Multi-channel radar aging monitoring method, system, equipment and medium

Abstract

The invention relates to the technical field of radars. By providing a multi-channel radar aging monitoring method, system and device and a medium, the method comprises the steps of performing sharing processing on aging monitoring data acquired by each channel by constructing a cooperative management architecture of a multi-channel radar to obtain multi-channel shared data; performing space-time sequence extraction processing on the aging parameters in the multi-channel shared data to obtain space-time sequence data of the multi-channel aging parameters for correlation modeling; performing correlation analysis processing on the aging parameters of the spatial dimension and the time dimension to obtain an inter-channel aging state correlation model; performing collaborative recognition processing on the global feature vectors of the multi-channel aging parameters to obtain an aging mode recognition result; and the response type of the cross-channel aging influence is judged and processed to obtain decision support information, so that the multi-channel data collaborative analysis capability is improved, the missed judgment risk of the equipment composite failure mode is reduced, and the fault repair period is shortened.

Description

Technical Field

[0001] This invention relates to the field of radar technology, and in particular to multi-channel radar aging monitoring methods, systems, equipment, and media. Background Technology

[0002] With the large-scale deployment of multi-sensor fusion and intelligent monitoring technologies, multi-channel radar systems are undertaking core sensing functions in scenarios such as security monitoring, industrial equipment status perception, low-altitude security, and smart homes. The long-term operational stability and aging status monitoring of radar systems are directly related to system reliability and represent a key technical challenge that continues to be addressed in the field of multi-radar collaborative applications.

[0003] Traditional technologies employ isolated single-channel monitoring to independently analyze radar aging parameters. However, this approach only outputs channel-level aging status and cannot achieve collaborative sharing and correlation analysis of multi-channel data. This results in a lack of ability to identify systemic risks such as "single-channel mutations triggering multi-channel cascading aging." Existing solutions primarily focus on monitoring the instantaneous values ​​of aging parameters in the current cycle, lacking time-series comparison technology for historical aging trends. This makes it difficult to capture time-series composite failure modes where chronic degradation and acute failure alternate. Furthermore, when multi-channel radars experience correlated aging, traditional technologies fail to construct inter-channel correlation models, and response strategies rely excessively on manual experience, leading to delayed maintenance decisions and long fault repair cycles. Summary of the Invention

[0004] Therefore, it is necessary to provide multi-channel radar aging monitoring methods, systems, equipment, and media to address the aforementioned technical issues, thereby enhancing the collaborative analysis capabilities of multi-channel data, reducing the risk of missed detection of composite failure modes of equipment, shortening the fault repair cycle, and ultimately improving the long-term reliability of radar systems.

[0005] In a first aspect, this application provides a multi-channel radar aging monitoring method, which includes:

[0006] By constructing a collaborative management architecture for multi-channel radar, the aging monitoring data collected from each channel is shared and processed to obtain multi-channel shared data.

[0007] Spatiotemporal sequence extraction processing is performed on the aging parameters in the multi-channel shared data to obtain the spatiotemporal sequence data of the multi-channel aging parameters for correlation modeling;

[0008] Based on the spatiotemporal sequence data of multi-channel aging parameters, correlation analysis is performed on the aging parameters in the spatial and temporal dimensions to obtain the correlation model of aging state between channels.

[0009] By utilizing the inter-channel aging state correlation model, the global feature vectors of multi-channel aging parameters are collaboratively identified to obtain aging pattern recognition results.

[0010] Based on the aging pattern recognition results, the response type of cross-channel aging impact is judged and processed to obtain decision support information.

[0011] In one embodiment, an inter-channel aging state correlation model is used to collaboratively identify the global feature vectors of multi-channel aging parameters to obtain aging pattern identification results, including:

[0012] Based on the inter-channel aging state correlation model, global feature aggregation is performed on the global feature vectors of multi-channel aging parameters to obtain a global feature vector set of multi-channel aging.

[0013] The global feature vector set of multi-channel aging is matched with the preset aging mode library to obtain the current aging mode matching result.

[0014] Based on the current aging pattern matching results, multi-channel aging impact prediction processing is performed to obtain aging pattern identification results.

[0015] In one embodiment, based on the inter-channel aging state correlation model, global feature aggregation processing is performed on the global feature vectors of multi-channel aging parameters to obtain a global feature vector set for multi-channel aging, including:

[0016] Based on the inter-channel correlation weight matrix in the inter-channel aging state association model, weight mapping is performed on each feature dimension of the global feature vector of multi-channel aging parameters to obtain the weight-adjusted global feature vector.

[0017] The following formula is used to perform multi-dimensional feature fusion on the weighted global feature vector, jointly aggregating the spatiotemporal features of each channel on the time and space axes to obtain the aggregated global feature vector set:

[0018]

[0019] Among them, F global Let C represent the aggregated global feature vector set, and ω represent the total number of channels. c Let represent the weight adjustment coefficient for the c-th channel, α represent the aggregated weight along the time axis, β represent the aggregated weight along the spatial axis, T represent the length of the time dimension, and S represent the length of the spatial dimension. This represents the feature vector of the c-th channel at time t. This represents the feature vector of the c-th channel at spatial position s;

[0020] The aggregated global feature vector set is subjected to normalization constraint processing to obtain the global feature vector set for multi-channel aging.

