Method and device for detecting abnormal voltage of flexible DC converter valve submodule

Abstract

The present invention provides a method and device for detecting voltage anomalies in a flexible direct current (DC) converter valve submodule. The method comprises: obtaining discrete voltage data of the flexible direct current (DC) converter valve submodule, performing distribution verification on the discrete voltage data, and obtaining a distribution verification result; if the distribution verification result is a passed verification, performing a preliminary fitting on the discrete voltage data by adjusting the length of the time period to determine the inflection point position; using a maximum likelihood estimation algorithm and a skewed distribution algorithm to determine an abnormality judgment limit value according to the inflection point position, and using a preset sliding window to perform continuous calculation of the time period to obtain a continuous calculation result; performing voltage anomaly detection on the flexible direct current (DC) converter valve submodule according to the abnormality judgment limit value and the continuous calculation result to obtain a voltage anomaly detection result. The present invention uses the skewed distribution and sliding window method to identify potential persistent voltage anomaly problems in the flexible direct current (DC) converter valve submodule, facilitates warnings for potential problems, and effectively prevents failures in the flexible direct current (DC) converter valve submodule.

Description

Technical Field

[0001] The present invention relates to the technical field of flexible direct current converter valves, and in particular to a method and device for detecting voltage anomalies in submodules of a flexible direct current converter valve. Background Art

[0002] Diagnosing abnormal voltage in flexible direct current (HVDC) converter valve submodules is a key area of ​​power electronics, primarily used in flexible direct current (HVDC) systems within power electronics devices. This technology addresses the operational safety and stability of HVDC systems, as well as the need for power equipment status monitoring and fault diagnosis.

[0003] The High Voltage Direct Current (HVDC) system is a system that uses direct current (DC) transmission technology to transmit high-capacity electricity over long distances. One of its key components is the High Voltage Direct Current (HVDC) converter, and the High Voltage Direct Current (HVDC) valve submodule is one of the components of the converter.

[0004] With the increasing diversification of electric energy sources, HVDC systems offer unique advantages in long-distance, high-capacity power transmission, leading to their increasing application in power transmission. The converter in an HVDC system is the core device connecting the AC and DC systems, and its stability and reliability are crucial to the operation of the entire system. Therefore, to ensure the safe operation of HVDC systems, timely and accurate detection and diagnosis of abnormal voltage faults in the flexible DC converter submodules is crucial to avoid potential system accidents and downtime losses.

[0005] Currently, due to the recent development of HVDC Flexible projects, limited data accumulation, and the challenging nature of anomaly diagnosis, most supervised algorithms are largely unusable. Furthermore, due to the poor interpretability of unsupervised learning algorithms, such as clustering algorithms, they are ineffective in diagnosing similar anomalies in power systems. In summary, current algorithms for HVDC Flexible transmission are limited to indicator-based condition monitoring and verification algorithms based on power mechanism formulas. However, these algorithms are only effective during major faults, which are generally short-lived and unable to generate meaningful action. Key algorithms for alarm and early warning are lacking for HVDC Flexible transmission. Summary of the Invention

[0006] In response to the problems existing in the prior art, the main purpose of the embodiments of the present invention is to provide a method and device for detecting voltage anomalies in a flexible direct current converter valve submodule, so as to accurately detect potential continuous abnormal voltage problems in the flexible direct current converter valve submodule and effectively prevent failures of the flexible direct current converter valve submodule.

[0007] To achieve the above objectives, an embodiment of the present invention provides a method for detecting abnormal voltage in a flexible direct current converter valve submodule, the method comprising:

[0008] Obtain discrete voltage data of the flexible DC converter valve submodule, perform distribution verification on the discrete voltage data, and obtain distribution verification results;

[0009] If the distribution verification result is passed, the discrete voltage data is preliminarily fitted by adjusting the time period to determine the inflection point position;

[0010] Using the maximum likelihood estimation algorithm and the skewed distribution algorithm, the abnormality judgment limit value is determined according to the inflection point position, and the preset sliding window is used to perform continuous calculations over the time period to obtain continuous calculation results;

[0011] According to the abnormal judgment limit value and the continuous calculation results, the voltage abnormality detection is performed on the flexible DC converter valve submodule to obtain the voltage abnormality detection result.

[0012] Optionally, in one embodiment of the present invention, performing distribution verification on discrete voltage data to obtain a distribution verification result includes:

[0013] Perform occurrence statistics on discrete voltage data and determine the significance level test value based on the obtained occurrence count;

[0014] Compare the significance level test value with the preset significance threshold, and obtain the distribution verification result based on the comparison result.

