Special equipment state identification method and system based on electrical parameter mode matching

Abstract

The invention discloses a special equipment state identification method and system based on electrical parameter mode matching, and relates to the technical field of data processing. The method comprises the following steps: performing special equipment state sensitivity verification on multiple monitoring indexes of the power distribution cabinet to obtain a dynamic grading sensitive index set; generating an elevator state mode library; calling multi-stage real-time operation parameters; triggering hierarchical equipment state mapping, and outputting a hierarchical state confidence vector; performing state transition prediction, and outputting a state transition probability matrix; fusing the hierarchical state confidence vector and the state transition probability matrix, and outputting real-time fault probability distribution; and according to the real-time fault probability distribution, generating a multi-stage early warning signal, and triggering a cascade safety protection response. The technical problem that state recognition of special equipment depends on a sensor in the prior art is solved, accurate recognition of the elevator running state based on the electrical parameters is achieved, and therefore the technical effects that dependence on an external sensor is reduced, and the real-time performance and reliability of state monitoring are improved are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a special equipment state identification method and system based on electrical parameter pattern matching. Background Art

[0002] During the operation of special equipment, its safety status monitoring and fault identification are important links to ensure the stable operation of the equipment and the safety of personnel. In existing technologies, the identification of the operating status of elevators mainly relies on multiple types of sensors installed in key parts of the equipment, which collect physical quantities such as vibration, temperature, and displacement for analysis. However, this type of method has the following shortcomings: On the one hand, the installation and maintenance costs of sensors are high, and they are easily affected by factors such as installation location, environmental interference, and the sensor's own performance degradation, making it difficult to ensure the stability and accuracy of the monitoring data; on the other hand, the variation patterns of electrical parameters under different operating conditions have not been fully utilized. Existing solutions lack the ability to recognize patterns based on electrical parameters, making it difficult to continuously achieve accurate status identification and trend prediction when sensors are limited or fail. Summary of the Invention

[0003] The present application provides a special equipment status identification method and system based on electrical parameter pattern matching, which solves the technical problem in the prior art that special equipment status identification relies on sensors.

[0004] In a first aspect of the present application, a method for identifying a special equipment state based on electrical parameter pattern matching is provided, the method comprising: The special equipment status sensitivity of multiple monitoring indicators of the distribution cabinet is verified to obtain a dynamic hierarchical sensitive indicator set; a multi-level status pattern sub-library is pre-built offline based on the dynamic hierarchical sensitive indicator set, and an elevator status pattern library is generated through hierarchical association integration; based on the dynamic hierarchical sensitive indicator set, multi-level real-time operating parameters are retrieved from the distribution cabinet step by step according to the sensitivity priority; the multi-level real-time operating parameters are dynamically loaded into the elevator status pattern library, triggering hierarchical equipment status mapping, and outputting a hierarchical state confidence vector; state transition prediction is performed on the hierarchical state confidence vector, and a state transition probability matrix is ​​output; the hierarchical state confidence vector and the state transition probability matrix are integrated to output a real-time fault probability distribution; a multi-level early warning signal is generated according to the real-time fault probability distribution to trigger a step-by-step safety protection response.

[0005] A second aspect of the present application provides a special equipment status identification system based on electrical parameter pattern matching, the system comprising: Sensitivity verification module: performs special equipment status sensitivity verification on multiple monitoring indicators of the distribution cabinet to obtain a dynamic hierarchical sensitive indicator set; pattern library construction module: pre-builds a multi-level state pattern sub-library offline based on the dynamic hierarchical sensitive indicator set, and generates an elevator state pattern library through hierarchical association integration; parameter retrieval module: retrieves multi-level real-time operating parameters from the distribution cabinet step by step according to the sensitivity priority based on the dynamic hierarchical sensitive indicator set; state mapping module: dynamically loads the multi-level real-time operating parameters into the elevator state pattern library, triggers hierarchical equipment state mapping, and outputs a hierarchical state confidence vector; state transition prediction module: performs state transition prediction on the hierarchical state confidence vector, and outputs a state transition probability matrix; fault probability output module: integrates the hierarchical state confidence vector and the state transition probability matrix, and outputs a real-time fault probability distribution; early warning response module: generates a multi-level early warning signal according to the real-time fault probability distribution, and triggers a step-by-step safety protection response.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: First, the sensitivity of multiple monitoring indicators of the power distribution cabinet to special equipment status is verified to obtain a dynamic hierarchical sensitive indicator set. Next, a multi-level state pattern sub-library is pre-built offline based on the dynamic hierarchical sensitive indicator set, and an elevator state pattern library is generated through hierarchical association and integration. Simultaneously, based on the dynamic hierarchical sensitive indicator set, multiple levels of real-time operating parameters are retrieved from the power distribution cabinet step by step according to sensitivity priority. Furthermore, these multiple levels of real-time operating parameters are dynamically loaded into the elevator state pattern library, triggering hierarchical equipment state mapping and outputting a hierarchical state confidence vector. Next, state transition prediction is performed on the hierarchical state confidence vector, outputting a state transition probability matrix. The hierarchical state confidence vector and the state transition probability matrix are then combined to output a real-time fault probability distribution. Finally, a multi-level warning signal is generated based on the real-time fault probability distribution, triggering a step-by-step safety protection response. This solves the technical problem of special equipment status identification relying on sensors in the prior art, achieving accurate identification of elevator operating status based on electrical parameters, thereby reducing reliance on external sensors and improving the real-time and reliability of state monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0008] Figure 1 A flow chart of a method for identifying the state of special equipment based on electrical parameter pattern matching provided in an embodiment of the present application; Figure 2 A structural diagram of a special equipment status identification system based on electrical parameter pattern matching provided in an embodiment of the present application.