[0021] In one embodiment, the global feature vector set of multi-channel aging is compared with a preset aging pattern library using similarity matching to obtain the current aging pattern matching result, including:

[0022] The global feature vector set of multi-channel aging is subjected to feature dimensionality reduction processing to obtain a dimensionality-reduced subset of global feature vectors;

[0023] The subset of global feature vectors after dimensionality reduction is compared with the feature vectors of each mode in the preset aging mode library to perform preliminary similarity calculation and obtain a set of candidate aging modes.

[0024] Based on the inter-channel weight matrix of the inter-channel aging state association model, the candidate aging mode set is subjected to weighted similarity optimization processing to obtain the optimized similarity matching result.

[0025] Based on the optimized similarity matching results, the current aging mode matching result is determined.

[0026] In one embodiment, for spatiotemporal sequence data based on multi-channel aging parameters, correlation analysis is performed on the aging parameters in the spatial and temporal dimensions to obtain an inter-channel aging state correlation model, including:

[0027] Spatial dimension correlation analysis was performed on the spatiotemporal sequence data of multi-channel aging parameters to obtain the correlation index of aging parameters between adjacent channels;

[0028] The spatiotemporal sequence data of multi-channel aging parameters are compared and processed in terms of time dimension trend to obtain the deviation index of aging trend of each channel.

[0029] A correlation model of aging status between channels is constructed based on the correlation index of aging parameters between adjacent channels and the deviation index of aging trends of each channel.

[0030] In one embodiment, the spatiotemporal sequence data of multi-channel aging parameters are subjected to time-dimensional trend comparison processing to obtain the deviation index of the aging trend of each channel, including:

[0031] From the spatiotemporal sequence data of multi-channel aging parameters, extract the time series subset of aging parameters for each channel within a preset historical period to obtain the historical aging parameter time series dataset for each channel.

[0032] Trend feature extraction processing is performed on the historical aging parameter time series dataset of each channel to obtain the historical aging trend feature sequence of each channel;

[0033] The time series data of aging parameters of each channel in the current period are compared with the historical aging trend feature sequence of the corresponding channel to calculate the deviation, and the time series deviation value set of each channel is obtained.

[0034] The deviation degree is quantified based on the time deviation value set of each channel to obtain the deviation index of the aging trend of each channel.

[0035] In one embodiment, based on the aging pattern recognition results, the response type of cross-channel aging impact is judged and processed to obtain decision support information, including:

[0036] The aging pattern recognition results are processed to classify the aging patterns, resulting in the classification types of the aging patterns.

[0037] Based on the aging mode's hierarchical type, a set of corresponding response triggering conditions is obtained by matching the preset cross-channel response rule base.

[0038] Based on the set of response trigger conditions, determine the response type for cross-channel aging;

[0039] Integrate response types and corresponding handling recommendations to generate decision support information.

[0040] Secondly, this application also provides a multi-channel radar aging monitoring system, which includes:

[0041] The data sharing module is used to share and process the aging monitoring data collected from each channel by building a collaborative management architecture for multi-channel radar, so as to obtain multi-channel shared data.

[0042] The spatiotemporal extraction module is used to perform spatiotemporal sequence extraction processing on aging parameters in multi-channel shared data to obtain spatiotemporal sequence data of multi-channel aging parameters for correlation modeling.

[0043] The correlation modeling module is used to perform correlation analysis on the aging parameters in the spatial and temporal dimensions based on the spatiotemporal sequence data of multi-channel aging parameters, and obtain the correlation model of aging state between channels.

[0044] The pattern recognition module is used to collaboratively identify the global feature vectors of aging parameters across multiple channels using an inter-channel aging state correlation model, thereby obtaining aging pattern recognition results.

[0045] The response decision module is used to judge and process the response type of cross-channel aging impact based on the aging pattern recognition results, and obtain decision support information.

[0046] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods in the first aspect of this application.

[0047] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods in the first aspect of this application.

[0048] The multi-channel radar aging monitoring method, system, equipment, and medium provided in this application achieve data sharing among aging monitoring channels by constructing a collaborative management architecture, breaking down the barriers of isolated single-channel data and laying the foundation for collaborative analysis of multi-channel data, effectively improving the collaborative analysis capability of multi-channel data. Based on this, spatiotemporal sequence extraction of aging parameters from the shared data is performed, and correlation analysis is conducted using spatiotemporal dimensions to construct an inter-channel aging state correlation model. This model can comprehensively capture the aging characteristics of equipment in time and space, thereby reducing the risk of missing composite failure modes. Furthermore, the aforementioned correlation model is used to collaboratively identify global feature vectors to obtain aging mode identification results, and based on this, the response type of cross-channel aging impact is determined, and decision support information is generated, which can quickly provide a basis for maintenance decisions and shorten the fault repair cycle. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart of a multi-channel radar aging monitoring method in one embodiment of the present invention;

[0051] Figure 2 This is a flowchart illustrating how, in one embodiment of the present invention, a global feature vector set of multi-channel aging is matched with a preset aging mode library to obtain the current aging mode matching result.