[0015] Optionally, in one embodiment of the present invention, using a maximum likelihood estimation algorithm and a skewed distribution algorithm to determine an abnormality judgment limit value according to the inflection point position includes:

[0016] The maximum likelihood estimation algorithm and the skewed distribution algorithm are used to calculate the skewed probability density function of the data in each time period according to the inflection point position;

[0017] According to the skewed probability density function, the abnormality judgment limit value is determined.

[0018] Optionally, in one embodiment of the present invention, using a maximum likelihood estimation algorithm and a skewed distribution algorithm to calculate the skewed probability density function of the data in each time period according to the inflection point position includes:

[0019] According to the inflection point position, the optimal time period data is determined, and the maximum likelihood estimation algorithm is used to fit the optimal time period data;

[0020] The skewed distribution algorithm is used to perform skewed distribution processing on the optimal time period data after fitting, and the skewed probability density function of the data in each time period is obtained.

[0021] An embodiment of the present invention further provides a device for detecting abnormal voltage of a flexible direct current converter valve submodule, the device comprising:

[0022] A distribution verification module is used to obtain discrete voltage data of the flexible direct current converter valve submodule, perform distribution verification on the discrete voltage data, and obtain distribution verification results;

[0023] The inflection point location module is used to perform a preliminary fit on the discrete voltage data by adjusting the time period if the distribution verification result is passed, and determine the inflection point location;

[0024] A judgment limit value module is used to determine the abnormality judgment limit value according to the inflection point position using the maximum likelihood estimation algorithm and the skewed distribution algorithm, and to perform continuous calculations over a time period using a preset sliding window to obtain continuous calculation results;

[0025] The anomaly detection module is used to perform voltage anomaly detection on the flexible direct current converter valve submodule based on the anomaly judgment limit value and the continuous calculation result to obtain the voltage anomaly detection result.

[0026] Optionally, in one embodiment of the present invention, the distribution verification module includes:

[0027] A significance test unit is used to count the number of occurrences of discrete voltage data and determine a significance level test value based on the obtained number of occurrences;

[0028] The distribution verification unit is used to compare the significance level test value with a preset significance threshold, and obtain the distribution verification result based on the comparison result.

[0029] Optionally, in one embodiment of the present invention, the threshold value determination module includes:

[0030] The probability density unit is used to calculate the skewed probability density function of the data in each time period according to the inflection point position using the maximum likelihood estimation algorithm and the skewed distribution algorithm;

[0031] The judgment limit value unit is used to determine the abnormal judgment limit value according to the skewed probability density function.

[0032] Optionally, in one embodiment of the present invention, the probability density unit includes:

[0033] The data fitting subunit is used to determine the optimal time period data according to the inflection point position and fit the optimal time period data using the maximum likelihood estimation algorithm;

[0034] The probability density subunit is used to perform skewed distribution processing on the fitted optimal time period data using a skewed distribution algorithm to obtain a skewed probability density function of the data in each time period.

[0035] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the above method is implemented when the processor executes the program.

[0036] The present invention also provides a computer-readable storage medium storing a computer program for executing the above method on a computer.

[0037] The present invention also provides a computer program product, comprising a computer program / instructions, which implement the steps of the above method when executed by a processor.

[0038] The present invention uses the discrete voltage data of the flexible direct current converter valve submodule and, based on the potential distribution relationship in the discrete voltage data, uses a relevant algorithm to find out the potential regularity, and uses the skewed distribution and sliding window method to find out the potential continuous voltage anomaly problem of the flexible direct current converter valve submodule, so as to facilitate the alarm of potential problems and effectively prevent the flexible direct current converter valve submodule from failing. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for describing 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 paying any creative work.