[0009] Explanation of the accompanying symbols: sensitivity verification module 11, pattern library construction module 12, parameter retrieval module 13, state mapping module 14, state transition prediction module 15, fault probability output module 16, early warning response module 17. DETAILED DESCRIPTION

[0010] The present application solves the technical problem in the prior art that special equipment status identification relies on sensors by providing a special equipment status identification method and system based on electrical parameter pattern matching.

[0011] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0012] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0013] Example 1, as Figure 1 As shown, the present application provides a special equipment state identification method based on electrical parameter pattern matching, wherein the method includes: The sensitivity of special equipment status is verified for multiple monitoring indicators of distribution cabinets to obtain a set of dynamic graded sensitive indicators.

[0014] Based on the application environment characteristics and equipment model code of the distribution cabinet, a device fingerprint is constructed, and the full-dimensional operation log data of multiple devices of the same model are called as the retrieval condition. Subsequently, the collected operation logs are aggregated according to the indicator type to form a multi-state time series data set corresponding to multiple monitoring indicators. For the multi-state time series data of each monitoring indicator, the data is divided using the equipment status label to construct a baseline state record set and multiple groups of fault state record sets. By calculating the KL divergence between the baseline state and each fault state record set, the sensitivity score of each monitoring indicator to different fault states is quantified. Combined with the recurrence frequency of the fault state, the sensitivity score of each fault state is weighted and summarized to obtain the comprehensive operation sensitivity of each monitoring indicator. Finally, the monitoring indicators are dynamically graded according to the comprehensive operation sensitivity to form a dynamic graded sensitive indicator set.

[0015] Furthermore, the sensitivity of the special equipment status is verified for multiple monitoring indicators of the distribution cabinet to obtain a dynamic hierarchical sensitive indicator set. The method includes: The application environment characteristics of the distribution cabinet are retrieved, and the application environment characteristics and the distribution cabinet model code are used as device fingerprint code retrieval conditions, and multiple full-dimensional operation logs of the devices are called; the multiple full-dimensional operation logs of the devices are aggregated based on the indicator type to obtain multiple multi-state time series data sets of multiple operation indicators; the state sensitivity of special equipment is quantified based on the multiple multi-state time series data sets, and multiple comprehensive operation sensitivities are output; based on the multiple comprehensive operation sensitivities, the multiple operation indicators are dynamically graded, and the dynamic graded sensitive indicator set is output.

[0016] First, retrieve the application environment characteristics of the target distribution cabinet, including but not limited to the installation location, power supply topology, load type, ambient temperature and humidity range, etc., and obtain the distribution cabinet model code; combine the above application environment characteristics with the model code to generate a unique device fingerprint code, and use it as a retrieval condition to call the full-dimensional operation logs of multiple similar distribution cabinets that match the device fingerprint from the historical operation database. The operation logs include time series records of each monitoring indicator under multiple operating states. Then, aggregate the full-dimensional operation logs of the multiple devices according to the indicator type, classify the monitoring data of the same type into the corresponding operating indicators, and form a multi-state time series data set corresponding to multiple operating indicators, wherein the multiple states include a baseline state and several fault states, and each state corresponds to a monitoring record of one or more time periods. Next, based on the multiple multi-state time series data sets, a quantitative analysis of the state sensitivity of each operating indicator is performed. Specifically, using the indicator distribution characteristics of the baseline state as a reference, the distribution difference measurement value between the indicator in each fault state and the baseline state is calculated. The measurement can use Kullback-Leibler divergence (KL divergence), Jensen-Shannon divergence (JSD), or other statistical distance algorithms. The difference measurement value of each fault state is weighted and summed according to the historical recurrence frequency of the fault state to obtain the comprehensive operation sensitivity of the operating indicator. Finally, all operating indicators are sorted in descending order according to their comprehensive operation sensitivity and dynamically graded according to preset grading thresholds or quantile intervals. Indicators with higher sensitivity are classified as high sensitivity, those with medium sensitivity are classified as medium sensitivity, and those with lower sensitivity are classified as low sensitivity. A dynamically graded sensitive indicator set containing indicator identification, sensitivity score, sensitivity level, and main associated fault types is output for subsequent state model construction and real-time state recognition.

[0017] Furthermore, the method of quantifying the state sensitivity of special equipment based on the multiple multi-state time series data sets and outputting multiple comprehensive operation sensitivities includes: The multiple multi-state time series data sets are decomposed based on the device state labels to obtain multiple baseline state record sets and multiple groups of fault state record sets; based on the multiple baseline state record sets, the multiple groups of fault state record sets are quantified using KL divergence to output multiple groups of fault state KL sensitivity scores; based on the recurrence frequency of M fault states, the multiple groups of fault state KL sensitivity scores are weighted within the group to output the multiple comprehensive operation sensitivities.

