Operation non-uniformity analysis method for driving circuit

Abstract

The invention provides an operation non-uniformity analysis method for a driving circuit, and belongs to the technical field of data processing, and the method comprises the steps: collecting the real-time operation data of all operation parameters of the driving circuit in real time, analyzing the real-time operation data, and determining a real-time feature matrix; acquiring a plurality of historical feature matrixes, and determining whether the real-time operation data is non-uniform data or not; for the real-time operation data which are non-uniform data, determining a deviation matrix of the real-time operation data and a second non-uniform matrix, and determining a non-uniform level of each operation parameter in the second non-uniform matrix; and generating a non-uniform report for the real-time operation data which is non-uniform data. The accuracy and real-time performance of the non-uniformity detection of the driving circuit are realized, the accuracy and response efficiency of non-uniformity identification are improved, the resource allocation and safety management are optimized, the efficiency and real-time performance of non-uniformity processing are remarkably improved, and the safe and stable operation of the driving circuit is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for analyzing uneven operation of a driving circuit. Background Art

[0002] In the prior art, the abnormality detection of the driving circuit mostly relies on the traditional single threshold or simple comparison algorithm, that is, the abnormality is identified by monitoring whether the operating parameters exceed the fixed preset threshold. This method works well when detecting simple abnormalities, but has significant shortcomings when dealing with complex multi-parameter and dynamic abnormalities. For example, when the circuit operation state is affected by multiple factors and there is a coupling relationship between the factors, a single threshold detection often has difficulty in identifying potential problems, and is not accurate enough in multi-level and multi-dimensional abnormality identification, which may lead to false alarms or missed alarms.

[0003] Therefore, the present invention provides a method for analyzing operation unevenness of a driving circuit. Summary of the invention

[0004] The present invention provides a method for analyzing uneven operation of a driving circuit. The method determines a real-time feature matrix by analyzing real-time operation data, analyzes the real-time feature matrix of the real-time operation data and multiple historical feature matrices, determines whether the real-time operation data is uneven data, and for real-time operation data that is uneven data, determines the uneven level of each operation parameter and generates an uneven report, thereby realizing precise and real-time uneven detection of the driving circuit, improving the accuracy and response efficiency of uneven identification, optimizing resource allocation and safety management, significantly improving the efficiency and real-time performance of uneven processing, and ensuring safe and stable operation of the driving circuit.

[0005] The present invention provides a method for analyzing uneven operation of a driving circuit, comprising: 101: collecting real-time operation data of all operation parameters of the driving circuit in real time, and analyzing the real-time operation data to determine a real-time characteristic matrix; 102: Acquire multiple historical feature matrices, analyze the real-time feature matrix of the real-time operation data and the multiple historical feature matrices, and determine whether the real-time operation data is uneven data; 103: for the real-time operation data being non-uniform data, determining a deviation matrix and a second non-uniform matrix of the real-time operation data, and determining a non-uniform level of each operation parameter in the second non-uniform matrix; 104 : For the real-time operation data that is non-uniform data, generate a non-uniform report based on the real-time feature matrix, the first non-uniform matrix, and non-uniform levels of all operation parameters in the non-uniform vector.

[0006] According to a method for analyzing uneven operation of a driving circuit provided by the present invention, real-time operation data of all operation parameters of the driving circuit are collected in real time, including: Analyze the characteristics of the driving circuit to determine the collection requirements of each operating parameter data, and determine the corresponding sensor group based on the collection requirements of the operating parameter data; Analyze potential non-uniformity sources of the driving circuit, determine multiple acquisition locations for acquiring each operating parameter data and the location priority of each acquisition location; Based on all the collection positions for collecting the operating parameter data and the position priority of each collection position, a sensor group corresponding to the operating parameter data is installed on the driving circuit; Based on the sensor group, the operating parameter data of each operating parameter within a specified time period is collected to determine the real-time operating data.

[0007] According to a method for analyzing uneven operation of a driving circuit provided by the present invention, real-time operation data is analyzed to determine a real-time characteristic matrix, including: Preprocessing the operating parameter data of each operating parameter in the real-time operating data, performing feature extraction on the preprocessed operating parameter data of each operating parameter, and determining a parameter feature vector of the operating parameter corresponding to each operating parameter data; A real-time feature matrix of the real-time operation data is determined based on the parameter feature vectors of all the operation parameters.