[0040] Figure 1 This is a flow chart of a method for detecting abnormal voltage in a flexible direct current converter valve submodule according to an embodiment of the present invention;

[0041] Figure 2 A flow chart showing how to obtain a distribution verification result in an embodiment of the present invention;

[0042] Figure 3 This is a flow chart of determining an abnormality judgment limit value in an embodiment of the present invention;

[0043] Figure 4 A flow chart of obtaining a skewed probability density function in an embodiment of the present invention;

[0044] Figure 5 Schematic diagram of the single valve tower time length and the standard deviation of the fitting distribution in an embodiment of the present invention;

[0045] Figures 6A-6C This is a schematic diagram of the change in the distribution when a submodule is bypassed in an embodiment of the present invention;

[0046] Figures 7A-7C This is a schematic diagram of the change in the distribution when a submodule is bypassed in another embodiment of the present invention;

[0047] Figure 8 This is a structural diagram of a device for detecting abnormal voltage in a flexible DC converter valve submodule according to an embodiment of the present invention;

[0048] Figure 9 This is a schematic diagram of the structure of a distributed verification module in an embodiment of the present invention;

[0049] Figure 10 Schematic diagram of the structure of the threshold value determination module in an embodiment of the present invention;

[0050] Figure 11 Schematic diagram of the structure of a probability density unit in an embodiment of the present invention;

[0051] Figure 12 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] Embodiments of the present invention provide a method and device for detecting voltage anomaly in a flexible direct current (DC) converter valve submodule.

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] The commonly used means in existing technologies are as follows: Sensor technology: Utilize high-precision sensors to monitor the voltage of the flexible DC converter valve submodule in real time to achieve sensitive detection of voltage anomalies. Data analysis and algorithms: Use advanced data analysis technologies and algorithms to process and analyze the monitored voltage data in real time to identify abnormal patterns and fault characteristics. Artificial intelligence and machine learning: Use artificial intelligence and machine learning technologies to establish a model for voltage anomaly diagnosis to achieve automatic identification and classification of different fault types. Online monitoring system: Build an online monitoring system that integrates sensors, data analysis, and diagnostic algorithms to achieve real-time monitoring and abnormal diagnosis of the voltage of the flexible DC converter valve submodule.

[0055] However, in DC transmission projects, data storage is typically enormous. If data from every collector at every frequency were recorded, the database would grow exponentially, and data storage usage would become a significant issue. Therefore, most DC transmission projects do not retain all data, but only critical data. For example, if data such as submodule voltage and temperature were collected at a fixed frequency per module, the large number of submodules in each bridge arm—typically thousands in total—would create a massive amount of data to store. Therefore, data is typically divided into valve towers, with each tower storing only the minimum and maximum values, rather than all data. This extreme compression of data dimensions and the shift from relatively continuous data to highly discrete data complicates algorithm execution.

[0056] like Figure 1 The flowchart of a method for detecting voltage anomalies in a flexible DC converter valve submodule according to an embodiment of the present invention is shown. The method for detecting voltage anomalies in a flexible DC converter valve submodule provided by the embodiment of the present invention includes, but is not limited to, a computer. The present invention utilizes discrete voltage data from the flexible DC converter valve submodule, employs a correlation algorithm based on the potential distribution relationships within the discrete voltage data, and uses a skewed distribution and sliding window approach to identify potential persistent voltage anomalies in the flexible DC converter valve submodule, thereby facilitating an alert for potential problems and effectively preventing failures in the flexible DC converter valve submodule. The method shown in the figure includes:

[0057] Step S1, obtaining discrete voltage data of the flexible direct current converter valve submodule, performing distribution verification on the discrete voltage data, and obtaining a distribution verification result;

[0058] Step S2: If the distribution verification result is passed, the discrete voltage data is preliminarily fitted by adjusting the time period to determine the inflection point position;

[0059] Step S3, using the maximum likelihood estimation algorithm and the skewed distribution algorithm to determine the abnormality judgment limit value according to the inflection point position, and using a preset sliding window to perform continuous calculations over a time period to obtain continuous calculation results;

[0060] Step S4: performing voltage anomaly detection on the flexible DC converter valve submodule according to the abnormality judgment limit value and the continuous calculation result to obtain a voltage anomaly detection result.

[0061] Among them, for the discrete voltage data of the flexible direct current converter valve submodule, a preset number of discrete voltage data of each submodule are collected, for example, five data of the maximum and minimum values ​​at each moment are obtained. To determine whether the collected discrete voltage data is usable, it is necessary to perform preliminary regularity verification on the discrete voltage data, that is, distribution verification, to obtain the distribution verification result.

[0062] As an embodiment of the present invention, Figure 2 As shown, the discrete voltage data is distributed and verified, and the distribution verification results include:

[0063] Step S11, performing a count of the number of occurrences of discrete voltage data, and determining a significance level test value based on the obtained number of occurrences;

[0064] Step S12: comparing the significance level test value with a preset significance threshold, and obtaining a distribution verification result based on the obtained comparison result.