[0018] First, using the device state labels corresponding to each multi-state time series dataset, the datasets are decomposed into states. Data in normal operation are divided into a baseline state record set, and data in different fault states are divided into multiple fault state record sets, each corresponding to a fault type. Then, using the multiple baseline state record sets as references, the distribution difference between each fault state record set and the corresponding baseline state record set for each operating indicator is measured. Specifically, probability density estimates are performed on the baseline state record set and the fault state record set to obtain the baseline distribution and fault distribution, and the Kullback–Leibler divergence (KL divergence) is used to calculate the distribution difference between the two to obtain the KL sensitivity score of the operating indicator under the fault state. Next, corresponding weight coefficients are constructed based on the recurrence frequency of M fault states in the historical operating data. The weight coefficients are proportional to the fault recurrence frequency and are normalized. The KL sensitivity scores of each operating indicator under each fault state are weighted and summed according to the corresponding weight coefficients to obtain the comprehensive operational sensitivity of the operating indicator. Finally, the comprehensive operational sensitivity of all operational indicators is output as the input basis for the subsequent generation of dynamic hierarchical sensitive indicator sets.

[0019] A multi-level state pattern sub-library is pre-built offline based on the dynamic hierarchical sensitive indicator set, and an elevator state pattern library is generated through hierarchical association integration.

[0020] Based on a set of dynamically graded sensitive indicators, the sensitive indicators are decomposed according to hierarchical relationships to obtain a multi-level sensitive indicator set. Using the first-level sensitive indicators as screening criteria, multiple sets of corresponding sample fault state time series data are retrieved from the full-dimensional operation log of the equipment. Combined with different fault level annotations, multiple sets of multi-state sample data sets are formed. These sample data are analyzed for indicator fluctuation scales, representative state feature templates are extracted, and a first-level state pattern sub-library is constructed. Subsequently, according to the hierarchical order of sensitive indicators, the second-level and higher-level state pattern sub-libraries are iteratively constructed, and a bidirectional mapping index is established between the multi-level sensitive indicators and the multi-level state pattern sub-libraries. Finally, through hierarchical association and integration, a complete elevator state pattern library is generated, providing a structured pattern foundation for subsequent real-time state matching and identification.

[0021] Furthermore, a multi-level state pattern sub-library is pre-built offline based on the dynamic hierarchical sensitive indicator set, and an elevator state pattern library is generated through hierarchical association integration. The method includes: The dynamic hierarchical sensitive indicator set is decomposed based on the hierarchical relationship to obtain multi-level sensitive indicators; using the specific indicator set of the first-level sensitive indicators as the screening condition, multiple groups of sample fault state time series data sets of multiple sample fault states under multiple groups of sample fault levels are retrieved from the full-dimensional operation logs of the multiple devices; by performing indicator fluctuation scale analysis on the multiple groups of sample fault state time series data sets, multiple groups of state feature templates are output; the multiple sample fault states, multiple groups of sample fault levels and multiple groups of sample state feature templates are hierarchically associated and stored to obtain a first-level state pattern sub-library; a multi-level state pattern sub-library is iteratively constructed in hierarchical order, and a bidirectional mapping relationship index table between the multi-level sensitive indicators and the multi-level state pattern sub-library is constructed to generate the elevator state pattern library.

[0022] The dynamic hierarchical sensitive indicator set is decomposed hierarchically based on sensitivity levels. Each operating indicator is divided into first-level sensitive indicators, second-level sensitive indicators, and so on, and finally, Nth-level sensitive indicators, according to their sensitivity from high to low. Each level of sensitive indicator set is used to construct a corresponding state pattern sub-library. Using specific indicator items in the first-level sensitive indicator set as screening criteria, historical data containing these indicator items is retrieved from the full-dimensional operation logs of multiple devices. During the retrieval process, the data is classified according to multiple preset fault states (such as door machine faults, brake faults, and traction system faults) and multiple sets of fault levels (such as minor, moderate, and severe), resulting in multiple sets of sample fault state time series data sets. Next, these multiple sets of sample fault state time series data sets are subjected to indicator fluctuation scale analysis, including but not limited to calculating mean fluctuation amplitude, standard deviation change rate, peak-to-valley ratio, transient fluctuation energy, and spectral features to capture the statistical and time series characteristics of the sensitive indicators under different states. The analysis results are packaged into multiple sets of state feature templates, each corresponding to a fault state and its level. Subsequently, the aforementioned multiple sample fault states, multiple sets of sample fault levels, and corresponding state feature templates are hierarchically associated and stored to construct a first-level state pattern sub-library. This first-level state pattern sub-library stores state labels, feature templates, and their associations with sensitive indicator items. On this basis, the aforementioned data retrieval, feature template generation, and hierarchical association storage steps are repeated for the second through Nth level sensitive indicators in hierarchical order to iteratively construct a multi-level state pattern sub-library. During this construction process, a bidirectional mapping relationship index table is established between the multi-level sensitive indicators and the corresponding multi-level state pattern sub-libraries. This allows for rapid locating the corresponding pattern sub-library using the sensitive indicators during operation, and conversely, the associated sensitive indicator set can be retrieved using the state pattern sub-library. Ultimately, all multi-level state pattern sub-libraries and the bidirectional mapping relationship index table are uniformly stored to form the elevator state pattern library.

[0023] Based on the dynamic hierarchical sensitive indicator set, multi-level real-time operating parameters are retrieved from the power distribution cabinet step by step according to the sensitivity priority.

[0024] The sensitivity levels of all monitoring indicators in the dynamic hierarchical sensitivity indicator set are obtained and sorted from high to low based on sensitivity to form a sensitivity priority list. Subsequently, the real-time operating parameters of the corresponding indicators are retrieved from the real-time data acquisition system of the power distribution cabinet according to this sensitivity priority list. The real-time data of highly sensitive indicators is first collected as the first-level operating parameters. The real-time data of medium-sensitivity indicators is then retrieved as the second-level operating parameters. This process is repeated step by step, and the real-time operating parameters corresponding to each sensitivity level are retrieved step by step to form a multi-level real-time operating parameter system.