[0052] Figure 3 This is a structural diagram of a multi-channel radar aging monitoring system according to one embodiment of the present invention. Detailed Implementation

[0053] To make the above-mentioned objects, features, and advantages of this application more apparent and understandable, the specific embodiments of this application will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0054] First, the application scenarios of the embodiments of this application are described. The embodiments of this application provide a multi-channel radar aging monitoring method, system, device, and medium applicable to, but not limited to, long-term operational status monitoring of multi-channel radar systems. For example, in security monitoring scenarios, multi-channel radar needs to continuously monitor specific areas around the clock, and the aging status of each channel directly affects monitoring accuracy and system stability. This method can achieve coordinated monitoring of the aging status of each channel. In industrial equipment status sensing scenarios, multi-channel radar is used to sense the operating status of production line equipment in real time. Effective monitoring of the radar's own aging status can ensure the reliability of equipment status sensing data.

[0055] For example, the multi-channel radar aging monitoring method, system, device and medium provided in the embodiments of this application can also be applied to other application scenarios. This is only an example and does not limit the specific application scenarios.

[0056] like Figure 1 As shown, this application provides a multi-channel radar aging monitoring method, which includes:

[0057] S101: By constructing a collaborative management architecture for multi-channel radar, the aging monitoring data collected from each channel is shared and processed to obtain multi-channel shared data.

[0058] For example, a unified collaborative management architecture is planned and built for each independent channel of a multi-channel radar system. This architecture is compatible with the data formats of each channel and includes mechanisms for data transmission, aggregation, and access control to ensure the standardization and security of data flow. Based on this architecture, the aging monitoring data acquisition process for each channel begins, collecting various parameter information related to the radar's aging status.

[0059] The architecture's built-in data sharing mechanism standardizes the raw aging monitoring data collected from each channel, eliminating data format differences between channels, and then aggregates and integrates the data according to preset rules. Through this series of processes, multi-channel shared data that can be interoperated and used across multiple channels is obtained.

[0060] S102: Perform spatiotemporal sequence extraction processing on the aging parameters in the multi-channel shared data to obtain spatiotemporal sequence data of the multi-channel aging parameters for correlation modeling.

[0061] For example, various aging parameters directly related to the radar's aging state are filtered from multi-channel shared data. These parameters include, but are not limited to, signal attenuation, noise interference, and frequency stability indicators. For the time dimension, aging parameters for each channel are continuously extracted at preset time intervals to generate parameter sequences for each channel at different time points, thereby reflecting the dynamic change trend of the parameters over time.

[0062] Meanwhile, for the spatial dimension, the aging parameters of different channels at the same time point are extracted accordingly, and the correlation sequence of parameters of each channel in the spatial distribution is constructed to reflect the parameter distribution characteristics between channels.

[0063] The sequences extracted from the time dimension are correlated and integrated with those extracted from the spatial dimension to ensure accurate correspondence between time nodes and channel positions. After this processing, spatiotemporal sequence data of multi-channel aging parameters for correlation modeling are obtained.

[0064] S103: Based on the spatiotemporal sequence data of multi-channel aging parameters, correlation analysis is performed on the aging parameters in the spatial and temporal dimensions to obtain the correlation model of aging state between channels.

[0065] For example, based on spatiotemporal sequence data of multi-channel aging parameters, for the spatial dimension, the aging parameters of each channel at the same time node are selected, and the synchronicity, correlation and mutual influence of parameter changes between different channels are analyzed, and indicators that can reflect the spatial correlation characteristics between channels are extracted.

[0066] Meanwhile, focusing on the time dimension, aging parameter sequences of the same channel at different time periods are extracted, and the evolution trend, fluctuation pattern and degree of deviation from normal state of the parameters over time are compared and analyzed to generate indicators characterizing the time evolution characteristics of each channel.

[0067] By integrating spatial correlation indicators with temporal trend indicators, a mapping relationship between the two is established, clarifying the influence weight of different spatiotemporal feature combinations on the correlation of channel aging states. Through the structured integration and modeling of the above correlation relationships, an inter-channel aging state correlation model is obtained.

[0068] S104: Using the inter-channel aging state correlation model, the global feature vectors of multi-channel aging parameters are collaboratively identified to obtain the aging pattern recognition result.

[0069] For example, a global feature vector covering key aging characteristics of each channel is extracted from the multi-channel aging parameters. These vectors need to contain typical aging attributes in the spatiotemporal dimensions. Based on the channel association rules and influence weights defined by the inter-channel aging state association model, cross-channel collaborative analysis is performed on the global feature vector, placing the features of each channel in the overall association framework to analyze the interaction and linkage patterns between features.

[0070] In the above process, combined with the established aging state association patterns in the model, pattern matching is performed on the feature vectors after collaborative analysis to determine their degree of fit with known aging patterns, and the dominant features and association patterns exhibited by the current multi-channel aging are identified. Through the above systematic collaborative identification process, the aging pattern identification results are obtained.

[0071] S105: Based on the aging pattern recognition results, the response type of cross-channel aging impact is judged and processed to obtain decision support information.

[0072] For example, based on the aging pattern recognition results, the scope, degree, and correlation characteristics of the current cross-channel aging impact are clarified, including key information such as the number of affected channels and the diffusion trend of the aging pattern. Referring to the preset response rule system, which includes the handling standards and priority classifications corresponding to different aging patterns, the nature of the cross-channel aging impact is determined, identifying whether it belongs to a response type such as early warning, local maintenance, system adjustment, or emergency repair.