[0008] According to a method for analyzing uneven operation of a driving circuit provided by the present invention, a plurality of historical feature matrices are obtained, a real-time feature matrix of real-time operation data and the plurality of historical feature matrices are analyzed, and whether the real-time operation data is uneven data is determined, including: Performing a first extraction of a historical feature matrix in a historical analysis report of a first specified number of specified time periods before a specified time period of the real-time operation data; performing a first sorting on the first specified number of historical feature matrices extracted first based on a specified time period of each historical feature matrix, and determining a sequence label of each historical feature matrix extracted first based on the first sorting result; Based on a first specified number of historical feature matrices, sequence labels of each historical feature matrix, and a real-time feature matrix, determining an uneven value of each feature of a parameter feature vector of each operating parameter, and determining a first uneven matrix of real-time operating data; If the first non-uniform matrix of the real-time operation data is not empty, determining that the real-time operation data is non-uniform data; Otherwise, the real-time operation data is determined to be uniform operation data, and a uniform report is generated based on the real-time feature matrix of the real-time operation data and the non-uniform value of each feature in the parameter feature vector of each operation parameter.

[0009] According to a method for analyzing uneven operation of a driving circuit provided by the present invention, based on a first specified number of historical feature matrices, a sequence label of each historical feature matrix, and a real-time feature matrix, an uneven value of each feature of a parameter feature vector of each operating parameter is determined, and a first uneven matrix of real-time operating data is determined, including: Determine an uneven value of each feature in a parameter feature vector for each running parameter based on all historical feature matrices, sequence labels of each historical feature matrix, and a real-time feature matrix; in, represents the uneven value of the jth feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix, represents the third decay parameter of the sequence label based on the historical feature matrix, The eigenvalue of the historical feature matrix of the kth sequence label for the jth feature in the parameter feature vector of the i-th running parameter, represents the eigenvalue of the jth feature in the parameter feature vector of the i-th operating parameter in the real-time feature matrix, N1 represents the number of historical feature matrices, represents the first attenuation parameter of the j-th column eigenvalue in the real-time feature matrix, represents the second attenuation parameter of the jth column eigenvalue in the real-time feature matrix, N2 represents the number of eigenvalues ​​in the parameter eigenvector, represents the number of operating parameters in the real-time feature matrix, Indicates adjustment parameters; Compare the uneven value of each feature of the parameter feature vector of each operating parameter with a preset uneven threshold value corresponding to each feature; If there is an uneven value of any feature in the parameter feature vector of the operating parameter that is greater than an uneven threshold of the corresponding feature, determining an uneven vector of the operating parameter based on the uneven values ​​of all features in the parameter feature vector of the operating parameter; in, represents the non-uniform vector based on the i-th operating parameter in the real-time feature matrix, They respectively represent the uneven values ​​of the 1st feature and the N2th feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix; otherwise, adjusting the non-uniform values ​​of all characteristics of the operating parameter to 0, and determining the non-uniform vector of the operating parameter; If the non-uniform values ​​in the non-uniform vectors of all operating parameters in the real-time feature matrix are all 0, it is determined that the first non-uniform matrix of the real-time operating data is empty; Otherwise, a first non-uniform matrix of real-time operation data is determined based on the non-uniform vectors of all operation parameters: ; in, , They respectively represent the uneven vectors based on the 1st operating parameter, the i-th operating parameter, and the N3-th operating parameter in the real-time feature matrix.

[0010] According to a method for analyzing uneven operation of a driving circuit provided by the present invention, for real-time operation data that is uneven data, a deviation matrix of the real-time operation data and a second uneven matrix are determined, comprising: Performing a second extraction of a second specified number of historical feature matrices in which the first uneven matrix is ​​empty in the historical analysis report of a second specified number of specified time periods before the specified time period of the real-time running data of the uneven data; Determine a fitting feature matrix based on a second specified number of historical feature matrices extracted from the second step; Determine a deviation matrix of real-time operation data based on the fitting feature matrix and the real-time feature matrix; All non-uniform values ​​of 0 in the non-uniform vectors of all operating parameters in the first non-uniform matrix are adjusted based on the deviation matrix, a second non-uniform matrix is ​​determined based on the adjusted first non-uniform matrix, and the operating parameters of the adjusted non-uniform vectors in the second non-uniform matrix are marked.

[0011] According to the present invention, a method for analyzing uneven operation of a driving circuit is provided, and the uneven level of each operating parameter in a second uneven matrix is ​​determined, including: A level division rule is determined based on the second non-uniform matrix, and a non-uniform level of each operating parameter in the non-uniform vector is determined based on the level division rule, wherein the non-uniform level includes a first non-uniform level, a second non-uniform level, a third non-uniform level and a fourth non-uniform level.