[0025] The multi-level real-time operating parameters are dynamically loaded into the elevator state pattern library, hierarchical device state mapping is triggered, and a hierarchical state confidence vector is output.

[0026] Furthermore, the multi-level real-time operating parameters are dynamically loaded into the elevator state pattern library, hierarchical device state mapping is triggered, and a hierarchical state confidence vector is output. The method includes: The first-level operating parameter sequence of the first-level sensitive indicator is retrieved from the distribution cabinet in real time; the first-level operating parameter sequence is feature extracted to obtain a first-level feature vector; the first-level feature vector is loaded into the first-level state pattern sub-library of the elevator state pattern library, and the initial state mapping set is matched and output; if the initial state mapping set is an empty set, a monitoring cycle of the first-level sensitive indicator is performed until the initial state mapping set is a non-empty set, thereby triggering a hierarchical device state mapping.

[0027] The first-level operating parameter sequence corresponding to the first-level sensitive indicator is retrieved from the real-time data acquisition system of the distribution cabinet. The sequence contains the continuous sampling data of the sensitive indicator within a preset time window; feature extraction is performed on the first-level operating parameter sequence to extract a multidimensional feature vector containing time domain and frequency domain features. Specific features include mean, variance, kurtosis, kurtosis, spectral energy distribution, etc., to form a first-level feature vector; the first-level feature vector is loaded into the first-level state pattern sub-library in the elevator state pattern library, and the initial state mapping set is matched and output by calculating the similarity between the first-level feature vector and each state feature template in the first-level state pattern sub-library; if the initial state mapping set is an empty set, indicating that the current feature vector cannot match any known state template, the system continues to perform a real-time monitoring cycle on the first-level sensitive indicator, repeatedly retrieves the operating parameter sequence and performs feature extraction and matching until the match is successful, and a non-empty initial state mapping set is obtained, triggering subsequent hierarchical equipment state mapping processing.

[0028] Furthermore, the first-level feature vector is loaded into the first-level state pattern sub-library of the elevator state pattern library, and the initial state mapping set is matched and outputted. The method includes: The method comprises the following steps: traversing and calculating the Euclidean distance matrix of multiple groups of sample state feature templates in the first-level feature vector and the first-level state pattern sub-library, and outputting multiple groups of state similarities; retrieving multiple real-time fault levels from the multiple groups of sample fault levels according to the maximum values ​​of the multiple groups of state similarities in descending order within the group, and using the multiple maximum similarities as multiple real-time fault probabilities of the multiple real-time fault levels; traversing the multiple real-time fault probabilities based on a preset similarity threshold, and screening P real-time fault levels and P real-time fault probabilities of P types of sample fault states from the multiple real-time fault levels; associating and storing the P types of sample fault states, the P real-time fault levels and the P real-time fault probabilities, and outputting the initial state mapping set; if the multiple real-time fault probabilities are all less than the preset similarity threshold, the initial state mapping set is an empty set.

[0029] First, the Euclidean distance matrix between the first-level feature vector and multiple groups of sample state feature templates in the first-level state pattern sub-library is traversed and calculated to measure the similarity between the input feature and each sample template. Then, the Euclidean distance is converted into a similarity index (for example, by mapping using an inverse or exponential function), and the similarity values ​​of each template are sorted in descending order within the group, and multiple sample states corresponding to the maximum similarity values ​​are screened out. Based on the historical fault level information corresponding to these sample states, multiple real-time fault levels are retrieved, and the corresponding maximum similarity values ​​are used as the real-time fault probabilities of these fault levels, forming multiple pairs of real-time fault levels and real-time fault probabilities. Next, based on a preset similarity threshold, the real-time fault probabilities are traversed to screen out the top P sample fault states with similarities above the threshold, along with their corresponding P real-time fault levels and real-time fault probabilities. Finally, the P sample fault states screened out, the corresponding P real-time fault levels, and real-time fault probabilities are associated and stored, and output as an initial state mapping set. If all real-time fault probabilities are below the preset similarity threshold, the initial state mapping set is determined to be empty, indicating that the current input feature fails to effectively match any known state template.

[0030] Furthermore, the multi-level real-time operating parameters are dynamically loaded into the elevator state pattern library, hierarchical device state mapping is triggered, and a hierarchical state confidence vector is output. The method further includes: If the initial state mapping set is a non-empty set, the second-level sensitive indicators are retrieved from the dynamic hierarchical sensitive indicator set; based on the indicator composition of the second-level sensitive indicators, the distribution cabinet backtracks to retrieve the second-level operating parameter sequence, and obtains the second-level feature vector through feature extraction; the first-level feature vector and the second-level feature vector are fused and loaded into the second-level state pattern sub-library of the elevator state pattern library, and the first verification state mapping set is matched and output; the first verification state mapping set is used to verify the state credibility of the initial state mapping set, and the first state confidence is output; if the first state confidence is higher than the preset confidence threshold, the initial state mapping set and the first verification state mapping set are weightedly fused to generate the hierarchical state confidence vector; if the first state confidence is lower than the preset confidence threshold, the next-level sensitive indicator is iteratively called for verification until the confidence meets the standard, and the multi-level mapping results are integrated to output the hierarchical state confidence vector.

