribution between disturbances in different directions, early signs of "directional corrosion instability" can be identified before significant mechanical deformation occurs in the equipment. This enables advance judgment of hidden risks under long-term environmental effects on track-moving equipment. 2. In this invention, the intelligent identification and dynamic early warning of equipment risk propagation paths are achieved through perturbation network analysis based on asymmetric stability index. Based on the changing law of network information impedance, the risk transmission direction and intensity of different directional structures can be quantified, thereby realizing real-time diagnosis and closed-loop control of equipment safety risks. Attached Figure Description

[0015] Figure 1 This is a flowchart of one embodiment of the present invention. Detailed Implementation

[0016] Example 1, as Figure 1 As shown, the specific implementation steps of the safety diagnosis method for coal port transportation equipment proposed in this invention are as follows: S1. Obtain environmental data and transportation equipment data of the coal port area, perform time-series alignment according to the preset time window and generate a disturbance behavior data sequence, and construct an environmental directional entropy field based on the disturbance behavior data sequence; The environmental directional entropy field is used to describe the spatial distribution characteristics of port area environmental data. S2. Discretize the transportation equipment data according to a preset time window and perform corresponding mapping processing in the environmental directional entropy field to obtain a set of state disturbance trajectories. Then, perform joint calculations on the amplitude and direction of the set of state disturbance trajectories to generate a set of disturbance phase vectors of the transportation equipment in the environmental directional entropy field. The state disturbance trajectory set is a set of trajectories formed by mapping equipment operation data in the environmental directional entropy field. The disturbance phase vector set is generated based on the state disturbance trajectory set and is used to describe the temporal disturbance characteristics of the transportation equipment operation state under the influence of the directional environment. S3. Use the information resonance identification model to combine and calculate the disturbances in all directions in the disturbance phase vector set to obtain the resonant disturbance node set. Perform back-folding inference on the resonant disturbance nodes that satisfy the continuous offset resonance condition to calculate the asymmetric stability index. The information resonance identification model is used to identify resonant perturbation nodes that satisfy the continuous offset resonance condition. S4. Construct a disturbance network topology based on the asymmetric stability index, sort and simplify the disturbance network topology, determine the disturbance propagation path and the information impedance of the disturbance propagation path, and diagnose the safety of transportation equipment based on the information impedance of the disturbance propagation path. The disturbance network topology is a directed weighted network constructed with resonant disturbance nodes as nodes and the information transmission relationship between resonant disturbance nodes as edges. It is used to characterize the transmission structure and mutual influence relationship of different directional disturbance behaviors in the time and space dimensions.

[0017] In this embodiment S1, the disturbance behavior data sequence is a time-series multi-source data set formed by synchronizing port area environmental data and transportation equipment data within a preset time window, used to represent the impact of port area environmental disturbance behavior on the operating status of transportation equipment; the environmental directional entropy field is a spatial information entropy distribution field calculated based on the distribution probability of port area environmental data in the disturbance behavior data sequence, with spatial direction angle as the independent variable. The specific method for constructing an environmental directional entropy field based on disturbance behavior data sequences includes: dividing the port area environmental data within each preset time window into spatial directional partitions, calculating the information entropy density value corresponding to each directional partition, and using the directional interval corresponding to the prevailing wind direction as a reference axis, normalizing and interpolating the information entropy density values ​​of all directional intervals through the directional entropy density function to form a continuous environmental directional entropy field.

[0018] In this embodiment S1, the port area environmental data is obtained through meteorological sensing units, temperature and humidity acquisition units, salt spray concentration sensing units, and wind direction detection devices deployed in the wharf operation area, along the track, and above the equipment. The transportation equipment data is collected in real time through current sensors, strain gauges, torque detection devices, and angular velocity sampling modules installed at the structural nodes of the equipment. The data is time-stamped at the acquisition end with a uniform sampling period and the data stream is aggregated through the port area monitoring server. The port area environmental data of the coal port includes multi-dimensional physical quantity data that can characterize the external environmental state of the port area, specifically including environmental monitoring parameters such as wind direction, wind speed, air humidity, temperature gradient, salt spray concentration, and air pressure change rate. The transportation equipment data includes dynamic signal data that can reflect the operating status of the track-moving equipment, specifically including operating status parameters such as drive current, track strain, wheel pressure, rotational angular velocity, and motor power signal.

[0019] In this embodiment S1, the preset time window is a fixed or sliding time interval determined based on the operating characteristic cycle of port equipment. It is used to ensure the synchronous alignment of environmental data and equipment data in the time dimension. The size of the time window can be set according to the port operation rhythm, equipment travel cycle, or wind direction change rate. When performing time alignment, the system first performs synchronous filtering of environmental data and equipment data according to the start and end time of the time window, and uses a time interpolation algorithm to fill in the missing sampling points. Then, it performs normalization processing on the two types of data with the same time index and forms a one-to-one corresponding data pair. All environmental data and equipment data in all time windows are arranged in the order of sampling time to form a disturbance behavior data sequence. This data sequence constitutes a multi-source synchronous time series set. Each time slice contains a set of environmental parameters and their corresponding equipment operating parameters, which are used to reflect the dynamic impact of port environmental disturbances on the operating status of transportation equipment.