[0065] Among them, the number of occurrences of all data is counted, and it is determined that the number of occurrences of each valve tower submodule in the data in the long-term data is not much different, as shown in Table 1.

[0066] Table 1

[0067]

[0068]

[0069] Furthermore, for a period of time, the distribution verification of the statistical data of the same valve tower is conducted to determine whether the conditions are met. Specifically, based on the number of occurrences obtained, the significance level test value is determined in a conventional manner, and the obtained significance level test value is compared with the preset significance threshold. Based on the comparison result, the distribution verification result is determined. Specifically, for example, if 14 days are used, the α value of the significance level test for fitting the Gaussian distribution will be around 10%. If 30 days are used, the α value in the significance level test is only 5% (calculated using the Andrew test), and the pvalue values ​​in the chi-square test are all above 0.7. The obtained distribution verification result satisfies the distribution fitting requirements. In addition, if the distribution of discrete voltage data is verified according to different methods, both the pvalue value and the significance level α of the verification means need to meet the requirements.

[0070] Furthermore, by adjusting the length of time, a fitting attempt is made to the skewed distribution, i.e., a preliminary fitting is performed. The standard deviation value after fitting is observed to see whether the value tends to be stable after a certain period of time, thereby determining the inflection point.

[0071] As an embodiment of the present invention, Figure 3 As shown in the figure, the maximum likelihood estimation algorithm and the skewed distribution algorithm are used to determine the abnormality judgment limit value according to the inflection point position, including:

[0072] Step S31, using the maximum likelihood estimation algorithm and the skewed distribution algorithm to calculate the skewed probability density function of the data in each time period according to the inflection point position;

[0073] Step S32: determining an abnormality judgment limit value according to the skewed probability density function.

[0074] Among them, after determining the inflection point position, the optimal time period data can be determined at the same time, that is, the time period in which the standard deviation after the inflection point position decreases gradually slows down and tends to be stable is taken as the optimal time period, and the data corresponding to the optimal time period is the optimal time period data.

[0075] In this embodiment, if Figure 4 As shown, using the maximum likelihood estimation algorithm and the skewed distribution algorithm, the skewed probability density function of the data in each time period is calculated according to the inflection point position, including:

[0076] Step S311, determining the optimal time period data according to the inflection point position, and fitting the optimal time period data using the maximum likelihood estimation algorithm;

[0077] Step S312 , using a skewed distribution algorithm to perform skewed distribution processing on the fitted optimal time period data to obtain a skewed probability density function of the data in each time period.

[0078] Among them, the maximum likelihood estimation algorithm is used to fit the distribution of the optimal time period data, and then the skewed distribution algorithm is used to perform skewed distribution processing on the fitted optimal time period data to obtain the skewed probability density function of the data in each time period.

[0079] Specifically, a skewed distribution refers to a frequency distribution of variable values ​​that deviates from symmetry. The degree of deviation can be expressed as a coefficient of deviation. In a skewed distribution, if the shape of the probability density function is tilted to the left, it is called a left-skewed distribution (negatively skewed distribution), and if it is tilted to the right, it is called a right-skewed distribution (positively skewed distribution).

[0080] Furthermore, based on the skewed probability density function, we select a threshold value for determining anomalies, namely the anomaly judgment threshold. If the threshold is exceeded, it is considered that the submodule has exceeded the distribution during this period. However, the distribution is likely due to random probability distribution. To reduce this effect, we use a sliding window to continuously calculate the time period when determining anomalies.

[0081] Specifically, a preset sliding window is used to perform continuous calculations over time periods, generating a set of results (with repeated times within each time period). The continuity and duration of a module's exceeding the distribution impact within the continuous calculation results are used to determine whether submodule voltage instability has occurred. This allows for voltage anomaly detection of the flexible DC converter valve submodule, resulting in a voltage anomaly detection result that includes both normal and abnormal results.

[0082] In a specific embodiment of the present invention, the biased distribution-based diagnosis of voltage anomalies in flexible DC converter valve submodules can be linked to anomaly detection and fault diagnosis technologies in power systems. In the field of power systems, there are some similarities or analogies with the diagnosis of voltage anomalies in flexible DC converter valve submodules. The following are some related existing technologies:

[0083] Statistical analysis method: Distribution-based anomaly detection methods usually use statistical analysis techniques such as mean, variance, Gaussian distribution, etc. Then, by monitoring the actual voltage data, check whether it deviates significantly from the historical value to identify anomalies.