[0031] If the initial state mapping set is a non-empty set, the second-level sensitive indicator set is retrieved from the dynamic hierarchical sensitive indicator set; according to the indicator composition of the second-level sensitive indicator, the corresponding second-level operating parameter sequence is retrieved from the historical data of the distribution cabinet, and features are extracted to obtain the second-level feature vector; the first-level feature vector and the second-level feature vector are fused and loaded into the second-level state pattern sub-library of the elevator state pattern library, and the fused features are matched using the state feature template of the sub-library to output the first verification state mapping set; the state credibility of the initial state mapping set is verified using the first verification state mapping set, and the first state confidence is calculated and output; when the first state confidence is higher than the preset confidence threshold, the initial state mapping set is fused with the first verification state mapping set based on the weighted fusion algorithm to generate the final hierarchical state confidence vector; if the first state confidence is lower than the preset confidence threshold, the subsequent lower-level sensitive indicators are iteratively called, and the above verification process is repeated until the confidence reaches the preset standard, and the mapping results of each level are integrated to output a comprehensive hierarchical state confidence vector.

[0032] Furthermore, the method includes: using the first verification state mapping set to perform state credibility verification on the initial state mapping set and outputting a first state confidence level; Perform state conflict detection on the first verification state mapping set and the initial state mapping set, filter and output the first conflict state mapping set and the initial conflict mapping set; calculate the first JSD distribution similarity of the first conflict state mapping set and the initial conflict mapping set as the first state confidence.

[0033] The fault state types and fault levels in the first verification state mapping set and the initial state mapping set are compared one by one to identify state conflicts between the two mapping sets. This means that the first conflicting state mapping set and the initial conflicting mapping set are selected. Conflicts refer to pairs of states in the two mapping sets that have inconsistent fault state types or significantly different fault levels. Based on the selected first conflicting state mapping set and the initial conflicting mapping set, the Jensen-Shannon divergence (JSD) distribution similarity of their corresponding fault state probability distributions is calculated. The specific calculation steps include smoothing the two probability distributions, calculating the intermediate mixed distribution, and then obtaining the similarity value according to the JSD definition. Finally, the JSD distribution similarity is used as a measure of the first state confidence, reflecting the consistency and credibility between the initial state mapping set and the first verification state mapping set.

[0034] Perform state transition prediction on the hierarchical state confidence vector and output a state transition probability matrix.

[0035] Furthermore, the hierarchical state confidence vector is subjected to state transition prediction, and a state transition probability matrix is ​​outputted. The method includes: Interactively obtain multiple sample state confidence vector sequences, and output a state transition frequency chain by performing state transition frequency statistics on the multiple sample state confidence vector sequences; construct a reference state transition frequency matrix based on the state transition frequency chain; perform state transition probability normalization correction on the hierarchical state confidence vector based on the reference state transition frequency matrix, and output the state transition probability matrix.

[0036] Interactively obtain a state confidence vector sequence of multiple historical samples, where the state confidence vector sequence covers the time evolution process of normal and multiple fault states; obtain a state transition frequency chain by statistically analyzing the transition relationship between adjacent states in the state confidence vector sequences of multiple samples, which records the number and frequency of transitions between different states; construct a reference state transition frequency matrix based on the state transition frequency chain, where the elements in the matrix represent the transition frequency between states, the rows of the matrix represent the current state, and the columns represent the possible next state; for the hierarchical state confidence vector at the current moment, perform normalization correction on the state transition probability based on the reference state transition frequency matrix, and generate a state transition probability matrix after correction to reflect the transition probability distribution between each state; finally, output the state transition probability matrix to provide a basis for subsequent real-time fault probability distribution calculation and multi-level warning signal generation.

[0037] The hierarchical state confidence vector and the state transition probability matrix are integrated to output a real-time fault probability distribution.

[0038] First, the hierarchical state confidence vector at the current moment is used as the initial probability distribution of the fault state. Then, based on the state transition probability matrix, the initial probability distribution is weightedly corrected and time-series predicted, and the conditional transition probability of each fault state is calculated to obtain the state probability distribution adjusted based on the historical transition law. Subsequently, the corrected probability distribution is fused with the initial confidence vector, and a comprehensive real-time fault probability distribution is generated through methods such as weighted averaging or Bayesian updating. Finally, the real-time fault probability distribution is output to support fault diagnosis, risk assessment, and the generation of multi-level warning signals.

[0039] A multi-level warning signal is generated according to the real-time fault probability distribution to trigger a cascade safety protection response.

[0040] Based on the probability values ​​corresponding to each fault state in the real-time fault probability distribution, multi-level warning thresholds are set, corresponding to low-level warning, medium-level warning and high-level warning levels respectively; the probability of each fault state is compared with the preset threshold to determine the current fault risk level; when the fault probability exceeds the corresponding threshold, a warning signal of the corresponding level is generated; according to the generated multi-level warning signal, the corresponding level of safety protection response measures are triggered, including but not limited to alarm prompts, operation restrictions, automatic shutdown and activation of emergency fault handling procedures.