[0020] In this embodiment S1, the determination of the dominant wind direction reference axis is achieved by fusing historical wind direction statistics and real-time wind direction data in the port area. First, statistical analysis is performed on wind direction records from the past year in the port area to calculate the weighted average angle of wind frequency occurrence in each direction, thus obtaining the historical dominant wind direction. During the equipment's operating cycle, the system synchronously collects real-time wind direction information. When the deviation between the real-time wind direction and the historical dominant wind direction is less than a preset threshold, the historical dominant wind direction is used as the reference axis; otherwise, the moving average direction of the real-time wind direction is used as the reference axis. The dominant wind direction reference axis is used to establish a spatial direction coordinate system, its function being to unify the directional identification of environmental data in the spatial dimension, and it does not participate in the numerical calculation of information entropy density. In the calculation of entropy density, each environmental parameter (including wind speed, humidity, salt spray concentration, temperature gradient, etc.) is normalized in each directional interval to eliminate the dimensional differences between different physical quantities. The normalization process uses a linear standardization method to transform each parameter value to the [0,1] interval. The frequency of occurrence of the normalized environmental parameters in the same time window within the spatial directional interval is statistically analyzed to obtain the directional distribution probability. The information entropy density value is calculated from the distribution probability of each directional interval and is used to measure the degree of dispersion of environmental disturbance information in each direction. The probability distribution of information entropy density is based on the distribution probability of environmental parameters in the spatial directional interval, and the dominant wind direction reference axis only determines the calibration method of the directional interval.

[0021] In this embodiment S1, during the process of constructing the environmental directional entropy field based on the disturbance behavior data sequence, all sampling points are first divided into several spatial directional partitions according to wind direction in the port area spatial coordinate system. The spatial directional partition refers to several directional intervals divided at fixed angular intervals with the prevailing wind direction as the reference axis. Each directional interval corresponds to a subset of environmental data with similar directional characteristics. The information entropy density value is a measure of the uncertainty of the distribution of environmental parameters within the directional interval. Its calculation method is as follows: perform entropy calculation on the probability distribution of environmental parameters within each directional interval. The information entropy density value is equal to negative one multiplied by the sum of the products of each probability value and its logarithm. After calculation, the information entropy density values ​​of all directional intervals are combined into a directional entropy density set, and the directional interval corresponding to the prevailing wind direction is used as a normalization reference to construct a directional entropy density function. The directional entropy density function is used to map the discrete directional entropy density set into a continuous spatial entropy field. Its mathematical form can be implemented using existing smoothing interpolation functions or kernel density estimation functions. The function can be any one of Gaussian kernel function, spline interpolation function, or bilinear interpolation function, used to establish a continuous entropy density transition relationship in each directional interval. Finally, by normalizing the continuous entropy distribution output by the directional entropy density function, the environmental directional entropy field is obtained.

[0022] In this embodiment S2, the state disturbance trajectory set is a time-series trajectory set formed by discretizing the transportation equipment data within a preset time window and mapping it according to the directional partitioning result of the environmental directional entropy field. Each time-series trajectory in the time-series trajectory set consists of a time index, directional coordinates and corresponding equipment state value, which is used to record the discrete response process of the transportation equipment's operating state in different directional entropy density ranges. The method for discretizing transportation equipment data according to a preset time window and performing corresponding mapping processing in the environmental directional entropy field includes: within each preset time window, calculating the difference amplitude of adjacent sampling points based on the rate of change of equipment operating data; when the difference amplitude exceeds a dynamic threshold, determining the sampling point as an effective disturbance point; matching the equipment state parameters of the effective disturbance point with the directional interval under the same time window in the environmental directional entropy field; connecting the effective disturbance points in chronological order according to the matching results to form a directional disturbance trajectory line; and aggregating all directional disturbance trajectory lines according to the preset time window number and directional partition number to generate a state disturbance trajectory set.

[0023] In this embodiment S2, the disturbance phase vector set is a composite vector set generated based on the state disturbance trajectory set. Each disturbance phase vector consists of a disturbance amplitude component and a direction phase component, which is used to describe the temporal disturbance characteristics of the transportation equipment in the environmental directional entropy field. The method for performing joint amplitude and direction calculations on the set of state disturbance trajectories includes: standardizing the disturbance amplitude of each trajectory line in the set of state disturbance trajectories; performing vector projection operation on the amplitude sequence and the environmental direction angle data within the same direction interval to calculate the phase angle difference; forming a composite vector by combining the amplitude sequence and the phase angle difference within each preset time window according to the corresponding index; and performing weighted aggregation on the composite vector set of all preset time windows to form a disturbance phase vector set.

[0024] In this embodiment S2, the generation of the disturbance phase vector is not achieved by directly performing vector projection calculations on the amplitude sequence and direction angle, but by processing it through amplitude modulation of the directional unit vector. First, the direction angle corresponding to each direction interval is converted into a directional unit vector to characterize the spatial action direction of the environmental disturbance in that direction. Then, the amplitude of each effective disturbance point in the state disturbance trajectory set is used as a modulation coefficient to act on the directional unit vector, so that the amplitude reflects the disturbance intensity and the directional unit vector reflects the direction of the disturbance action. The above modulation results form a disturbance direction response feature vector, and a disturbance phase vector set is constructed in chronological order to describe the joint evolution characteristics of the disturbance in time and direction. The directional phase component is used to characterize the bias trend of the disturbance in the directional entropy field, and the amplitude component is used to characterize the disturbance intensity. The two are combined in vector form to form the disturbance phase vector, avoiding the model inconsistency problem caused by a single scalar angle directly participating in the vector calculation.

[0025] In this embodiment S2, the rate of change of equipment operating data refers to the speed at which the operating status parameters of the transportation equipment change at adjacent sampling times. It is calculated by taking the difference between the equipment operating parameters at two consecutive sampling times and dividing by the corresponding sampling time interval. The rate of change is used to measure the fluctuation intensity of the equipment status in the time dimension. The difference amplitude between adjacent sampling points is the absolute difference between the equipment operating parameters between each sampling point. It is calculated by taking the absolute value of the difference between each sampling point and its predecessor in the same parameter dimension. It is used to reflect the disturbance amplitude of the equipment status signal in a short time interval. The dynamic threshold is the amplitude limit adaptively determined based on the overall fluctuation level of the disturbance behavior data sequence in each time window. It is calculated by multiplying the moving average of all difference amplitudes in the time window by an adjustment coefficient to obtain the threshold. The adjustment coefficient is an empirical coefficient. The setting of the dynamic threshold is used to ensure that the disturbance detection sensitivity is adaptively adjusted when environmental conditions change.