[0073] In the above process, by combining the correlation characteristics of aging patterns, the applicable conditions and potential impacts of each response type are analyzed, and the optimal response strategy is selected. Then, the results of the response type determination and corresponding implementation suggestions are integrated to form structured information containing the direction of action, key operational points, and resource allocation references, thus obtaining decision support information.

[0074] The multi-channel radar aging monitoring method provided in one embodiment of this application realizes the sharing of aging monitoring data of each channel by constructing a collaborative management architecture, breaking the barrier of isolated single-channel data, laying the foundation for collaborative analysis of multi-channel data, and effectively improving the collaborative analysis capability of multi-channel data. On this basis, the aging parameters in the shared data are extracted in a spatiotemporal sequence, and correlation analysis is performed in combination with the spatiotemporal dimension to construct an aging state correlation model between channels. This can capture the aging characteristics of the equipment in time and space more comprehensively, thereby reducing the risk of missing the detection of composite failure modes of the equipment. Furthermore, the above correlation model is used to collaboratively identify the global feature vector to obtain the aging mode identification result, and based on this, the response type of cross-channel aging impact is determined and decision support information is generated, which can quickly provide a basis for maintenance decisions and shorten the fault repair cycle.

[0075] In one embodiment, an inter-channel aging state correlation model is used to collaboratively identify the global feature vectors of multi-channel aging parameters to obtain aging pattern identification results, including:

[0076] (1) Based on the inter-channel aging state correlation model, the global feature vectors of the multi-channel aging parameters are processed by global feature aggregation to obtain the global feature vector set of multi-channel aging.

[0077] For example, based on the aging state association model between channels, the weight ratio and association rules of each channel in the global features are clarified. According to the above rules, the global feature vector of the multi-channel aging parameters is integrated and the features are enhanced. The scattered single-channel features are incorporated into the overall association framework to form a set that can reflect the collaborative aging characteristics of multiple channels, and the global feature vector set of multi-channel aging is obtained.

[0078] (2) Perform similarity matching processing between the global feature vector set of multi-channel aging and the preset aging mode library to obtain the current aging mode matching result.

[0079] For example, based on the global feature vector set of multi-channel aging, a preset aging mode library is invoked. This library contains feature vectors of various typical aging modes and their corresponding mode attributes. By comparing the similarity between the global feature vector set and the feature vectors of each mode in the library, several modes with the highest matching degree are selected as candidates to obtain the current aging mode matching result.

[0080] (3) Perform multi-channel aging impact prediction processing based on the current aging pattern matching results to obtain aging pattern identification results.

[0081] For example, by combining the current aging pattern matching results and referring to the channel influence transmission path and aging diffusion law defined in the inter-channel aging state correlation model, the channel range that the matching pattern may affect, the chain reaction it causes, and the development trend of the aging degree are analyzed. The subsequent impact of multi-channel aging is prospectively judged, and the aging pattern identification results are obtained.

[0082] In one embodiment, based on the inter-channel aging state correlation model, global feature aggregation processing is performed on the global feature vectors of multi-channel aging parameters to obtain a global feature vector set for multi-channel aging, including:

[0083] (1) Based on the inter-channel correlation weight matrix in the inter-channel aging state correlation model, weight mapping is performed on each feature dimension of the global feature vector of the multi-channel aging parameters to obtain the weight-adjusted global feature vector.

[0084] For example, an inter-channel correlation weight matrix is ​​extracted from the inter-channel aging state correlation model. This matrix contains information on the correlation strength between the feature dimensions of each channel. Based on the weight values ​​of the corresponding dimensions in the matrix, the values ​​of each feature dimension in the global feature vector of the multi-channel aging parameters are adjusted to make the feature vector more closely reflect the actual correlation between channels, resulting in a weight-adjusted global feature vector.

[0085] (2) Using the following formula, perform multi-dimensional feature fusion on the weighted global feature vector, and jointly aggregate the spatiotemporal features of each channel on the time axis and the spatial axis to obtain the aggregated global feature vector set:

[0086]

[0087] Among them, F global Let C represent the aggregated global feature vector set, and ω represent the total number of channels. c Let represent the weight adjustment coefficient for the c-th channel, α represent the aggregated weight along the time axis, β represent the aggregated weight along the spatial axis, T represent the length of the time dimension, and S represent the length of the spatial dimension. This represents the feature vector of the c-th channel at time t. Let represent the feature vector of the c-th channel at spatial position s.

[0088] For example, based on the weighted global feature vectors, and combined with the preset aggregation rules of the time axis and spatial axis, the feature vectors of each channel at different time points and the feature vectors at different spatial locations are jointly integrated. During the integration process, according to the weight adjustment coefficients of each channel and the aggregation weights of the time axis and spatial axis, the spatiotemporal features are comprehensively calculated and merged to form a set covering multi-dimensional related information, resulting in the aggregated global feature vector set.

[0089] (3) Normalize the aggregated global feature vector set to obtain the global feature vector set of multi-channel aging.

[0090] For example, for the aggregated global feature vector set, a standardization process is adopted to constrain the numerical range of each feature vector in the set to a unified range, eliminate the influence of differences in the units of different features, ensure the consistency and comparability of feature vectors, and obtain a multi-channel aged global feature vector set.