[0041] In summary, the embodiments of the present application have at least the following technical effects: First, the sensitivity of multiple monitoring indicators of the power distribution cabinet to special equipment status is verified to obtain a dynamic hierarchical sensitive indicator set. Next, a multi-level state pattern sub-library is pre-built offline based on the dynamic hierarchical sensitive indicator set, and an elevator state pattern library is generated through hierarchical association and integration. Simultaneously, based on the dynamic hierarchical sensitive indicator set, multiple levels of real-time operating parameters are retrieved from the power distribution cabinet step by step according to sensitivity priority. Furthermore, these multiple levels of real-time operating parameters are dynamically loaded into the elevator state pattern library, triggering hierarchical equipment state mapping and outputting a hierarchical state confidence vector. Next, state transition prediction is performed on the hierarchical state confidence vector, outputting a state transition probability matrix. The hierarchical state confidence vector and the state transition probability matrix are then combined to output a real-time fault probability distribution. Finally, a multi-level warning signal is generated based on the real-time fault probability distribution, triggering a step-by-step safety protection response. This solves the technical problem of special equipment status identification relying on sensors in the prior art, achieving accurate identification of elevator operating status based on electrical parameters, thereby reducing reliance on external sensors and improving the real-time and reliability of state monitoring.

[0042] Embodiment 2 is based on the same inventive concept as the special equipment status identification method based on electrical parameter pattern matching in the above embodiment. Figure 2 As shown, the present application provides a special equipment status identification system based on electrical parameter pattern matching, wherein the system includes: Sensitivity verification module 11: performs special equipment state sensitivity verification on multiple monitoring indicators of the distribution cabinet to obtain a dynamic hierarchical sensitive indicator set; pattern library construction module 12: pre-constructs a multi-level state pattern sub-library offline based on the dynamic hierarchical sensitive indicator set, and generates an elevator state pattern library through hierarchical association integration; parameter retrieval module 13: retrieves multi-level real-time operating parameters from the distribution cabinet step by step according to the sensitivity priority based on the dynamic hierarchical sensitive indicator set; state mapping module 14: dynamically loads the multi-level real-time operating parameters into the elevator state pattern library, triggers hierarchical equipment state mapping, and outputs a hierarchical state confidence vector; state transition prediction module 15: performs state transition prediction on the hierarchical state confidence vector and outputs a state transition probability matrix; fault probability output module 16: integrates the hierarchical state confidence vector and the state transition probability matrix to output a real-time fault probability distribution; early warning response module 17: generates a multi-level early warning signal based on the real-time fault probability distribution to trigger a step safety protection response.

[0043] Furthermore, the sensitivity verification module 11 is used to perform the following method: The application environment characteristics of the distribution cabinet are retrieved, and the application environment characteristics and the distribution cabinet model code are used as device fingerprint code retrieval conditions, and multiple full-dimensional operation logs of the devices are called; the multiple full-dimensional operation logs of the devices are aggregated based on the indicator type to obtain multiple multi-state time series data sets of multiple operation indicators; the state sensitivity of special equipment is quantified based on the multiple multi-state time series data sets, and multiple comprehensive operation sensitivities are output; based on the multiple comprehensive operation sensitivities, the multiple operation indicators are dynamically graded, and the dynamic graded sensitive indicator set is output.

[0044] Furthermore, the pattern library construction module 12 is used to execute the following method: The dynamic hierarchical sensitive indicator set is decomposed based on the hierarchical relationship to obtain multi-level sensitive indicators; using the specific indicator set of the first-level sensitive indicators as the screening condition, multiple groups of sample fault state time series data sets of multiple sample fault states under multiple groups of sample fault levels are retrieved from the full-dimensional operation logs of the multiple devices; by performing indicator fluctuation scale analysis on the multiple groups of sample fault state time series data sets, multiple groups of state feature templates are output; the multiple sample fault states, multiple groups of sample fault levels and multiple groups of sample state feature templates are hierarchically associated and stored to obtain a first-level state pattern sub-library; a multi-level state pattern sub-library is iteratively constructed in hierarchical order, and a bidirectional mapping relationship index table between the multi-level sensitive indicators and the multi-level state pattern sub-library is constructed to generate the elevator state pattern library.

[0045] Furthermore, the state mapping module 14 is configured to execute the following method: The first-level operating parameter sequence of the first-level sensitive indicator is retrieved from the distribution cabinet in real time; the first-level operating parameter sequence is feature extracted to obtain a first-level feature vector; the first-level feature vector is loaded into the first-level state pattern sub-library of the elevator state pattern library, and the initial state mapping set is matched and output; if the initial state mapping set is an empty set, a monitoring cycle of the first-level sensitive indicator is performed until the initial state mapping set is a non-empty set, thereby triggering a hierarchical device state mapping.

[0046] Furthermore, the state mapping module 14 is configured to execute the following method: If the initial state mapping set is a non-empty set, the second-level sensitive indicators are retrieved from the dynamic hierarchical sensitive indicator set; based on the indicator composition of the second-level sensitive indicators, the distribution cabinet backtracks to retrieve the second-level operating parameter sequence, and obtains the second-level feature vector through feature extraction; the first-level feature vector and the second-level feature vector are fused and loaded into the second-level state pattern sub-library of the elevator state pattern library, and the first verification state mapping set is matched and output; the first verification state mapping set is used to verify the state credibility of the initial state mapping set, and the first state confidence is output; if the first state confidence is higher than the preset confidence threshold, the initial state mapping set and the first verification state mapping set are weightedly fused to generate the hierarchical state confidence vector; if the first state confidence is lower than the preset confidence threshold, the next-level sensitive indicator is iteratively called for verification until the confidence meets the standard, and the multi-level mapping results are integrated to output the hierarchical state confidence vector.