[0026] In this embodiment S2, the effective disturbance point is a sampling point whose differential amplitude exceeds the dynamic threshold. This sampling point represents the time node when the transportation equipment undergoes a significant state change due to directional environmental disturbance within the current time window. When matching the equipment state parameters of the effective disturbance point with the directional intervals under the same time window in the environmental directional entropy field, the system determines the corresponding time window based on the sampling time index of the effective disturbance point, calculates the directional interval to which it belongs based on the spatial coordinates of the point and the environmental directional angle, and retrieves the entropy density values ​​of the same interval in the environmental directional entropy field to establish a one-to-one correspondence and obtain the matching result. The matching result includes four components: time index, directional interval number, equipment state parameter value, and corresponding directional entropy density value, which are used to describe the correspondence between equipment state disturbance and environmental directional characteristics.

[0027] In this embodiment S2, when the directional interval numbers of multiple consecutive valid disturbance points remain consistent or change continuously in adjacent directional intervals, the system determines that the sequence of disturbance points is a directional consistent sequence and connects them in the order of sampling time to form a directional disturbance trajectory line. The directional disturbance trajectory line refers to a discrete curve in the environmental directional entropy field with time as the horizontal axis, directional interval as the vertical axis, and equipment state change as the trajectory path, which is used to represent the disturbance evolution process of transportation equipment under the influence of a specific directional environment. By repeating the above operation in all time windows, a state disturbance trajectory set composed of multiple directional disturbance trajectory lines is obtained. The state disturbance trajectory set is also used for subsequent disturbance phase vector calculation.

[0028] In this embodiment S2, during the process of generating a set of disturbance phase vectors based on the set of state disturbance trajectories, each disturbance phase vector consists of a disturbance amplitude component and a directional phase component. The disturbance amplitude component is the standardized value of the disturbance amplitude in each time window of the corresponding trajectory line. Its calculation method is to divide the disturbance amplitude of each directional disturbance trajectory line by the maximum amplitude of the same trajectory line in the entire sampling period to ensure that the amplitude component is normalized in the [0,1] interval. The directional phase component is the phase offset between the disturbance direction angle and the prevailing wind direction angle in the same time window. Its calculation method is to calculate the vector angle between the two direction angles and take its sign. The positive and negative signs are used to distinguish the offset direction. The temporal disturbance characteristics of the transportation equipment in the environmental directional entropy field include three dimensions: the trend of disturbance amplitude change, the directional phase drift rate, and the concentration of disturbance energy, which correspond to the equipment response intensity, the disturbance phase evolution speed, and the aggregation of disturbance energy in the directional space, respectively.

[0029] In this embodiment S2, the disturbance amplitude of each trajectory line in the state disturbance trajectory set is composed of the amplitude components of all effective disturbance points on the trajectory line. During calculation, the amplitude components of each trajectory line are formed into an amplitude sequence according to the time index. When performing vector projection operation on the amplitude sequence and the environmental orientation angle data, the amplitude sequence is the amplitude vector of continuous time points in the trajectory line, and the environmental orientation angle data is the orientation angle value of the corresponding orientation interval in the environmental directional entropy field. The projection operation is to calculate the cosine of the angle between the amplitude vector and the orientation angle vector in the two-dimensional coordinate system and multiply it by the amplitude magnitude to obtain the amplitude vector. Phase angle difference; the phase angle difference is used to characterize the degree of phase shift of the device's disturbance response direction relative to the dominant direction of the environment; the weighted aggregation operation of the composite vector set of all preset time windows can be achieved by weighted averaging or weighted superposition, where the weight is determined by the variance of the disturbance amplitude in each time window, and the larger the variance, the higher the weight; the result of weighted aggregation is a weighted comprehensive vector set of disturbance amplitude and phase shift in all time windows, which finally forms a disturbance phase vector set, used for the input analysis of the subsequent information resonance identification model and the calculation of resonance disturbance node identification.

[0030] In this embodiment S2, the dynamic threshold is not set using a fixed empirical coefficient, but rather by statistically analyzing the distribution of disturbance intensity changes within the time window and combining it with historical data to form an adaptive update mechanism. First, the differential amplitude of equipment state parameters within multiple operating cycles is statistically modeled, and its mean and fluctuation range are calculated to form an initial reference value. Then, a sliding window method is used to calculate the average value of all differential amplitudes within the current time window in real time, and an adjustment coefficient is determined by combining it with the fluctuation range of historical data. This adjustment coefficient can be automatically updated through data training to ensure the adaptability and repeatability of the threshold. When mapping effective disturbance points to the environmental directional entropy field, the system uses the equipment running direction vector as the spatial mapping reference. The track walking equipment in the coal port area usually runs along the track centerline direction, and the system uses the track centerline direction as the equipment running direction reference, combining it with the real-time travel direction angle of the equipment to form a mapping parameter. The equipment state parameters of the effective disturbance points are associated with this running direction parameter and mapped to the corresponding direction interval in the environmental directional entropy field. This mapping process is based only on the overall running direction information of the equipment and does not depend on the specific component positions of the equipment, thereby avoiding the uncertainty brought about by changes in the structural dimensions of large equipment to the direction mapping.