[0084] Machine learning technology: Machine learning algorithms, such as support vector machines (SVMs) and neural networks, are used to train and classify voltage data from flexible DC converter valve submodules to identify normal and abnormal conditions. These algorithms can learn patterns of voltage anomalies from large amounts of data and perform classification and diagnosis during real-time monitoring.

[0085] Bayesian Network: A Bayesian network is a probabilistic graphical model that describes the dependencies between the voltage of the flexible direct current converter submodule and other related variables. Bayesian network inference enables anomaly detection and diagnosis, while taking into account the complex relationships between multiple factors.

[0086] State Estimation and Kalman Filtering: State estimation techniques and Kalman filtering methods use system dynamic models and measurement data to estimate the actual state of the flexible DC converter valve submodule voltage and compare it with the expected state. Deviations between the actual state and the expected state may indicate an abnormal situation.

[0087] These technologies can all be used to diagnose abnormal voltages in flexible DC converter valve submodules, but they require appropriate adjustments and optimization based on specific system characteristics and data availability. Furthermore, since flexible DC technology has only been around for a short time and data accumulation is limited, the practical application of these methods remains to be verified. Therefore, when selecting the most appropriate technology, it is necessary to consider system complexity, data availability, and actual application requirements.

[0088] Therefore, there are currently few fault diagnosis algorithms for flexible DC converter valves, and most of them are based on data indicator threshold judgment methods based on mechanisms and defect manifestations. No other fault diagnosis methods based on flexible DC converter valve data and statistical algorithms have been found.

[0089] In this embodiment, for the submodule voltage of discrete voltage data, only five maximum and minimum values ​​are obtained at each moment. To determine whether the data is usable, preliminary regularity verification of the data is required, and the method is as follows.

[0090] (1) Count the number of occurrences of all data and make sure that the number of occurrences of each submodule of the same valve tower in the long-term data is not much different. The statistical example is shown in Table 1.

[0091] (2) Verify the distribution of the statistical data of the same valve tower over a period of time to see if it meets the requirements. The verification example is as follows. If 14 days are used, the α value of the significance level test for fitting the Gaussian distribution will be around 10%. If 30 days are used, the α value in the significance level test is only 5% (calculated using the Andrew test). The p-value values ​​in the chi-square test are all above 0.7, which meets the distribution fitting requirements. Verify the original data distribution according to different methods so that the p-value value and significance level α of the verification method meet the requirements.

[0092] (3) Try to fit the skewed distribution by adjusting the length of time, and observe the value of the standard deviation after fitting to see whether the value tends to be stable after a certain period of time, such as Figure 5 As shown. Figure 5 It can be seen that 15 days after September 15 (the optimal time period), the rate of decline of the standard deviation gradually slows down and tends to be stable. This time period is the inflection point.

[0093] After the above verification conditions are met, the distribution of the original discrete voltage data can be fitted by using the maximum likelihood estimation method based on the obtained optimal time period (the inflection point position obtained by the above (3)).

[0094] Among them, maximum likelihood estimation means that given a probability distribution D, its probability density function (continuous distribution) or probability aggregation function (discrete distribution) is assumed to be f D , and a distribution parameter θ, a sample with n values ​​X1,X2,...,X can be drawn from this distribution n , by using f D , we can calculate its probability.

[0095] P=(x1,x2,..,x n )=f D (x1,x2,...,x n |θ) (1)

[0096] But we may not know the value of θ, even though we know that the sampled data comes from distribution D. So how can we estimate θ? A natural idea is to draw a sample X1, X2, ..., X with n values ​​from this distribution. n , and then use these sampled data to estimate θ.

[0097] Once obtained, an estimate of θ can be found from it. Maximum likelihood estimation will find the most likely value of θ (that is, among all possible values ​​of θ, find a value that maximizes the "likelihood" of this sample). This method is different from some other estimation methods, such as unbiased estimation of θ. Unbiased estimation does not necessarily output the most likely value, but rather outputs a value of θ that is neither overestimated nor underestimated.

[0098] To implement the maximum likelihood estimation method mathematically, we first need to define the likelihood:

[0099] lik(θ)=f D (x1,x2,..,x n |θ) (2)

[0100] And over all values ​​of θ, this function is maximized. The value that maximizes the likelihood is called the maximum likelihood estimate of θ.