[0012] According to a method for analyzing uneven operation of a driving circuit provided by the present invention, a level division rule is determined based on a second uneven matrix, and an uneven level of each operating parameter in an uneven vector is determined based on the level division rule, including: For an uneven vector in the second uneven matrix, where the uneven value of a feature included in the uneven vector is greater than the uneven threshold of the corresponding feature and the number of features exceeds the first number threshold, determining the operating parameter of the uneven vector to be a first uneven level; For an uneven vector in the second uneven matrix, where the uneven value of a feature contained in the uneven vector is greater than the uneven threshold of the corresponding feature and the number of features is between the second number threshold and the first number threshold, determining the operating parameter of the uneven vector to be a second uneven level; Determining the operating parameter marked in the second non-uniform matrix as a fourth non-uniform level; The operating parameters of all the non-uniform vectors in the second non-uniform matrix except the non-uniform vectors of the operating parameters of the first non-uniform level, the second non-uniform level and the fourth non-uniform level are determined as the third non-uniform level.

[0013] Compared with the prior art, the present invention has the following beneficial effects: By analyzing the real-time operation data to determine the real-time feature matrix, analyzing the real-time feature matrix of the real-time operation data and multiple historical feature matrices, it is determined whether the real-time operation data is uneven data. For real-time operation data that is uneven data, the uneven level of each operation parameter is determined, and an uneven report is generated. This realizes the precise and real-time detection of unevenness in the drive circuit, improves the accuracy and response efficiency of uneven identification, optimizes resource allocation and safety management, significantly improves the efficiency and real-time performance of uneven processing, and ensures the safe and stable operation of the drive circuit. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0015] Figure 1 It is a flow chart of a method for analyzing uneven operation of a driving circuit provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Embodiment 1: The embodiment of the present invention provides a method for analyzing uneven operation of a driving circuit, such as Figure 1 As shown, including: 101: collecting real-time operation data of all operation parameters of the driving circuit in real time, and analyzing the real-time operation data to determine a real-time characteristic matrix; 102: Acquire multiple historical feature matrices, analyze the real-time feature matrix of the real-time operation data and the multiple historical feature matrices, and determine whether the real-time operation data is uneven data; 103: for the real-time operation data being non-uniform data, determining a deviation matrix and a second non-uniform matrix of the real-time operation data, and determining a non-uniform level of each operation parameter in the second non-uniform matrix; 104 : For the real-time operation data that is non-uniform data, generate a non-uniform report based on the real-time feature matrix, the first non-uniform matrix, and non-uniform levels of all operation parameters in the non-uniform vector.

[0018] In this embodiment, for the real-time operation data of the uneven data, an uneven report is generated by combining the real-time feature matrix, the first uneven matrix (for preliminary uneven analysis) and the uneven level of each parameter. The report comprehensively analyzes the current operation status and uneven level to provide a basis for system optimization and preventive maintenance.

[0019] In this embodiment, the deviation matrix is ​​a difference matrix that quantifies the deviation of the real-time data from the reference state.

[0020] In this embodiment, the unevenness level indicates the classification of unevenness degrees, and is used to manage unevenness in a hierarchical manner, thereby optimizing response measures.

[0021] The beneficial effects of the above technical solution are as follows: by analyzing the real-time operation data to determine the real-time feature matrix, analyzing the real-time feature matrix of the real-time operation data and multiple historical feature matrices, determining whether the real-time operation data is non-uniform data, and for real-time operation data that is non-uniform data, determining the non-uniformity level of each operation parameter, and generating a non-uniformity report, the non-uniformity detection of the drive circuit is realized in a precise and real-time manner, the accuracy and response efficiency of non-uniformity identification are improved, resource allocation and safety management are optimized, the efficiency and real-time performance of non-uniformity processing are significantly improved, and the safe and stable operation of the drive circuit is ensured.

[0022] Embodiment 2: The embodiment of the present invention provides a method for analyzing uneven operation of a driving circuit, which collects real-time operation data of all operation parameters of the driving circuit in real time, including: Analyze the characteristics of the driving circuit to determine the collection requirements of each operating parameter data, and determine the corresponding sensor group based on the collection requirements of the operating parameter data; Analyze potential non-uniformity sources of the driving circuit, determine multiple acquisition locations for acquiring each operating parameter data and the location priority of each acquisition location; Based on all the collection positions for collecting the operating parameter data and the position priority of each collection position, a sensor group corresponding to the operating parameter data is installed on the driving circuit; Based on the sensor group, the operating parameter data of each operating parameter within a specified time period is collected to determine the real-time operating data.