[0031] In this embodiment S3, the information resonance identification model is a statistical inference model based on the perturbation phase vector set. By establishing the resonance energy distribution matrix of the perturbation phase components in different directions, it is used to identify the resonance perturbation relationship that satisfies the continuous offset resonance condition in the time series. The method for combining and calculating perturbations in all directions of the perturbation phase vector set using the information resonance identification model includes: establishing a time-series vector set for each directional component of the perturbation phase vector set, calculating the cross-correlation coefficient between adjacent directional components; when the cross-correlation coefficient exceeds a preset correlation threshold, determining that the directional pair is a resonance directional pair; performing energy spectrum analysis on all resonance directional pairs, calculating the phase resonance energy value and forming a resonance energy distribution matrix, and performing eigenvalue decomposition on the resonance energy distribution matrix to extract the main eigenvectors as feature representations of the resonance center; and forming a resonance perturbation node set with the resonance energy peak position and its corresponding directional pair as nodes.

[0032] In this embodiment S3, the information resonance identification model is an unsupervised statistical inference model constructed based on the perturbation phase vector set. It does not rely on the model training process or parameter optimization process, nor does it involve the convergence control of traditional machine learning models. This model identifies resonance behavior from the directional cooperative change characteristics by statistically analyzing the phase change trends of perturbations in different directions within multiple time windows in the perturbation phase vector set. It belongs to the diagnostic discrimination model based on perturbation cooperativeness. To solve the problem that the cross-correlation coefficient can only reflect linear correlation characteristics, this embodiment does not use the cross-correlation coefficient as the sole criterion for resonance determination. Instead, the following combined judgment mechanism is used to determine the resonance node: First, the cross-correlation coefficients between the directional components in the perturbation phase vector set are calculated to identify initial candidate direction pairs with a consistent directional trend. Perturbation energy spectrum enhancement analysis is then performed on the candidate direction pairs, using the spatial distribution of perturbation energy as the screening criterion for resonance determination, thereby reducing misjudgments caused by a single calculation indicator. The preset correlation threshold is determined based on historical port operation data and environmental change experience calibration or an adaptive method, and is used to limit perturbation direction pairs that exhibit a clear trend of coordinated response under typical directional corrosion scenarios. If both the cross-correlation coefficient and the energy enhancement index meet the threshold conditions, the direction pair is determined as the input object for the resonance perturbation node.

[0033] In this embodiment S3, the information resonance recognition model is a statistical inference tool without training parameters, implemented based on a statistical modeling method using cross-correlation matrix and energy spectrum analysis. During model construction, the directional components of the perturbation phase vector set are first used as input variables, organized into a multi-dimensional time series set. Preprocessing operations are performed on the set, including removing trend terms, performing normalization, and interpolating to compensate for missing points, to ensure consistency in numerical scale and temporal continuity of each directional component. The system calculates the cross-correlation coefficients between each directional component in the multi-dimensional time series set, forming a cross-correlation matrix to describe the correlation between perturbation components of different directions. Each element of the cross-correlation matrix represents the synchronization strength of a pair of directional components during time variation. The system performs threshold filtering on the cross-correlation matrix, retaining only directional pairs with correlation coefficients reaching a preset threshold. This threshold is determined through statistical analysis of historical perturbation data to ensure that only significantly correlated directional relationships are retained. The set of filtered directional pairs is... A preliminary set of resonance direction pairs is used in the subsequent energy spectrum analysis stage. During this stage, the system transforms the time series of each direction pair from the time domain to the frequency domain, calculating the distribution of its signal energy along the frequency dimension. By identifying the main energy peaks in the energy spectrum, the dominant resonance frequency and corresponding energy intensity of the direction pair are determined. The energy peaks of all direction pairs constitute an energy distribution matrix, with direction pairs as row and column indices and energy intensities as matrix elements, used to characterize the energy coupling relationship between direction pairs. The system performs eigenvalue decomposition analysis on the energy distribution matrix to extract the dominant patterns of energy distribution in the direction space. During the analysis, the system identifies the feature vector that best explains the overall energy distribution characteristics and defines this vector as the main feature vector. The set of directions corresponding to the main feature vector is the resonance center direction set, used to characterize the dominant directions of energy concentration and resonance under directional perturbation. The system stores this main feature vector and its corresponding energy amplitude information as the core feature structure of the information resonance identification model, thereby constructing the static benchmark of the model.

[0034] In this embodiment S3, during the operation phase of the information resonance identification model, the system receives a new set of disturbance phase vectors as input, repeatedly calculates the cross-correlation matrix and energy distribution matrix, and performs similarity analysis between the calculation results and the baseline feature structure. When the similarity between the new energy distribution direction and the main feature vector exceeds a preset threshold, the system determines that a new resonance disturbance node has appeared in the current disturbance environment and outputs the direction index and corresponding time window identifier of the node. The information resonance identification model reflects the interdependence between directions through the cross-correlation matrix and reveals the aggregation pattern of disturbance energy in different directions in the frequency domain through energy spectrum analysis, thereby realizing the identification and quantification of multi-directional disturbance resonance behavior.

[0035] In this embodiment S3, when the information resonance identification model is used to combine and calculate the perturbations in all directions in the perturbation phase vector set, the cross-correlation coefficient refers to the measure of the linear correlation between the phase components of perturbations in different directions, which is used to characterize the similarity strength between directional perturbations. Its value range is [-1,1]. The preset correlation threshold is an empirical threshold obtained by statistically analyzing historical perturbation data during the training phase. Its determination method is to calculate the 95th percentile value of the distribution of all cross-correlation coefficients or the stable upper bound value calculated based on the maximum likelihood estimation.