[0091] like Figure 2 As shown, the global feature vector set of multi-channel aging is compared with a preset aging pattern library using similarity matching to obtain the current aging pattern matching result, including:

[0092] S201: Perform feature dimensionality reduction on the global feature vector set of multi-channel aging to obtain a dimensionality-reduced subset of global feature vectors.

[0093] For example, the global feature vector set of multi-channel aging is simplified by retaining key features and removing redundant information, thereby compressing the dimensionality of the feature vectors and obtaining a more dimensionally reduced subset of global feature vectors while maintaining the expressive power of the core features.

[0094] S202: Perform preliminary similarity calculation on the subset of global feature vectors after dimensionality reduction and the feature vectors of each mode in the preset aging mode library to obtain a set of candidate aging modes.

[0095] For example, using a subset of the dimensionality-reduced global feature vectors as a benchmark, a pre-defined aging mode library is invoked. This library stores standard feature vectors corresponding to various known aging modes. By calculating the similarity between the subset and the feature vectors of each mode in the library, several modes with high similarity are selected to generate a set of candidate aging modes for further selection.

[0096] S203: Based on the inter-channel weight matrix of the inter-channel aging state association model, the candidate aging mode set is subjected to weighted similarity optimization processing to obtain the optimized similarity matching result.

[0097] For example, an inter-channel weight matrix is ​​introduced into the inter-channel aging state association model. This matrix reflects the influence weight of different channels in the aging association. Based on the weight ratio of each channel in the matrix, the similarity values ​​of each mode in the candidate aging mode set are weighted and adjusted to strengthen the influence of key channel association features on the similarity results, thus obtaining optimized similarity matching results.

[0098] S204: Determine the current aging mode matching result based on the optimized similarity matching result.

[0099] For example, based on the optimized similarity matching results, according to the preset matching threshold and priority rules, the mode with the highest matching degree and that meets the judgment conditions is selected from the optimized results as the matching mode corresponding to the current multi-channel aging state, and the matching result of the current aging mode is determined.

[0100] In one embodiment, for spatiotemporal sequence data based on multi-channel aging parameters, correlation analysis is performed on the aging parameters in the spatial and temporal dimensions to obtain an inter-channel aging state correlation model, including:

[0101] (1) Spatial dimension correlation analysis was performed on the spatiotemporal sequence data of multi-channel aging parameters to obtain the correlation index of aging parameters between adjacent channels.

[0102] For example, for spatiotemporal sequence data of multi-channel aging parameters, focusing on the spatial dimension, channel groups located in adjacent positions or functionally related are selected, and the aging parameters of each group at the same time node are extracted. By analyzing the synchronicity, amplitude correlation, and mutual influence of parameter changes, the degree of correlation between the aging states of different channels is quantitatively characterized, and the correlation index of aging parameters between adjacent channels is obtained.

[0103] (2) Perform time dimension trend comparison processing on the spatiotemporal sequence data of multi-channel aging parameters to obtain the deviation index of aging trend of each channel.

[0104] For example, for spatiotemporal sequence data of multi-channel aging parameters, the focus shifts to the time dimension, extracting the aging parameter sequence of each channel within continuous historical periods, and fitting it to generate a benchmark trend curve reflecting its normal aging process. The parameter sequence of the current period is compared with the benchmark trend curve, and the degree of difference between the two in terms of change rate, fluctuation range, etc., is calculated to quantify the degree to which each channel deviates from the normal aging trajectory, thus obtaining the deviation index of the aging trend of each channel.

[0105] (3) Based on the correlation index of aging parameters between adjacent channels and the deviation index of aging trend of each channel, a correlation model of aging status between channels is constructed.

[0106] For example, the correlation index of aging parameters between adjacent channels and the deviation index of aging trends of each channel are integrated to establish a correlation mapping relationship between the two. By defining the index weight allocation rules and correlation logic, spatial correlation features and temporal trend features are incorporated into a unified model framework to form a structured expression that can comprehensively reflect the correlation law of aging state of multiple channels in the spatiotemporal dimension, and to construct an aging state correlation model between channels.

[0107] In one embodiment, the spatiotemporal sequence data of multi-channel aging parameters are subjected to time-dimensional trend comparison processing to obtain the deviation index of the aging trend of each channel, including:

[0108] (1) Extract the time series subset of aging parameters of each channel within a preset historical period from the spatiotemporal sequence data of the multi-channel aging parameters to obtain the historical aging parameter time series dataset of each channel.

[0109] For example, from the spatiotemporal sequence data of multi-channel aging parameters, according to the preset historical period range, all aging parameter records of each channel within the period are selected and sorted into independent sequence subsets in chronological order, thereby summarizing to obtain the historical aging parameter time series dataset of each channel.

[0110] (2) The historical aging parameter time series dataset of each channel is processed to extract trend features to obtain the historical aging trend feature sequence of each channel.

[0111] For example, for the historical aging parameter time series dataset of each channel, by extracting the key trend features such as the long-term change direction, fluctuation frequency and amplitude change patterns contained in the data, and arranging the above features in sequence according to time nodes, a typical trajectory that can reflect the historical aging process of each channel is constructed, and the historical aging trend feature sequence of each channel is obtained.