[0101] Furthermore, skewed distribution, also known as crooked distribution, refers to the frequency distribution of variable values ​​that deviates from symmetry, and the degree of deviation can be expressed by the deviation coefficient. In a skewed distribution, if the shape of the probability density function tilts to the left, it is called a left-skewed distribution (negatively skewed distribution), and if it tilts to the right, it is called a right-skewed distribution (positively skewed distribution). The tail of a left-skewed distribution is mainly concentrated on the left, while the tail of a right-skewed distribution is concentrated on the right. These two skewed distributions can be expressed by the following formulas:

[0102]

[0103] Using the above methods, we calculate the skewed probability density function of the data for each time period. Based on this probability density function, we select a threshold value (tentatively set at 3σ) to identify anomalies. If this threshold value is exceeded, it is considered that the module has exceeded the distribution during this period. However, this is likely due to random probability distribution. To mitigate this effect, we use a sliding window to perform continuous calculations over time periods when determining anomalies. This results in a set of results (with repeated periods within each time period), known as the continuous calculation results. The consistency and duration of a module's exceeding the distribution in the continuous calculation results are used to determine whether a submodule voltage instability anomaly has occurred.

[0104] Among them, such as Figures 6A-6C As shown in the figure, the distribution changes when a submodule is bypassed (the horizontal axis shows the number of times the number reaches its maximum (or minimum) value in the time period, and the vertical axis shows the number of modules corresponding to this number). As can be seen, number 106 gradually falls out of the distribution due to being bypassed, and can be clearly captured by the method of the present invention.

[0105] Further, such as Figures 7A-7CAnother set of schematic diagrams is shown. It can be seen that No. 69 gradually deviates from the distribution. Compared with the previous example, the value of No. 69 does not deviate too much from other modules, and does not exceed its rated voltage. The present invention can capture emergencies and potential influencing factors that are difficult to find through this means, while meeting the functions of status monitoring and equipment early warning.

[0106] The present invention uses discrete data of the flexible direct current converter valve submodule voltage, and according to the potential distribution relationship in the data, uses an algorithm to find the potential rules and finds the potential persistent problems of the flexible direct current converter valve submodule through a sliding window method, and issues an alarm for potential problems, effectively preventing submodule failures.

[0107] In summary, the present invention uses statistical thinking and anomaly detection algorithms to propose a basis for judging submodule voltage anomalies. Through the above steps, fault warning of flexible direct current submodules is achieved. The specific advantages are as follows:

[0108] 1) Use smaller and more discrete data

[0109] The present invention can use a smaller amount of data (a total of 10 measuring points with a maximum and minimum value) than the total number of sub-modules of each valve tower (generally more than 100 measuring points) to perform calculations, and does not require the use of massive data for anomaly detection, thereby solving the problems of difficult data storage and low algorithm calculation efficiency.

[0110] 2) Fault diagnosis and early warning of flexible direct current converter valves through algorithms

[0111] At present, most algorithms for flexible direct current converter valves are state monitoring or formula mechanism deduction, and there are not many algorithms. In addition, since flexible direct current technology is a new technology that has only been put into practical use recently, there are very few problems with fault cases. The present invention starts from the relationship that should exist between sub-modules, studies the potential relationship of the data, and uses statistical methods to evaluate the distribution of sub-module data, and can simultaneously realize the functions of fault diagnosis and fault warning.

[0112] 3) Effective early warning measures for flexible DC converter valves

[0113] Currently, flexible DC converters are based on state monitoring and formula mechanism derivation, and cannot obtain effective early warning strategies. By studying the potential relationship between data, the present invention can promptly identify the potential imbalance of sub-module voltages and provide timely warnings, which can effectively reduce the damage frequency of sub-module equipment.

[0114] 4) Use self-learning method to self-iterate fault diagnosis model

[0115] During actual operation, the present invention can conduct self-learning based on the continuity of the model's abnormal judgment of data, and continuously strengthen its ability to judge abnormal situations of concern to operation and maintenance personnel through the probability judgment of the original model according to the repair of module abnormal situations, and continuously repair its own handling of misjudgment situations.

[0116] like Figure 8 FIG2 is a schematic diagram of a structure of a device for detecting abnormal voltage of a flexible direct current converter valve submodule according to an embodiment of the present invention. The device shown in the figure includes:

[0117] The distribution verification module 10 is used to obtain discrete voltage data of the flexible direct current converter valve submodule, perform distribution verification on the discrete voltage data, and obtain a distribution verification result;

[0118] The inflection point location module 20 is configured to perform a preliminary fit on the discrete voltage data by adjusting the time period to determine the inflection point location if the distribution verification result is passed.