[0036] In this embodiment S3, when performing energy spectrum analysis on all resonance direction pairs, the system uses the time-series data of the perturbation phase vector as input and employs frequency domain transformation technology to calculate the distribution of signal energy in the frequency space. Energy spectrum analysis can also be achieved through techniques such as Fast Fourier Transform, Short-Time Fourier Transform, Continuous Wavelet Transform, or Power Spectral Density Estimation to reveal the concentrated characteristics of phase resonance energy in different frequency ranges. The selection of the energy spectrum analysis method is based on the characteristics of the concentrated perturbation amplitude change of the perturbation phase vector. When the perturbation signal exhibits characteristics of frequency stability and strong periodicity, Fast Fourier Transform can be used to identify spectral abrupt changes. When the perturbation signal has time-dependent enhancement characteristics but the overall frequency distribution is relatively stable, Short-Time Fourier Transform can be used. The method involves extracting energy change patterns using a windowed approach. When the disturbance signal exhibits multi-scale time response and directional change coupling phenomena, continuous wavelet transform can be used to achieve cross-scale energy concentration trend analysis. In scenarios where the disturbance signal has strong random fluctuations or a low signal-to-noise ratio, power spectral density estimation can be used to identify energy segments. All of the above methods can be adjusted by setting parameters such as window width, scale factor, or frequency band interval. The parameter settings are adaptively adjusted based on the statistical characteristics of historical disturbance response patterns during the operation of port equipment and the intensity of current environmental disturbances. This method is not used to establish a predictive model but for diagnostic feature extraction. Therefore, it is not required to uniformly select a fixed algorithm but to adopt an adaptive selection strategy based on disturbance characteristics, which only plays an enhancement analysis role in the phase resonance judgment process.

[0037] In this embodiment S3, the resonance energy distribution matrix is ​​a two-dimensional numerical matrix with resonance direction pairs as row and column indices and the energy spectrum peak values ​​of the corresponding direction pairs as matrix elements. It is used to characterize the resonance energy distribution relationship between each direction pair. The resonance energy distribution matrix is ​​obtained by calculating the main peak value of the energy spectrum for all identified resonance direction pairs and filling it into the corresponding matrix position. The rows of the matrix represent the source direction components, and the columns represent the response direction components. When performing eigenvalue decomposition on the resonance energy distribution matrix, the main eigenvector is obtained by finding the eigenvector corresponding to the largest eigenvalue. The main eigenvector is the eigenvector that maximizes the variance explained by the matrix. It represents the main distribution direction of the resonance energy in the direction space and is used as the feature representation of the resonance center. The feature representation of the resonance center is the main eigenvector because the vector corresponds to the main direction of energy distribution in the matrix. It can comprehensively describe the main resonance mode and energy concentration trend between different directional disturbances and provide the initial energy center coordinates for subsequent foldback inference.

[0038] In this embodiment S3, the continuous offset resonance condition refers to the condition that the phase center of the resonant perturbation node continuously shifts in the directional angle dimension within adjacent time windows and the shift amplitude exceeds the preset minimum phase drift threshold. This condition is used to determine whether the resonant perturbation node is in a continuous phase shift state during the time evolution process. The back-reflection inference is a spatial back-tracking calculation process based on the resonant perturbation node that satisfies the continuous offset resonance condition. It is used to determine the back propagation path of the resonant energy in the environmental directional entropy field and its energy back-reflection center.

[0039] In this embodiment S3, the determination of the phase center of the resonant disturbance node in the continuous offset resonance condition is based on the weighted average of the disturbance phase vector set in the corresponding directional angle dimension. The weighting factor is the disturbance amplitude corresponding to the disturbance phase vector component, which is used to reflect the dominant role of the strong disturbance point on the directional offset trend. The minimum phase drift threshold is based on the rate of change of the disturbance phase in the structurally stable stage in the historical operation data of coal port transportation equipment. First, the cumulative rate of change of the directional angle of the disturbance phase vector set is statistically analyzed within the preset historical time window, and its mean and standard deviation range are calculated. Then, the standard deviation is adaptively filtered according to the preset quantile interval to form a dynamic threshold for judging the continuous offset trend. When the phase center maintains a unidirectional cumulative offset in the directional angle dimension in at least two or more adjacent time windows, and the offset amplitude of adjacent windows exceeds the above minimum phase drift threshold, it is determined that the node meets the continuous offset resonance condition. This continuous offset is used to characterize the gradual accumulation trend of disturbance energy in the equipment structure response under the action of directional environment.

[0040] In this embodiment S3, during the judgment process of continuous offset resonance conditions, the condition that the phase center of the resonant disturbance node continuously shifts unidirectionally in the directional angle dimension within adjacent time windows and the shift amplitude exceeds the preset minimum phase drift threshold refers to the state in which the phase center position of the same resonant disturbance node continuously moves along a single directional angle change direction within adjacent time windows. Here, the phase center of the resonant disturbance node refers to the mean point of the phase distribution of the directional disturbance corresponding to the node within the time window, and the directional angle dimension refers to the angle coordinate axis defined by the prevailing wind direction in the port area spatial direction coordinate system. Unidirectional shift refers to the situation where the phase center continuously moves in the same direction without reverse change within consecutive time windows, and the shift amplitude is the angular difference between the phase centers of two adjacent time windows. When the shift amplitude exceeds the preset minimum phase drift threshold, it indicates that the phase evolution of the node exhibits a stable unidirectional shift characteristic, satisfying the continuous offset resonance condition. The preset minimum phase drift threshold can be adaptively calculated based on the mean and standard deviation of the phase drift rate distribution in historical data, and usually the mean plus one standard deviation is taken as the judgment boundary.