[0023] In this embodiment, the operating parameters may be voltage, current, temperature, frequency, phase, power, etc.

[0024] In this embodiment, the sensor group may be a voltage sensor group, a current sensor group, a temperature sensor group, a power sensor group, a frequency meter group, etc.

[0025] In this embodiment, the importance of various operating parameters of the driving circuit (such as voltage, current, temperature, etc.) to the system stability is evaluated, the data acquisition frequency and accuracy requirements are determined, and the appropriate sensor group is selected and configured based on the analysis results to ensure that the data acquisition requirements of each parameter can be met.

[0026] In this embodiment, potential sources of non-uniformity in the driving circuit (such as hot spots and load imbalance) are studied, key areas where non-uniformity may occur are identified, multiple data collection locations are determined for each operating parameter, and each location is given a priority to indicate its importance and the necessity of collection.

[0027] In this embodiment, according to the determined collection positions and position priorities, the sensor groups are installed at key points of the driving circuit to achieve optimal configuration of coverage and priority.

[0028] In this embodiment, during operation, the sensor group collects operating parameter data at a set time period to form real-time operating data of the drive circuit.

[0029] The beneficial effects of the above technical solution are: real-time acquisition of real-time operating data of all operating parameters of the driving circuit can improve the real-time detection and positioning capabilities of potential circuit inhomogeneities and inhomogeneities, and enhance the diagnosis and preventive maintenance effects of the circuit.

[0030] Embodiment 3: An embodiment of the present invention provides a method for analyzing uneven operation of a driving circuit, which analyzes real-time operation data to determine a real-time characteristic matrix, including: Preprocessing the operating parameter data of each operating parameter in the real-time operating data, performing feature extraction on the preprocessed operating parameter data of each operating parameter, and determining a parameter feature vector of the operating parameter corresponding to each operating parameter data; A real-time feature matrix of the real-time operation data is determined based on the parameter feature vectors of all the operation parameters.

[0031] In this embodiment, the raw data of each operating parameter in the real-time operating data is preprocessed, including steps such as denoising, filtering, and normalization, so as to improve the accuracy and stability of the data.

[0032] In this embodiment, key features such as average value, peak value, frequency component, trend change, etc. are extracted from the preprocessed data to form a parameter feature vector describing the state of the operating parameter.

[0033] In this embodiment, each operating parameter has a corresponding parameter feature vector, which is used to summarize and express the main features of the data.

[0034] In this embodiment, the parameter feature vectors of all operating parameters are combined to form a real-time feature matrix, which is a global description of the operating state of the entire drive circuit.

[0035] The beneficial effects of the above technical solution are as follows: analyzing real-time operation data to construct a real-time feature matrix can reduce computational complexity, improve the accuracy and response speed of data analysis, and lay the foundation for rapid uneven detection and intelligent diagnosis of circuits.

[0036] Embodiment 4: The embodiment of the present invention provides a method for analyzing uneven operation of a driving circuit, which obtains multiple historical feature matrices, analyzes the real-time feature matrix of real-time operation data and the multiple historical feature matrices, and determines whether the real-time operation data is uneven data, including: Performing a first extraction of a historical feature matrix in a historical analysis report of a first specified number of specified time periods before a specified time period of the real-time operation data; performing a first sorting on the first specified number of historical feature matrices extracted first based on a specified time period of each historical feature matrix, and determining a sequence label of each historical feature matrix extracted first based on the first sorting result; Based on a first specified number of historical feature matrices, sequence labels of each historical feature matrix, and a real-time feature matrix, determining an uneven value of each feature of a parameter feature vector of each operating parameter, and determining a first uneven matrix of real-time operating data; If the first non-uniform matrix of the real-time operation data is not empty, determining that the real-time operation data is non-uniform data; Otherwise, the real-time operation data is determined to be uniform operation data, and a uniform report is generated based on the real-time feature matrix of the real-time operation data and the non-uniform value of each feature in the parameter feature vector of each operation parameter.

[0037] In this embodiment, a first specified number of historical feature matrices are extracted from the historical analysis report before the specified time period, and these matrices contain status information of each operating parameter in the past multiple time periods.

[0038] In this embodiment, the extracted historical feature matrices are first sorted based on the time period to which they belong. After the sorting is completed, a sequence label is assigned to each matrix to reflect its time sequence.

[0039] In this embodiment, if the first non-uniform matrix is ​​not empty, the current real-time operation data is determined to be non-uniform data, indicating that the system has potential non-uniformity or fluctuation; if the first non-uniform matrix is ​​empty, it is determined to be uniform operation data, indicating that the system is in a stable state.