[0112] (3) Perform deviation calculation on the time series data of aging parameters of each channel in the current period and the historical aging trend feature sequence of the corresponding channel to obtain the set of time series deviation values ​​of each channel.

[0113] For example, the aging parameter time series data of each channel in the current cycle is collected, and compared with the historical aging trend feature sequence of the corresponding channel at each time node. The differences between the two in terms of parameter values, rate of change, etc. are calculated, and the above difference values ​​are summarized to form a set, thus obtaining the time series deviation value set of each channel.

[0114] (4) Based on the time deviation value set of each channel, the deviation degree is quantified to obtain the deviation index of the aging trend of each channel.

[0115] For example, based on the time deviation value set of each channel, a standardized quantification method is adopted to comprehensively consider the magnitude, duration and cumulative effect of the deviation value, and these deviation information are transformed into indicators that can intuitively reflect the degree of deviation from the historical trend, thus obtaining the deviation index of the aging trend of each channel.

[0116] In one embodiment, based on the aging pattern recognition results, the response type of cross-channel aging impact is judged and processed to obtain decision support information, including:

[0117] (1) Perform pattern type classification processing on the aging pattern recognition results to obtain the classification type of the aging pattern.

[0118] For example, the aging pattern recognition results are classified according to the number of channels involved, the aging diffusion rate and the scope of influence, and different levels of pattern types such as mild, moderate and severe are distinguished based on the preset classification criteria to obtain the aging pattern classification type.

[0119] (2) Based on the aging mode classification type, match the preset cross-channel response rule library to obtain the corresponding response trigger condition set.

[0120] For example, using the aging mode classification as the retrieval basis, a preset cross-channel response rule library is invoked. This library stores response trigger conditions corresponding to different levels of aging modes. Through matching and filtering, all trigger conditions that match the current classification are extracted, forming a set of associated conditions, thus obtaining the corresponding response trigger condition set.

[0121] (3) Determine the response type for cross-channel aging based on the set of response triggering conditions.

[0122] For example, the specific requirements of each condition in the response trigger condition set are analyzed, and combined with the actual operating status of the current multi-channel radar system, it is determined whether each condition is met. Then, the most suitable handling category is selected from the preset response type system to determine the response type for cross-channel aging.

[0123] (4) Integrate response types and corresponding handling suggestions to generate decision support information.

[0124] For example, the determined response type is integrated with the corresponding standard handling procedures, key points of operation, and resource allocation suggestions to form structured content that includes action direction, implementation steps, and precautions, thereby generating decision support information.

[0125] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0126] In one embodiment, such as Figure 3 As shown, this application also provides a multi-channel radar aging monitoring system 300, which includes:

[0127] The data sharing module 301 is used to share and process the aging monitoring data collected by each channel by constructing a collaborative management architecture for multi-channel radar, so as to obtain multi-channel shared data.

[0128] The spatiotemporal extraction module 302 is used to perform spatiotemporal sequence extraction processing on aging parameters in multi-channel shared data to obtain spatiotemporal sequence data of multi-channel aging parameters for correlation modeling.

[0129] The correlation modeling module 303 is used to perform correlation analysis on the aging parameters in the spatial and temporal dimensions based on the spatiotemporal sequence data of multi-channel aging parameters, and obtain the correlation model of aging state between channels.

[0130] The pattern recognition module 304 is used to perform collaborative recognition processing on the global feature vectors of multi-channel aging parameters using the inter-channel aging state correlation model to obtain aging pattern recognition results.

[0131] The response decision module 305 is used to judge and process the response type of cross-channel aging impact based on the aging pattern recognition results, and obtain decision support information.

[0132] Specifically, the data sharing module 301 establishes a collaborative management architecture for multi-channel radar. This architecture is compatible with the data formats of each channel and has data transmission, aggregation, and access control mechanisms. Based on this, the raw aging monitoring data collected by each channel is standardized, and after eliminating format differences, it is aggregated and integrated to obtain multi-channel shared data.

[0133] The spatiotemporal extraction module 302 determines the parameter range reflecting the aging characteristics of radar channels based on multi-channel shared data, extracts the time change sequence of each channel parameter according to the set time period, and simultaneously extracts the spatial distribution sequence of all channel parameters at the same time node. After matching and integrating to ensure accurate spatiotemporal correspondence, the spatiotemporal sequence data of multi-channel aging parameters used for correlation modeling is obtained.

[0134] The correlation modeling module 303 is based on the spatiotemporal sequence data of multi-channel aging parameters. It analyzes the synchronicity and correlation of parameter changes in different channels at the same time node in the spatial dimension and extracts spatial correlation indicators. In the time dimension, it extracts the historical periodic parameter sequence of each channel, compares the current and historical trends to obtain time evolution indicators, integrates the two types of indicators and establishes a mapping relationship to construct an aging state correlation model between channels.

[0135] The pattern recognition module 304 uses the inter-channel aging state association model to extract global feature vectors covering the key spatiotemporal aging attributes of each channel. It relies on the association rules and weighting mechanism in the model to perform cross-channel collaborative analysis, and combines known aging pattern features for matching to identify the dominant features and association patterns of current multi-channel aging, thus obtaining the aging pattern recognition results.