[0041] In this embodiment S3, when performing backtracking inference on resonant perturbation nodes that satisfy the continuous offset resonance condition, the system executes the following calculation steps: First, the resonant perturbation nodes within the continuous time window are arranged in chronological order to construct a phase trajectory curve; second, the first derivative of the phase trajectory curve is calculated to obtain the phase drift rate. When the sign of the phase drift rate is reversed, it is marked as the backtracking start point, indicating that the phase change trend has changed from positive to negative; with the backtracking start point as the reference, the resonant perturbation nodes within the previous time window are tracked in reverse, the phase drift amount of each time window is accumulated, and the energy decay rate is calculated; when the energy decay rate reaches the minimum value, the energy backtracking center position is determined, which is the energy reverse aggregation point; the backtracking rate is calculated with the energy backtracking center position as the center, which is used to characterize the degree of reverse aggregation of resonant energy. The higher the backtracking rate value, the more significant the energy aggregation trend in this direction interval.

[0042] In this embodiment S3, the asymmetric stability index is a scalar quantitative index calculated based on the combination of the reversion rate and the phase drift rate. It is used to characterize the stability state of the transportation equipment under directional disturbances. The value range of the asymmetric stability index is a positive real number. The higher the asymmetric stability index, the greater the degree of imbalance in the equipment's operating state under directional disturbances.

[0043] In this embodiment S3, the asymmetric stability index is a scalar quantitative index obtained by composite calculation of the reversion rate and the phase drift rate, used to characterize the operational stability of transportation equipment under directional disturbances. Among them, the reversion rate is a quantitative result of the energy convergence intensity, reflecting the degree of reverse convergence of directional disturbance energy in space; the phase drift rate is the rate at which the phase center changes with time, reflecting the dynamic change speed of the equipment response; the composite calculation is a calculation method that normalizes two different physical quantities and then performs functional coupling according to a specified weight. Its implementation technology can adopt a linear weighted function, a product coupling function, or a comprehensive evaluation function based on information entropy weighting.

[0044] In this embodiment S4, the specific method for constructing a perturbation network topology based on the asymmetric stability index is as follows: extract the asymmetric stability index from the resonant perturbation node and use it as input; calculate the synchronicity of the changes in the asymmetric stability index between any two resonant perturbation nodes; when the synchronicity exceeds a preset association threshold, establish a directed edge between the corresponding resonant perturbation nodes, and determine the direction of the edge according to the phase difference of the changes in the asymmetric stability index of the resonant perturbation node; the weight of each directed edge is determined by the relative ratio of the rates of change of the asymmetric stability index between the two resonant perturbation nodes; and a perturbation network topology containing multiple layers of directed edges is formed.

[0045] In this embodiment S4, the sorting and simplification of the perturbation network topology specifically involves: removing low-weight edges, merging edges in the same direction, and normalizing edge weights to generate a standardized perturbation network topology that can be used for propagation path calculation.

[0046] In this embodiment S4, the disturbance propagation path refers to the shortest directed path sequence connecting the source node and the target node in the disturbance network, which is used to represent the propagation path of directional disturbances between different spatial nodes; the information impedance of the disturbance propagation path is a metric that characterizes the difficulty of the disturbance behavior being transmitted on the disturbance propagation path; The method for calculating the information impedance of the disturbance propagation path is as follows: the system takes the reciprocal of the weight of each directed edge on the disturbance propagation path and accumulates them in the order of the path to obtain the information impedance value of the disturbance propagation path; when the path contains multiple intermediate nodes, the system normalizes the rate of change of the asymmetric stability index of each node in the path, and uses the normalization rate as the impedance correction factor to correct the information impedance value to obtain the information impedance of the disturbance propagation path.

[0047] In this embodiment S4, after obtaining the information impedance of the disturbance propagation path, the path corresponding to the minimum information impedance value is taken as the dominant disturbance propagation path, and the direction of this path is defined as the priority direction of disturbance propagation. The dominant propagation path reflects the optimal transmission path of the disturbance signal between nodes under the current directional disturbance conditions. When the minimum information impedance value of the dominant propagation path is in a long-term downward trend, it indicates that the disturbance energy is transmitted smoothly in the network and the equipment structure is in a stable operating state. When the information impedance of the dominant propagation path shows a continuous increase or a sudden increase, it indicates that there is an abnormal stagnation or reverse aggregation phenomenon in the transmission of disturbance energy, and the system determines that there is a potential risk of instability in the equipment in this area. The system further outputs safety control signals according to the location and direction of the abnormal path, including reducing the operating speed of the equipment in that direction, limiting the drive load, or triggering a local maintenance command. When the information impedance change trends of multiple propagation paths all show a synchronous increase, the system integrates the results into a global risk indicator and outputs the safety warning status of the overall equipment group.

[0048] In this embodiment S4, the synchronicity between any two resonant disturbance nodes is determined by the correlation characteristics of the changes in the asymmetric stability index within the same time window. Specifically, the correlation coefficient between the incremental sequences of the asymmetric stability index of each resonant disturbance node within a unit time window is calculated, or a dynamic time warping algorithm independent of sampling synchronicity is used to extract the consistency index of the changing trend. Based on the correlation characteristics of the node relationships corresponding to the confirmed instability events in the historical operation phase of the port area, and combined with the preset abnormal data identification window, an adaptive threshold optimization screening method is performed on the correlation characteristics to determine the association threshold. When determining the direction of the directed edges of the network, the time lag of the changes in the asymmetric stability index between different resonant disturbance nodes is used as the judgment basis. The network topology simplification process includes removing weakly correlated edges whose weighted correlation values ​​between nodes are lower than the upper bound of the historical noise fluctuation range, performing path merging on multiple edges with consistent directions and continuous phase evolution trends, and determining the representative path based on the principle of minimum information impedance. The weights of all retained edges are normalized to make the weights of each path in the network have a unified comparison benchmark, thereby forming a disturbance network structure for fault diagnosis and analysis.