[0119] The judgment limit value module 30 is used to determine the abnormality judgment limit value according to the inflection point position using the maximum likelihood estimation algorithm and the skewed distribution algorithm, and to perform continuous calculations over a time period using a preset sliding window to obtain continuous calculation results;

[0120] The abnormality detection module 40 is used to perform voltage abnormality detection on the flexible direct current converter valve submodule according to the abnormality judgment limit value and the continuous calculation result to obtain a voltage abnormality detection result.

[0121] As an embodiment of the present invention, Figure 9 As shown, the distribution verification module 10 includes:

[0122] The significance test unit 11 is used to count the number of occurrences of discrete voltage data and determine a significance level test value according to the obtained number of occurrences;

[0123] The distribution verification unit 12 is used to compare the significance level test value with a preset significance threshold, and obtain a distribution verification result based on the obtained comparison result.

[0124] As an embodiment of the present invention, Figure 10 As shown, the threshold value determination module 30 includes:

[0125] The probability density unit 31 is used to calculate the skewed probability density function of the data in each time period according to the inflection point position using the maximum likelihood estimation algorithm and the skewed distribution algorithm;

[0126] The judgment limit value unit 32 is used to determine the abnormality judgment limit value according to the skewed probability density function.

[0127] In this embodiment, if Figure 11As shown, the probability density unit 31 includes:

[0128] The data fitting subunit 311 is used to determine the optimal time period data according to the inflection point position and fit the optimal time period data using the maximum likelihood estimation algorithm;

[0129] The probability density subunit 312 is used to perform skewed distribution processing on the fitted optimal time period data using a skewed distribution algorithm to obtain a skewed probability density function of the data in each time period.

[0130] Based on the same application concept as the aforementioned method for detecting voltage anomalies in a flexible DC converter valve submodule, the present invention also provides the aforementioned device for detecting voltage anomalies in a flexible DC converter valve submodule. Because the principles underlying the device for detecting voltage anomalies in a flexible DC converter valve submodule are similar to those of the method for detecting voltage anomalies in a flexible DC converter valve submodule, the implementation of the device for detecting voltage anomalies in a flexible DC converter valve submodule can be referenced to the implementation of the method for detecting voltage anomalies in a flexible DC converter valve submodule, and any repetitive details will not be repeated.

[0131] The present invention uses the discrete voltage data of the flexible direct current converter valve submodule and, based on the potential distribution relationship in the discrete voltage data, uses a relevant algorithm to find out the potential regularity, and uses the skewed distribution and sliding window method to find out the potential continuous voltage anomaly problem of the flexible direct current converter valve submodule, so as to facilitate the alarm of potential problems and effectively prevent the flexible direct current converter valve submodule from failing.

[0132] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the above method is implemented when the processor executes the program.

[0133] The present invention also provides a computer program product, comprising a computer program / instructions, which implement the steps of the above method when executed by a processor.

[0134] The present invention also provides a computer-readable storage medium storing a computer program for executing the above method on a computer.

[0135] like Figure 12 As shown, the electronic device 600 may further include: a communication module 110, an input unit 120, an audio processor 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily have to include Figure 12 In addition, the electronic device 600 may also include all components shown in Figure 12 For components not shown, reference may be made to the prior art.

[0136] like Figure 12As shown, the central processing unit 100 is sometimes also referred to as a controller or an operation control unit, and may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operations of various components of the electronic device 600 .

[0137] Memory 140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information and may also store programs that execute the relevant information. The CPU 100 may execute the programs stored in memory 140 to implement information storage or processing.

[0138] The input unit 120 provides input to the CPU 100. The input unit 120 may be, for example, a keypad or touch input device. The power supply 170 is used to provide power to the electronic device 600. The display 160 is used to display objects such as images and text. The display may be, for example, an LCD display, but is not limited thereto.

[0139] The memory 140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), or a SIM card. Alternatively, it may be a memory that retains information even when power is off, can be selectively erased, and is provided with more data. Examples of such memory are sometimes referred to as EPROMs. The memory 140 may also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142 for storing application programs and function programs or processes for executing the operations of the electronic device 600 via the central processing unit 100.

[0140] The memory 140 may also include a data storage unit 143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 144 of the memory 140 may include various driver programs for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0141] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via an antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processor 100 to provide input signals and receive output signals, which may be the same as in a conventional mobile communication terminal.