[0136] Based on the aging pattern identification results, the response decision module 305 clarifies the scope, degree, and correlation characteristics of the cross-channel aging impact, determines the response type with reference to the preset response rule system, analyzes the applicable conditions of each type to select the optimal strategy, integrates the response type determination results and implementation suggestions to form structured information and obtain decision support information.

[0137] Pattern recognition module 304 is also used for:

[0138] Based on the inter-channel aging state correlation model, global feature aggregation is performed on the global feature vectors of multi-channel aging parameters to obtain a global feature vector set of multi-channel aging.

[0139] The global feature vector set of multi-channel aging is matched with the preset aging mode library to obtain the current aging mode matching result.

[0140] Based on the current aging pattern matching results, multi-channel aging impact prediction processing is performed to obtain aging pattern identification results.

[0141] Pattern recognition module 304 is also used for:

[0142] Based on the inter-channel correlation weight matrix in the inter-channel aging state association model, weight mapping is performed on each feature dimension of the global feature vector of multi-channel aging parameters to obtain the weight-adjusted global feature vector.

[0143] The following formula is used to perform multi-dimensional feature fusion on the weighted global feature vector, jointly aggregating the spatiotemporal features of each channel on the time and space axes to obtain the aggregated global feature vector set:

[0144]

[0145] Among them, F global Let C represent the aggregated global feature vector set, and ω represent the total number of channels. c Let represent the weight adjustment coefficient for the c-th channel, α represent the aggregated weight along the time axis, β represent the aggregated weight along the spatial axis, T represent the length of the time dimension, and S represent the length of the spatial dimension. This represents the feature vector of the c-th channel at time t. This represents the feature vector of the c-th channel at spatial position s;

[0146] The aggregated global feature vector set is subjected to normalization constraint processing to obtain the global feature vector set for multi-channel aging.

[0147] Pattern recognition module 304 is also used for:

[0148] The global feature vector set of multi-channel aging is subjected to feature dimensionality reduction processing to obtain a dimensionality-reduced subset of global feature vectors;

[0149] The subset of global feature vectors after dimensionality reduction is compared with the feature vectors of each mode in the preset aging mode library to perform preliminary similarity calculation and obtain a set of candidate aging modes.

[0150] Based on the inter-channel weight matrix of the inter-channel aging state association model, the candidate aging mode set is subjected to weighted similarity optimization processing to obtain the optimized similarity matching result.

[0151] Based on the optimized similarity matching results, the current aging mode matching result is determined.

[0152] The association modeling module 303 is also used for:

[0153] Spatial dimension correlation analysis was performed on the spatiotemporal sequence data of multi-channel aging parameters to obtain the correlation index of aging parameters between adjacent channels;

[0154] The spatiotemporal sequence data of multi-channel aging parameters are compared and processed in terms of time dimension trend to obtain the deviation index of aging trend of each channel.

[0155] A correlation model of aging status between channels is constructed based on the correlation index of aging parameters between adjacent channels and the deviation index of aging trends of each channel.

[0156] The association modeling module 303 is also used for:

[0157] From the spatiotemporal sequence data of multi-channel aging parameters, extract the time series subset of aging parameters for each channel within a preset historical period to obtain the historical aging parameter time series dataset for each channel.

[0158] Trend feature extraction processing is performed on the historical aging parameter time series dataset of each channel to obtain the historical aging trend feature sequence of each channel;

[0159] The time series data of aging parameters of each channel in the current period are compared with the historical aging trend feature sequence of the corresponding channel to calculate the deviation, and the time series deviation value set of each channel is obtained.

[0160] The deviation degree is quantified based on the time deviation value set of each channel to obtain the deviation index of the aging trend of each channel.

[0161] The response decision module 305 is also used for:

[0162] The aging pattern recognition results are processed to classify the aging patterns, resulting in the classification types of the aging patterns.

[0163] Based on the aging mode's hierarchical type, a set of corresponding response triggering conditions is obtained by matching the preset cross-channel response rule base.

[0164] Based on the set of response trigger conditions, determine the response type for cross-channel aging;

[0165] Integrate response types and corresponding handling recommendations to generate decision support information.

[0166] In one embodiment, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0167] In one embodiment, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0168] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0169] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A multi-channel radar aging monitoring method, characterized in that, The method includes: By constructing a collaborative management architecture for multi-channel radar, the aging monitoring data collected from each channel is shared and processed to obtain multi-channel shared data. The aging parameters in the multi-channel shared data are subjected to spatiotemporal sequence extraction processing to obtain spatiotemporal sequence data of multi-channel aging parameters for correlation modeling; Based on the spatiotemporal sequence data of the multi-channel aging parameters, correlation analysis is performed on the aging parameters in the spatial and temporal dimensions to obtain the correlation model of aging states between channels. Using the inter-channel aging state correlation model, the global feature vectors of the multi-channel aging parameters are collaboratively identified to obtain aging pattern identification results. Based on the aging pattern recognition results, the response type of cross-channel aging impact is judged and processed to obtain decision support information.

2. The multi-channel radar aging monitoring method according to claim 1, characterized in that, The step of using the inter-channel aging state correlation model to perform collaborative recognition processing on the global feature vectors of the multi-channel aging parameters to obtain aging pattern recognition results includes: Based on the inter-channel aging state correlation model, global feature aggregation processing is performed on the global feature vectors of the multi-channel aging parameters to obtain a global feature vector set for multi-channel aging. The global feature vector set of the multi-channel aging is matched with the preset aging mode library to obtain the current aging mode matching result. Based on the current aging pattern matching results, multi-channel aging impact prediction processing is performed to obtain the aging pattern identification results.