[0049] In this embodiment S4, the weights of the directed edges in the disturbance propagation path are used to characterize the information response strength and directional coupling degree during disturbance transmission. The larger the value, the higher the information transmission efficiency and the smaller the resistance. Therefore, when calculating the information impedance, the weights of each directed edge in the path are counted inversely and then accumulated in the order of the path. The accumulated result is used to measure the degree of information obstruction when the disturbance is transmitted along the path. The information impedance value is a reverse characteristic quantity of disturbance propagation, used to reflect the resistance accumulation of the disturbance in the structural response. The larger the value, the more suppressed the information transmission is, and the smaller the value, the smoother the information transmission is. In this embodiment, the reciprocal of the edge weights is used to represent the resistance index instead of the reciprocal of the transmission efficiency when calculating the information impedance. The calculation logic is used to diagnose the phase disturbance propagation trend rather than directly measure the energy transmission amount. Therefore, the path impedance is only used to assess the risk of directional adaptation of disturbance transmission and not as an indicator of disturbance scale.

[0050] In this embodiment S4, the rate of change of the asymmetric stability index of each resonant disturbance node in the path within the corresponding time window is normalized and the normalized rate is used as the impedance correction factor. The correction process adopts a multiplicative superposition method, that is, the information impedance accumulation result is multiplied by the correction factor formed based on the node phase evolution rate, so that the resistance index can reflect the dual comprehensive effect of disturbance transmission resistance and disturbance growth intensity. If the disturbance growth rate increases, even if the impedance value is low, the risk will increase due to the effect of the correction factor, thereby avoiding misjudging the operational safety situation by relying solely on transmission smoothness. This processing method meets the diagnostic calculation requirements and does not involve predictive mechanical modeling.

[0051] In this embodiment S4, the information impedance change trend is used to determine the disturbance transmission structure and the degree of dynamic coupling. The impedance change trend and the disturbance scale factor are analyzed simultaneously. When the path impedance value continues to decrease and the growth rate of the corresponding node's asymmetric stability index is in a low range, it is diagnosed that the structure is in an adaptive operation stage that can withstand disturbances. Conversely, when the impedance continues to decrease but the growth rate of the asymmetric stability index exceeds the historical risk threshold, it is identified as a high flow risk state and an abnormal diagnosis output is triggered. The threshold is determined by statistical analysis of fault scenarios that have occurred in the long-term operation data of the port area. By setting the fault evolution stage identification conditions, a qualitative-quantitative mapping relationship between disturbance impedance and equipment safety is established to prevent the information impedance trend from misjudging the actual equipment operation risk.

[0052] Example 2: The present invention proposes a safety diagnostic system for coal port transportation equipment, which is applied to the safety diagnostic method for coal port transportation equipment proposed in Example 1. It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the safety diagnostic method for coal port transportation equipment in Example 1.

[0053] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A method for safety diagnosis of transportation equipment in coal port areas, characterized in that, Includes the following steps: S1. Obtain environmental data and transportation equipment data of the coal port area, perform time-series alignment according to the preset time window and generate a disturbance behavior data sequence, and construct an environmental directional entropy field based on the disturbance behavior data sequence; The environmental directional entropy field is used to describe the spatial distribution characteristics of port area environmental data. S2. Discretize the transportation equipment data according to a preset time window and perform corresponding mapping processing in the environmental directional entropy field to obtain a set of state disturbance trajectories. Then, perform joint calculations on the amplitude and direction of the set of state disturbance trajectories to generate a set of disturbance phase vectors of the transportation equipment in the environmental directional entropy field. The state disturbance trajectory set is a set of trajectories formed by mapping equipment operation data in the environmental directional entropy field. The disturbance phase vector set is generated based on the state disturbance trajectory set and is used to describe the temporal disturbance characteristics of the transportation equipment operation state under the influence of the directional environment. S3. Use the information resonance identification model to combine and calculate the disturbances in all directions in the disturbance phase vector set to obtain the resonant disturbance node set. Perform back-folding inference on the resonant disturbance nodes that satisfy the continuous offset resonance condition to calculate the asymmetric stability index. The information resonance identification model is used to identify resonant perturbation nodes that satisfy the continuous offset resonance condition. S4. Construct a disturbance network topology based on the asymmetric stability index, sort and simplify the disturbance network topology, determine the disturbance propagation path and the information impedance of the disturbance propagation path, and diagnose the safety of transportation equipment based on the information impedance of the disturbance propagation path. The disturbance network topology is a directed weighted network constructed with resonant disturbance nodes as nodes and the information transmission relationship between resonant disturbance nodes as edges. It is used to characterize the transmission structure and mutual influence relationship of different directional disturbance behaviors in the time and space dimensions.

2. The method for safety diagnosis of coal port transportation equipment according to claim 1, characterized in that: In S1, the disturbance behavior data sequence is a time-series multi-source data set formed by synchronizing port area environmental data and transportation equipment data within a preset time window, used to represent the impact of port area environmental disturbance behavior on the operating status of transportation equipment; the environmental directional entropy field is a spatial information entropy distribution field calculated based on the distribution probability of port area environmental data in the disturbance behavior data sequence, with spatial direction angle as the independent variable. The specific method for constructing an environmental directional entropy field based on disturbance behavior data sequences includes: dividing the port area environmental data within each preset time window into spatial directional partitions, calculating the information entropy density value corresponding to each directional partition, and using the directional interval corresponding to the prevailing wind direction as a reference axis, normalizing and interpolating the information entropy density values ​​of all directional intervals through the directional entropy density function to form a continuous environmental directional entropy field.