[0040] In this embodiment, when the data is uniform, a uniform report is generated based on the real-time feature matrix and all non-uniform values ​​to summarize the health status and performance of the system.

[0041] The beneficial effects of the above technical solution are: obtaining multiple historical feature matrices, analyzing the real-time feature matrix of real-time operation data and multiple historical feature matrices, determining whether the real-time operation data is uneven data, being able to quickly identify uneven data, reducing the false alarm rate, and providing a data basis for determining the deviation matrix and the second uneven matrix.

[0042] Embodiment 5: An embodiment of the present invention provides an operation non-uniform analysis method for a driving circuit, which determines the non-uniform value of each feature of a parameter feature vector of each operating parameter based on a first specified number of historical feature matrices, a sequence label of each historical feature matrix, and a real-time feature matrix, and determines a first non-uniform matrix of real-time operation data, including: Determine an uneven value of each feature in a parameter feature vector for each running parameter based on all historical feature matrices, sequence labels of each historical feature matrix, and a real-time feature matrix; in, represents the uneven value of the jth feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix, represents the third decay parameter of the sequence label based on the historical feature matrix, The eigenvalue of the historical feature matrix of the kth sequence label for the jth feature in the parameter feature vector of the i-th running parameter, represents the eigenvalue of the jth feature in the parameter feature vector of the i-th operating parameter in the real-time feature matrix, N1 represents the number of historical feature matrices, represents the first attenuation parameter of the j-th column eigenvalue in the real-time feature matrix, represents the second attenuation parameter of the jth column eigenvalue in the real-time feature matrix, N2 represents the number of eigenvalues ​​in the parameter eigenvector, represents the number of operating parameters in the real-time feature matrix, Indicates adjustment parameters; Compare the uneven value of each feature of the parameter feature vector of each operating parameter with a preset uneven threshold value corresponding to each feature; If there is an uneven value of any feature in the parameter feature vector of the operating parameter that is greater than an uneven threshold of the corresponding feature, determining an uneven vector of the operating parameter based on the uneven values ​​of all features in the parameter feature vector of the operating parameter; in, represents the non-uniform vector based on the i-th operating parameter in the real-time feature matrix, They respectively represent the uneven values ​​of the 1st feature and the N2th feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix; otherwise, adjusting the non-uniform values ​​of all characteristics of the operating parameter to 0, and determining the non-uniform vector of the operating parameter; If the non-uniform values ​​in the non-uniform vectors of all operating parameters in the real-time feature matrix are all 0, it is determined that the first non-uniform matrix of the real-time operating data is empty; Otherwise, a first non-uniform matrix of real-time operation data is determined based on the non-uniform vectors of all operation parameters: ; in, , They respectively represent the uneven vectors based on the 1st operating parameter, the i-th operating parameter, and the N3-th operating parameter in the real-time feature matrix.

[0043] In this embodiment, The smaller the number of columns j representing the eigenvalues ​​in the real-time feature matrix, the more important the eigenvalues ​​in the j-th column are in the real-time feature matrix.

[0044] In this embodiment, The larger the sequence label representing the historical feature matrix, the greater the impact on the uneven value of each feature in the parameter feature vector of each operating parameter in the real-time feature matrix.

[0045] In this embodiment, represents the uneven value of the jth feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix; In this embodiment, Represents the standard deviation of the jth feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix.

[0046] In this embodiment, the uneven value of each feature of each operating parameter is compared with the corresponding preset uneven threshold one by one. If the uneven value of any feature exceeds the threshold, it indicates that the feature is uneven. Based on the uneven values ​​of all features, the uneven vector of the operating parameter is determined. If the uneven values ​​of all features do not exceed the threshold, it means that the feature is normal. The uneven values ​​of all features are adjusted to 0, and the uneven vector of the operating parameter is re-determined as an all-zero vector.

[0047] In this embodiment, the inhomogeneous vectors of all operating parameters are checked. If the inhomogeneous values ​​of all operating parameters are 0, the first inhomogeneous matrix of the real-time operating data is empty, indicating that the system operation is stable; if there are operating parameters whose inhomogeneous values ​​are not 0, the first inhomogeneous matrix is ​​constructed according to these inhomogeneous vectors to characterize the inhomogeneous state of the current system.

[0048] The beneficial effects of the above technical solution are as follows: based on a first specified number of historical feature matrices, the sequence labels of each historical feature matrix and the real-time feature matrix, the uneven value of each feature of the parameter feature vector of each operating parameter is determined, and the first uneven matrix of the real-time operating data is determined, which can refine the uneven detection and provide parameter-level and global comprehensive judgment capabilities, significantly improving the accuracy and intelligence level of real-time operating data analysis, and helping to quickly identify potential unevenness and optimize system operating efficiency.