[0047] Furthermore, the state mapping module 14 is configured to execute the following method: The method comprises the following steps: traversing and calculating the Euclidean distance matrix of multiple groups of sample state feature templates in the first-level feature vector and the first-level state pattern sub-library, and outputting multiple groups of state similarities; retrieving multiple real-time fault levels from the multiple groups of sample fault levels according to the maximum values ​​of the multiple groups of state similarities in descending order within the group, and using the multiple maximum similarities as multiple real-time fault probabilities of the multiple real-time fault levels; traversing the multiple real-time fault probabilities based on a preset similarity threshold, and screening P real-time fault levels and P real-time fault probabilities of P types of sample fault states from the multiple real-time fault levels; associating and storing the P types of sample fault states, the P real-time fault levels and the P real-time fault probabilities, and outputting the initial state mapping set; if the multiple real-time fault probabilities are all less than the preset similarity threshold, the initial state mapping set is an empty set.

[0048] Furthermore, the state mapping module 14 is configured to execute the following method: Perform state conflict detection on the first verification state mapping set and the initial state mapping set, filter and output the first conflict state mapping set and the initial conflict mapping set; calculate the first JSD distribution similarity of the first conflict state mapping set and the initial conflict mapping set as the first state confidence.

[0049] Furthermore, the state transition prediction module 15 is configured to execute the following method: Interactively obtain multiple sample state confidence vector sequences, and output a state transition frequency chain by performing state transition frequency statistics on the multiple sample state confidence vector sequences; construct a reference state transition frequency matrix based on the state transition frequency chain; perform state transition probability normalization correction on the hierarchical state confidence vector based on the reference state transition frequency matrix, and output the state transition probability matrix.

[0050] Furthermore, the sensitivity verification module 11 is used to perform the following method: The multiple multi-state time series data sets are decomposed based on the device state labels to obtain multiple baseline state record sets and multiple groups of fault state record sets; based on the multiple baseline state record sets, the multiple groups of fault state record sets are quantified using KL divergence to output multiple groups of fault state KL sensitivity scores; based on the recurrence frequency of M fault states, the multiple groups of fault state KL sensitivity scores are weighted within the group to output the multiple comprehensive operation sensitivities.

[0051] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0052] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0053] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A special equipment status identification method based on electrical parameter pattern matching, characterized in that: The method comprises: The sensitivity of special equipment status is verified for multiple monitoring indicators of the distribution cabinet to obtain a dynamic hierarchical sensitive indicator set; Pre-build a multi-level state mode sub-library offline based on the dynamic hierarchical sensitive indicator set, and generate an elevator state mode library through hierarchical association integration; According to the dynamic hierarchical sensitive indicator set, multi-level real-time operating parameters are retrieved from the power distribution cabinet step by step according to the sensitivity priority; Dynamically loading the multi-level real-time operating parameters into the elevator state pattern library, triggering hierarchical device state mapping, and outputting a hierarchical state confidence vector; Performing state transition prediction on the hierarchical state confidence vector and outputting a state transition probability matrix; fusing the hierarchical state confidence vector and the state transition probability matrix to output a real-time fault probability distribution; A multi-level warning signal is generated according to the real-time fault probability distribution to trigger a cascade safety protection response.

2. The special equipment state identification method based on electrical parameter pattern matching according to claim 1, characterized in that: The sensitivity verification of special equipment status is performed on multiple monitoring indicators of the power distribution cabinet to obtain a dynamic hierarchical sensitive indicator set. The method includes: Retrieve the application environment characteristics of the power distribution cabinet, use the application environment characteristics and the power distribution cabinet model code as device fingerprint code retrieval conditions, and call multiple device full-dimensional operation logs; Aggregating the full-dimensional operation logs of the multiple devices based on the indicator type to obtain multiple multi-state time series data sets of the multiple operation indicators; quantifying the state sensitivity of the special equipment based on the multiple multi-state time series data sets, and outputting multiple comprehensive operation sensitivities; Dynamically classify the multiple operating indicators based on the multiple operating comprehensive sensitivities, and output the dynamically classified sensitivity indicator set.

3. The special equipment state identification method based on electrical parameter pattern matching according to claim 2, characterized in that: Pre-building a multi-level state mode sub-library offline based on the dynamic hierarchical sensitive indicator set, and generating an elevator state mode library through hierarchical association integration, the method includes: Decomposing the dynamic hierarchical sensitive indicator set based on the hierarchical relationship to obtain multi-level sensitive indicators; Using a specific indicator set of the first-level sensitive indicators as a screening condition, retrieving multiple groups of sample fault state time series data sets of multiple sample fault states at multiple groups of sample fault levels from the full-dimensional operation logs of the multiple devices; Outputting multiple groups of state feature templates by performing indicator fluctuation scale analysis on the multiple groups of sample fault state time series data sets; hierarchically associatively storing the plurality of sample fault states, the plurality of groups of sample fault levels and the plurality of groups of sample state feature templates to obtain a first-level state pattern sub-library; A multi-level state mode sub-library is iteratively constructed in hierarchical order, and a bidirectional mapping relationship index table between the multi-level sensitive indicators and the multi-level state mode sub-library is constructed to generate the elevator state mode library.

4. The special equipment state identification method based on electrical parameter pattern matching according to claim 3, characterized in that: Dynamically loading the multi-level real-time operating parameters into the elevator state model library, triggering hierarchical device state mapping, and outputting a hierarchical state confidence vector, the method includes: Retrieving the first-level operating parameter sequence of the first-level sensitive indicator from the power distribution cabinet in real time; Performing feature extraction on the first-level operating parameter sequence to obtain a first-level feature vector; Loading the first-level feature vector into the first-level state pattern sub-library of the elevator state pattern library, matching and outputting an initial state mapping set; If the initial state mapping set is an empty set, a monitoring cycle of the first-level sensitive indicators is performed until the initial state mapping set is a non-empty set, thereby triggering hierarchical device state mapping.