3. The multi-channel radar aging monitoring method according to claim 2, characterized in that, The global feature vector of the multi-channel aging parameters is aggregated based on the inter-channel aging state correlation model to obtain a global feature vector set for multi-channel aging, including: Based on the inter-channel correlation weight matrix in the inter-channel aging state association model, weight mapping is performed on each feature dimension of the global feature vector of the multi-channel aging parameters to obtain the weight-adjusted global feature vector. Using the following formula, a multi-dimensional feature fusion operation is performed on the weighted global feature vector to jointly aggregate the spatiotemporal features of each channel on the time axis and the spatial axis, resulting in an aggregated global feature vector set: Among them, F global Let C represent the aggregated global feature vector set, and ω represent the total number of channels. c Let represent the weight adjustment coefficient for the c-th channel, α represent the aggregated weight along the time axis, β represent the aggregated weight along the spatial axis, T represent the length of the time dimension, and S represent the length of the spatial dimension. This represents the feature vector of the c-th channel at time t. This represents the feature vector of the c-th channel at spatial position s; The aggregated global feature vector set is subjected to normalization constraint processing to obtain the global feature vector set of the multi-channel aging.

4. The multi-channel radar aging monitoring method according to claim 2, characterized in that, The step of performing similarity matching processing between the global feature vector set of the multi-channel aging and the preset aging pattern library to obtain the current aging pattern matching result includes: The global feature vector set of the multi-channel aging is subjected to feature dimensionality reduction processing to obtain a dimensionality-reduced subset of global feature vectors. The subset of global feature vectors after dimensionality reduction is compared with the feature vectors of each mode in the preset aging mode library to perform preliminary similarity calculation, thereby obtaining a set of candidate aging modes. Based on the inter-channel weight matrix of the inter-channel aging state association model, the candidate aging mode set is subjected to weighted similarity optimization processing to obtain the optimized similarity matching result. Based on the optimized similarity matching results, the current aging mode matching result is determined.

5. The multi-channel radar aging monitoring method according to claim 1, characterized in that, The process of performing correlation analysis on the spatiotemporal sequence data based on the multi-channel aging parameters to obtain an inter-channel aging state correlation model includes: Spatial dimension correlation analysis is performed on the spatiotemporal sequence data of the multi-channel aging parameters to obtain the correlation index of aging parameters between adjacent channels; The spatiotemporal sequence data of the multi-channel aging parameters are subjected to time dimension trend comparison processing to obtain the deviation index of the aging trend of each channel. Based on the correlation index of aging parameters between adjacent channels and the deviation index of aging trends of each channel, an aging state correlation model between the channels is constructed.

6. The multi-channel radar aging monitoring method according to claim 5, characterized in that, The process of comparing the spatiotemporal sequence data of the multi-channel aging parameters in terms of time dimension trends yields deviation indices for the aging trends of each channel, including: From the spatiotemporal sequence data of the multi-channel aging parameters, extract the time-series subset of the aging parameters of each channel within a preset historical period to obtain the historical aging parameter time-series dataset of each channel. The historical aging parameter time series datasets of each channel are processed to extract trend features, resulting in the historical aging trend feature sequence of each channel. The time series data of aging parameters of each channel in the current period are compared with the historical aging trend feature sequence of the corresponding channel to calculate the deviation, and the time series deviation value set of each channel is obtained. The deviation degree is quantified based on the time deviation value set of each channel to obtain the deviation index of the aging trend of each channel.

7. The multi-channel radar aging monitoring method according to claim 1, characterized in that, The step of determining the response type of cross-channel aging impact based on the aging pattern identification result to obtain decision support information includes: The aging pattern recognition results are processed to classify the aging patterns, thereby obtaining the classification types of the aging patterns. Based on the aging mode's classification type, a preset cross-channel response rule base is matched to obtain the corresponding set of response triggering conditions. Based on the set of response triggering conditions, determine the response type for cross-channel aging; The decision support information is generated by integrating the response type with the corresponding handling suggestion information.

8. A multi-channel radar aging monitoring system, characterized in that, The system includes: The data sharing module is used to share and process the aging monitoring data collected from each channel by building a collaborative management architecture for multi-channel radar, so as to obtain multi-channel shared data. The spatiotemporal extraction module is used to perform spatiotemporal sequence extraction processing on the aging parameters in the multi-channel shared data to obtain spatiotemporal sequence data of the multi-channel aging parameters for correlation modeling. The correlation modeling module is used to perform correlation analysis on the aging parameters in the spatial and temporal dimensions based on the spatiotemporal sequence data of the multi-channel aging parameters, and to obtain the correlation model of aging state between channels. The pattern recognition module is used to perform collaborative recognition processing on the global feature vector of the multi-channel aging parameters using the inter-channel aging state association model to obtain the aging pattern recognition result. The response decision module is used to judge and process the response type of cross-channel aging impact based on the aging pattern recognition results, and obtain decision support information.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the multi-channel radar aging monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-channel radar aging monitoring method according to any one of claims 1 to 7.

Images