[0049] Embodiment 6: The embodiment of the present invention provides a method for analyzing uneven operation of a driving circuit, which determines a deviation matrix and a second uneven matrix of the real-time operation data for real-time operation data that is uneven data, including: Performing a second extraction of a second specified number of historical feature matrices in which the first uneven matrix is ​​empty in the historical analysis report of a second specified number of specified time periods before the specified time period of the real-time running data of the uneven data; Determine a fitting feature matrix based on a second specified number of historical feature matrices extracted from the second step; Determine a deviation matrix of real-time operation data based on the fitting feature matrix and the real-time feature matrix; All non-uniform values ​​of 0 in the non-uniform vectors of all operating parameters in the first non-uniform matrix are adjusted based on the deviation matrix, a second non-uniform matrix is ​​determined based on the adjusted first non-uniform matrix, and the operating parameters of the adjusted non-uniform vectors in the second non-uniform matrix are marked.

[0050] In this embodiment, historical data before a specified time period is analyzed, and a second specified number of historical feature matrices in which the uneven matrix is ​​empty (ie, there is no significant unevenness in these periods) are selected as reference data.

[0051] In this embodiment, a fitting feature matrix is ​​constructed based on the historical feature matrices extracted in the second step. The fitting feature matrix represents a standard state and is used to compare the current real-time data with the standard state in subsequent analysis to determine the degree of deviation.

[0052] In this embodiment, a deviation matrix is ​​calculated using the fitted characteristic matrix and the current real-time characteristic matrix. The deviation matrix quantifies the degree of non-uniformity of the real-time data relative to the standard state.

[0053] In this embodiment, for the deviation matrix, all the non-uniform values ​​(indicating uniformity) of which the non-uniform values ​​are 0 in the first non-uniform matrix are updated, and the deviation values ​​in the deviation matrix are assigned corresponding non-uniform values ​​to form an adjusted first non-uniform matrix.

[0054] In this embodiment, the second non-uniform matrix is ​​generated by the adjusted first non-uniform matrix, providing a more accurate non-uniform representation.

[0055] The beneficial effects of the above technical solution are: for real-time operating data that is uneven data, the deviation matrix of the real-time operating data and the second uneven matrix are determined, dynamic adaptive adjustment of uneven detection can be achieved, misjudgment caused by fixed thresholds can be avoided, the accuracy and sensitivity of real-time monitoring can be effectively improved, and the accuracy of uneven detection can be optimized.

[0056] Embodiment 7: The embodiment of the present invention provides a method for analyzing uneven operation of a driving circuit, and determines the uneven level of each operating parameter in a second uneven matrix, including: A level division rule is determined based on the second non-uniform matrix, and a non-uniform level of each operating parameter in the non-uniform vector is determined based on the level division rule, wherein the non-uniform level includes a first non-uniform level, a second non-uniform level, a third non-uniform level and a fourth non-uniform level.

[0057] In this embodiment, each non-uniformity level corresponds to a different non-uniformity severity, which helps determine the priority processing order of different operating parameters.

[0058] The beneficial effects of the above technical solution are as follows: the unevenness level of each operating parameter in the second uneven matrix is ​​determined, and automatic grading of unevenness levels is achieved, which can effectively improve the efficiency of unevenness analysis and processing, and enhance the operational reliability and safety of the system.

[0059] Embodiment 8: An embodiment of the present invention provides an operation non-uniform analysis method for a driving circuit, determining a level division rule based on a second non-uniform matrix, and determining the non-uniform level of each operation parameter in the non-uniform vector based on the level division rule, including: For an uneven vector in the second uneven matrix, where the uneven value of a feature included in the uneven vector is greater than the uneven threshold of the corresponding feature and the number of features exceeds the first number threshold, determining the operating parameter of the uneven vector to be a first uneven level; For an uneven vector in the second uneven matrix, where the uneven value of a feature contained in the uneven vector is greater than the uneven threshold of the corresponding feature and the number of features is between the second number threshold and the first number threshold, determining the operating parameter of the uneven vector to be a second uneven level; Determining the operating parameter marked in the second non-uniform matrix as a fourth non-uniform level; The operating parameters of all the non-uniform vectors in the second non-uniform matrix except the non-uniform vectors of the operating parameters of the first non-uniform level, the second non-uniform level and the fourth non-uniform level are determined as the third non-uniform level.