[0142] Based on different communication technologies, multiple communication modules 110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby implementing common telecommunication functions. The audio processor 130 may include any suitable buffer, decoder, amplifier, etc. Furthermore, the audio processor 130 is also coupled to the central processing unit 100, enabling local recording via the microphone 132 and playback of stored audio via the speaker 131.

[0143] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0144] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0145] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0146] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0147] Specific embodiments are used in the present invention to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for detecting abnormal voltage of a flexible direct current converter valve submodule, characterized in that: The method comprises: Obtaining discrete voltage data of the flexible direct current converter valve submodule, performing distribution verification on the discrete voltage data, and obtaining a distribution verification result; If the distribution verification result is passed, the discrete voltage data is preliminarily fitted by adjusting the length of the time period to determine the inflection point position; Using the maximum likelihood estimation algorithm and the skewed distribution algorithm, the abnormality judgment limit value is determined according to the inflection point position, and a preset sliding window is used to perform continuous calculations over a time period to obtain continuous calculation results; According to the abnormality judgment limit value and the continuous calculation result, voltage abnormality detection is performed on the flexible direct current converter valve submodule to obtain a voltage abnormality detection result.

2. The method according to claim 1, characterized in that Performing distribution verification on the discrete voltage data to obtain distribution verification results includes: Performing a number of occurrence statistics on the discrete voltage data, and determining a significance level test value based on the obtained number of occurrences; The significance level test value is compared with a preset significance threshold, and the distribution verification result is obtained based on the obtained comparison result.

3. The method according to claim 1, characterized in that Using the maximum likelihood estimation algorithm and the skewed distribution algorithm, determining the abnormality judgment limit value according to the inflection point position includes: Calculate the skewed probability density function of the data in each time period according to the inflection point position using the maximum likelihood estimation algorithm and the skewed distribution algorithm; The abnormality judgment limit value is determined according to the skewed probability density function.

4. The method according to claim 3, characterized in that Using the maximum likelihood estimation algorithm and the skewed distribution algorithm, the skewed probability density function of the data in each time period is calculated according to the inflection point position, including: Determine the optimal time period data according to the inflection point position, and fit the optimal time period data using a maximum likelihood estimation algorithm; The skewed distribution algorithm is used to perform skewed distribution processing on the optimal time period data after fitting, and the skewed probability density function of the data in each time period is obtained.

5. A device for detecting abnormal voltage of a flexible direct current converter valve submodule, characterized in that: The device comprises: A distribution verification module is used to obtain discrete voltage data of the flexible direct current converter valve submodule, perform distribution verification on the discrete voltage data, and obtain a distribution verification result; an inflection point location module, configured to perform a preliminary fit on the discrete voltage data by adjusting the length of the time period to determine the inflection point location if the distribution verification result is a passed verification; A judgment limit value module is used to determine the abnormality judgment limit value according to the inflection point position using the maximum likelihood estimation algorithm and the skewed distribution algorithm, and to perform continuous calculations over a time period using a preset sliding window to obtain continuous calculation results; The abnormality detection module is used to perform voltage abnormality detection on the flexible direct current converter valve submodule according to the abnormality judgment limit value and the continuous calculation result to obtain a voltage abnormality detection result.

6. The device according to claim 5, characterized in that The distribution verification module includes: A significance test unit, configured to perform a statistical analysis of the number of occurrences of the discrete voltage data and determine a significance level test value based on the obtained number of occurrences; The distribution verification unit is used to compare the significance level test value with a preset significance threshold, and obtain the distribution verification result based on the comparison result.

7. The device according to claim 5, characterized in that The threshold value determination module includes: A probability density unit is used to calculate the skewed probability density function of the data in each time period according to the inflection point position using a maximum likelihood estimation algorithm and a skewed distribution algorithm; The judgment limit value unit is used to determine the abnormal judgment limit value according to the skewed probability density function.

8. The device according to claim 7, characterized in that The probability density unit includes: A data fitting subunit, configured to determine the optimal time period data according to the inflection point position, and to fit the optimal time period data using a maximum likelihood estimation algorithm; The probability density subunit is used to perform skewed distribution processing on the fitted optimal time period data using a skewed distribution algorithm to obtain a skewed probability density function of the data in each time period.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program for causing a computer to execute the method according to any one of claims 1 to 4.

11. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images