5. The special equipment state identification method based on electrical parameter pattern matching according to claim 4, characterized in that: Dynamically loading the multi-level real-time operating parameters into the elevator state model library, triggering hierarchical device state mapping, and outputting a hierarchical state confidence vector, the method further includes: If the initial state mapping set is a non-empty set, retrieving the second-level sensitive indicator from the dynamic hierarchical sensitive indicator set; According to the indicator composition of the second-level sensitive indicators, the power distribution cabinet backtracks and retrieves the second-level operating parameter sequence, and obtains the second-level feature vector through feature extraction; The first-level feature vector and the second-level feature vector are fused and loaded into the second-level state pattern sub-library of the elevator state pattern library, and a first verification state mapping set is outputted through matching; Using the first verification state mapping set to perform state credibility verification on the initial state mapping set, and outputting a first state confidence level; If the first state confidence is higher than a preset confidence threshold, weightedly fusing the initial state mapping set and the first verification state mapping set to generate the hierarchical state confidence vector; If the first state confidence is lower than the preset confidence threshold, the next level of sensitive indicators are iteratively called for verification until the confidence meets the standard, and the multi-level mapping results are integrated to output a hierarchical state confidence vector.

6. The special equipment state identification method based on electrical parameter pattern matching according to claim 4, characterized in that: The first-level feature vector is loaded into the first-level state pattern sub-library of the elevator state pattern library, and an initial state mapping set is outputted through matching. The method includes: Traversing and calculating the Euclidean distance matrix between the first-level feature vector and multiple groups of sample state feature templates in the first-level state pattern sub-library, and outputting multiple groups of state similarities; According to the maximum values ​​of the multiple groups of state similarities arranged in descending order within the group, multiple real-time fault levels are retrieved from the multiple groups of sample fault levels, and the multiple maximum similarities are used as multiple real-time fault probabilities of the multiple real-time fault levels; Traversing the multiple real-time fault probabilities based on a preset similarity threshold, screening P real-time fault levels and P real-time fault probabilities of P types of sample fault states from the multiple real-time fault levels; Associatively storing the P sample fault states, the P real-time fault levels, and the P real-time fault probabilities, and outputting the initial state mapping set; If the multiple real-time fault probabilities are all smaller than a preset similarity threshold, the initial state mapping set is an empty set.

7. The special equipment state identification method based on electrical parameter pattern matching according to claim 5, characterized in that: Using the first verification state mapping set to perform state credibility verification on the initial state mapping set and outputting a first state confidence level, the method includes: Performing state conflict detection on the first verification state mapping set and the initial state mapping set, filtering and outputting a first conflicting state mapping set and an initial conflicting mapping set; A first JSD distribution similarity between the first conflict state mapping set and the initial conflict mapping set is calculated as the first state confidence.

8. The special equipment status identification method based on electrical parameter pattern matching according to claim 1, characterized in that: Performing state transition prediction on the hierarchical state confidence vector and outputting a state transition probability matrix, the method comprising: Interactively obtain multiple sample state confidence vector sequences, and output a state transition frequency chain by performing state transition frequency statistics on the multiple sample state confidence vector sequences; Constructing a reference state transition frequency matrix based on the state transition frequency chain; The state transition probability of the hierarchical state confidence vector is normalized and corrected based on the reference state transition frequency matrix, and the state transition probability matrix is ​​output.

9. The special equipment state identification method based on electrical parameter pattern matching according to claim 2, characterized in that: Quantifying the state sensitivity of special equipment based on the multiple multi-state time series data sets and outputting multiple comprehensive operation sensitivities, the method comprising: Decomposing the plurality of multi-state time series data sets based on device state labels to obtain a plurality of baseline state record sets and a plurality of fault state record sets; Taking the multiple reference state record sets as benchmarks, using KL divergence to quantify the multiple groups of fault state record sets, and outputting multiple groups of fault state KL sensitivity scores; According to the recurrence frequencies of the M fault states, intra-group weighting of the multiple groups of fault state KL sensitivity scores is performed to output the multiple operation comprehensive sensitivities.

10. Special equipment status identification system based on electrical parameter pattern matching, characterized in that: A system for implementing the special equipment state identification method based on electrical parameter pattern matching according to any one of claims 1 to 9, comprising: Sensitivity verification module: performs special equipment status sensitivity verification on multiple monitoring indicators of the distribution cabinet to obtain a dynamic hierarchical sensitive indicator set; Pattern library construction module: pre-constructs a multi-level state pattern sub-library offline based on the dynamic hierarchical sensitive indicator set, and generates an elevator state pattern library through hierarchical association integration; Parameter retrieval module: according to the dynamic hierarchical sensitive indicator set, multi-level real-time operating parameters are retrieved from the power distribution cabinet step by step according to the sensitivity priority; State mapping module: dynamically loads the multi-level real-time operating parameters into the elevator state pattern library, triggers hierarchical equipment state mapping, and outputs a hierarchical state confidence vector; State transition prediction module: performs state transition prediction on the hierarchical state confidence vector and outputs a state transition probability matrix; Fault probability output module: fuses the hierarchical state confidence vector and state transition probability matrix to output real-time fault probability distribution; Early warning response module: generates multi-level early warning signals according to the real-time fault probability distribution, and triggers cascade safety protection response.

Images