[0060] In this embodiment, in the second uneven matrix, features contained in all uneven vectors are found, whose uneven values ​​exceed the uneven threshold value of the feature. If the number of these features exceeds the first number threshold, the operating parameter corresponding to the uneven vector is divided into the first uneven level, indicating that the parameter is severely uneven.

[0061] In this embodiment, in the second uneven matrix, if the number of features in the uneven vector that exceeds the uneven threshold is between the second number threshold and the first number threshold, the corresponding operating parameter is classified into the second uneven level, indicating that the parameter has moderate unevenness.

[0062] In this embodiment, the operating parameters marked in the second non-uniformity matrix are directly defined as the fourth non-uniformity level, indicating low non-uniformity.

[0063] In this embodiment, in addition to the above levels, the non-uniform vectors of other operating parameters are classified as the third non-uniform level, indicating that these parameters are slightly non-uniform or in a normal state.

[0064] The beneficial effects of the above technical solution are as follows: based on the second uneven matrix, the level division rules are determined, and based on the level division rules, the uneven level of each operating parameter in the uneven vector is determined, which can accurately distinguish the uneven levels of different degrees, realize the intelligent classification of the unevenness, enable the system to give priority to severe unevenness, optimize resource allocation and safety management, and significantly improve the efficiency and real-time performance of uneven processing.

[0065] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0066] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0067] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for analyzing uneven operation of a driving circuit, characterized in that: include: 101: collecting real-time operation data of all operation parameters of the driving circuit in real time, and analyzing the real-time operation data to determine a real-time characteristic matrix; 102: Acquire multiple historical feature matrices, analyze the real-time feature matrix of the real-time operation data and the multiple historical feature matrices, and determine whether the real-time operation data is uneven data; 103: for the real-time operation data being non-uniform data, determining a deviation matrix and a second non-uniform matrix of the real-time operation data, and determining a non-uniform level of each operation parameter in the second non-uniform matrix; 104 : For the real-time operation data that is non-uniform data, generate a non-uniform report based on the real-time feature matrix, the first non-uniform matrix, and non-uniform levels of all operation parameters in the non-uniform vector.

2. A method for analyzing uneven operation of a driving circuit according to claim 1, characterized in that: Real-time acquisition of real-time operating data of all operating parameters of the drive circuit, including: Analyze the characteristics of the driving circuit to determine the collection requirements of each operating parameter data, and determine the corresponding sensor group based on the collection requirements of the operating parameter data; Analyze potential non-uniformity sources of the driving circuit, determine multiple acquisition locations for acquiring each operating parameter data and the location priority of each acquisition location; Based on all the collection positions for collecting the operating parameter data and the position priority of each collection position, a sensor group corresponding to the operating parameter data is installed on the driving circuit; Based on the sensor group, the operating parameter data of each operating parameter within a specified time period is collected to determine the real-time operating data.

3. A method for analyzing uneven operation of a driving circuit according to claim 2, characterized in that: Analyze real-time operation data to determine real-time feature matrix, including: Preprocessing the operating parameter data of each operating parameter in the real-time operating data, performing feature extraction on the preprocessed operating parameter data of each operating parameter, and determining a parameter feature vector of the operating parameter corresponding to each operating parameter data; A real-time feature matrix of the real-time operation data is determined based on the parameter feature vectors of all the operation parameters.

4. The method for analyzing uneven operation of a driving circuit according to claim 1, characterized in that: Acquire multiple historical feature matrices, analyze the real-time feature matrix of the real-time operation data and the multiple historical feature matrices, and determine whether the real-time operation data is uneven data, including: Performing a first extraction of a historical feature matrix in a historical analysis report of a first specified number of specified time periods before a specified time period of the real-time operation data; performing a first sorting on the first specified number of historical feature matrices extracted first based on a specified time period of each historical feature matrix, and determining a sequence label of each historical feature matrix extracted first based on the first sorting result; Based on a first specified number of historical feature matrices, sequence labels of each historical feature matrix, and a real-time feature matrix, determining an uneven value of each feature of a parameter feature vector of each operating parameter, and determining a first uneven matrix of real-time operating data; If the first non-uniform matrix of the real-time operation data is not empty, determining that the real-time operation data is non-uniform data; Otherwise, the real-time operation data is determined to be uniform operation data, and a uniform report is generated based on the real-time feature matrix of the real-time operation data and the non-uniform value of each feature in the parameter feature vector of each operation parameter.