3. The method for safety diagnosis of coal port transportation equipment according to claim 2, characterized in that: In S2, the state disturbance trajectory set is a time-series trajectory set formed by discretizing the transportation equipment data within a preset time window and mapping it according to the directional partitioning result of the environmental directional entropy field. Each time-series trajectory in the time-series trajectory set consists of a time index, directional coordinates and corresponding equipment state value, which is used to record the discrete response process of the transportation equipment's operating state in different directional entropy density intervals. The method for discretizing transportation equipment data according to a preset time window and performing corresponding mapping processing in the environmental directional entropy field includes: within each preset time window, calculating the difference amplitude of adjacent sampling points based on the rate of change of equipment operating data; when the difference amplitude exceeds a dynamic threshold, determining the sampling point as an effective disturbance point; matching the equipment state parameters of the effective disturbance point with the directional interval under the same time window in the environmental directional entropy field; connecting the effective disturbance points in chronological order according to the matching results to form a directional disturbance trajectory line; and aggregating all directional disturbance trajectory lines according to the preset time window number and directional partition number to generate a state disturbance trajectory set.

4. The method for safety diagnosis of coal port transportation equipment according to claim 3, characterized in that: In S2, the disturbance phase vector set is a composite vector set generated based on the state disturbance trajectory set. Each disturbance phase vector consists of a disturbance amplitude component and a direction phase component, which are used to describe the temporal disturbance characteristics of the transportation equipment in the environmental directional entropy field. The method for performing joint amplitude and direction calculations on the set of state disturbance trajectories includes: standardizing the disturbance amplitude of each trajectory line in the set of state disturbance trajectories; Within the same directional interval, a vector projection operation is performed on the amplitude sequence and the ambient azimuth angle data to calculate the phase angle difference; the amplitude sequence and phase angle difference within each preset time window are combined into a composite vector according to their corresponding indices; weighted aggregation is performed on the composite vector set of all preset time windows to form a perturbation phase vector set.

5. The method for safety diagnosis of coal port transportation equipment according to claim 4, characterized in that: In S3, the information resonance identification model is a statistical reasoning model based on the perturbation phase vector set. By establishing the resonance energy distribution matrix of the perturbation phase components in different directions, it is used to identify the resonance perturbation relationship that satisfies the continuous offset resonance condition in the time series. The method for combining and calculating perturbations in all directions of the perturbation phase vector set using the information resonance identification model includes: establishing a time-series vector set for each directional component of the perturbation phase vector set, calculating the cross-correlation coefficient between adjacent directional components; when the cross-correlation coefficient exceeds a preset correlation threshold, determining that the directional pair is a resonance directional pair; performing energy spectrum analysis on all resonance directional pairs, calculating the phase resonance energy value and forming a resonance energy distribution matrix, and performing eigenvalue decomposition on the resonance energy distribution matrix to extract the main eigenvectors as feature representations of the resonance center; and forming a resonance perturbation node set with the resonance energy peak position and its corresponding directional pair as nodes.

6. The method for safety diagnosis of coal port transportation equipment according to claim 5, characterized in that: In S3, the continuous offset resonance condition refers to the condition that the phase center of the resonant perturbation node continuously shifts in the directional angle dimension within adjacent time windows and the shift amplitude exceeds the preset minimum phase drift threshold. It is used to determine whether the resonant perturbation node is in a continuous phase shift state during the time evolution process. The back-reflection inference is a spatial back-tracking calculation process based on the resonant perturbation node that satisfies the continuous offset resonance condition. It is used to determine the back propagation path of the resonant energy in the environmental directional entropy field and its energy back-reflection center.

7. The method for safety diagnosis of coal port transportation equipment according to claim 6, characterized in that: In S3, the asymmetric stability index is a scalar quantitative index calculated based on the combination of the reversion rate and the phase drift rate. It is used to characterize the stability state of the transportation equipment under directional disturbances. The asymmetric stability index is a positive real number. The higher the asymmetric stability index, the greater the imbalance of the equipment's operating state under directional disturbances.

8. The method for safety diagnosis of coal port transportation equipment according to claim 7, characterized in that: In step S4, the specific method for constructing the perturbation network topology based on the asymmetric stability index is as follows: extract the asymmetric stability index from the resonant perturbation node and use it as input; calculate the synchronicity of the changes in the asymmetric stability index between any two resonant perturbation nodes; when the synchronicity exceeds a preset association threshold, establish a directed edge between the corresponding resonant perturbation nodes, and determine the direction of the edge according to the phase difference of the changes in the asymmetric stability index of the resonant perturbation node; the weight of each directed edge is determined by the relative ratio of the change rates of the asymmetric stability index between the two resonant perturbation nodes; thus forming a perturbation network topology containing multiple layers of directed edges.

9. A method for safety diagnosis of coal port transportation equipment according to claim 8, characterized in that: In S4, the disturbance propagation path refers to the shortest directed path sequence connecting the source node and the target node in the disturbance network, which is used to represent the propagation path of directional disturbances between different spatial nodes; the information impedance of the disturbance propagation path is a measure of the difficulty of the disturbance behavior being transmitted on the disturbance propagation path. The method for calculating the information impedance of the disturbance propagation path is as follows: the system takes the reciprocal of the weight of each directed edge on the disturbance propagation path and accumulates them in the order of the path to obtain the information impedance value of the disturbance propagation path; when the path contains multiple intermediate nodes, the system normalizes the rate of change of the asymmetric stability index of each node in the path, and uses the normalization rate as the impedance correction factor to correct the information impedance value to obtain the information impedance of the disturbance propagation path.

10. A safety diagnostic system for coal port transportation equipment, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes a computer program to implement the safety diagnosis method for coal port transportation equipment as described in any one of claims 1-9.

Images