5. A method for analyzing uneven operation of a driving circuit according to claim 4, characterized in that: Based on a first specified number of historical feature matrices, sequence labels of each historical feature matrix, and a real-time feature matrix, determining an uneven value of each feature of a parameter feature vector of each operating parameter, and determining a first uneven matrix of real-time operating data, including: Determine an uneven value of each feature in a parameter feature vector for each running parameter based on all historical feature matrices, sequence labels of each historical feature matrix, and a real-time feature matrix; in, represents the uneven value of the jth feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix, represents the third decay parameter of the sequence label based on the historical feature matrix, The eigenvalue of the historical feature matrix of the kth sequence label for the jth feature in the parameter feature vector of the i-th running parameter, represents the eigenvalue of the jth feature in the parameter feature vector of the i-th operating parameter in the real-time feature matrix, N1 represents the number of historical feature matrices, represents the first attenuation parameter of the j-th column eigenvalue in the real-time feature matrix, represents the second attenuation parameter of the jth column eigenvalue in the real-time feature matrix, N2 represents the number of eigenvalues ​​in the parameter eigenvector, represents the number of operating parameters in the real-time feature matrix, Indicates adjustment parameters; Compare the uneven value of each feature of the parameter feature vector of each operating parameter with a preset uneven threshold value corresponding to each feature; If there is an uneven value of any feature in the parameter feature vector of the operating parameter that is greater than an uneven threshold of the corresponding feature, determining an uneven vector of the operating parameter based on the uneven values ​​of all features in the parameter feature vector of the operating parameter; in, represents the non-uniform vector based on the i-th operating parameter in the real-time feature matrix, They respectively represent the uneven values ​​of the 1st feature and the N2th feature in the parameter feature vector based on the i-th operating parameter in the real-time feature matrix; otherwise, adjusting the non-uniform values ​​of all characteristics of the operating parameter to 0, and determining the non-uniform vector of the operating parameter; If the non-uniform values ​​in the non-uniform vectors of all operating parameters in the real-time feature matrix are all 0, it is determined that the first non-uniform matrix of the real-time operating data is empty; Otherwise, a first non-uniform matrix of real-time operation data is determined based on the non-uniform vectors of all operation parameters: ; in, , They respectively represent the uneven vectors based on the 1st operating parameter, the i-th operating parameter, and the N3-th operating parameter in the real-time feature matrix.

6. A method for analyzing uneven operation of a driving circuit according to claim 5, characterized in that: For real-time operation data that is non-uniform data, determining a deviation matrix of the real-time operation data and a second non-uniform matrix includes: Performing a second extraction of a second specified number of historical feature matrices in which the first uneven matrix is ​​empty in the historical analysis report of a second specified number of specified time periods before the specified time period of the real-time running data of the uneven data; Determine a fitting feature matrix based on a second specified number of historical feature matrices extracted from the second step; Determine a deviation matrix of real-time operation data based on the fitting feature matrix and the real-time feature matrix; All non-uniform values ​​of 0 in the non-uniform vectors of all operating parameters in the first non-uniform matrix are adjusted based on the deviation matrix, a second non-uniform matrix is ​​determined based on the adjusted first non-uniform matrix, and the operating parameters of the adjusted non-uniform vectors in the second non-uniform matrix are marked.

7. The method for analyzing uneven operation of a driving circuit according to claim 1, characterized in that: And determining the unevenness level of each operating parameter in the second unevenness matrix, including: A level division rule is determined based on the second non-uniform matrix, and a non-uniform level of each operating parameter in the non-uniform vector is determined based on the level division rule, wherein the non-uniform level includes a first non-uniform level, a second non-uniform level, a third non-uniform level and a fourth non-uniform level.

8. A method for analyzing uneven operation of a driving circuit according to claim 7, characterized in that: Determining a level division rule based on the second non-uniform matrix, and determining the non-uniform level of each operating parameter in the non-uniform vector based on the level division rule, including: For an uneven vector in the second uneven matrix, where the uneven value of a feature included in the uneven vector is greater than the uneven threshold of the corresponding feature and the number of features exceeds the first number threshold, determining the operating parameter of the uneven vector to be a first uneven level; For an uneven vector in the second uneven matrix, where the uneven value of a feature contained in the uneven vector is greater than the uneven threshold of the corresponding feature and the number of features is between the second number threshold and the first number threshold, determining the operating parameter of the uneven vector to be a second uneven level; Determining the operating parameter marked in the second non-uniform matrix as a fourth non-uniform level; The operating parameters of all the non-uniform vectors in the second non-uniform matrix except the non-uniform vectors of the operating parameters of the first non-uniform level, the second non-uniform level and the fourth non-uniform level are determined as the third non-uniform level.

Images