A distributed power data intelligent evaluation and abnormal diagnosis system and method thereof

Abstract

This invention discloses a distributed power data intelligent assessment and anomaly diagnosis system and method, belonging to the field of power big data analysis technology. It includes a data acquisition module for acquiring power generation data from distributed power sources, classifying the power generation data into high-generation, medium-generation, and low-generation ranges, and obtaining the power range corresponding to each range. This invention achieves intelligent assessment and anomaly detection of distributed power generation through refined range division, dynamic correlation analysis, multi-dimensional feature extraction, and quantitative trend diagnosis. Its core advantage lies in the deep integration of electrical theory (such as vertical control) with data-driven methods, ensuring the scientific nature of the diagnosis while improving real-time performance and operability, providing strong technical support for the stable operation of distributed energy systems.

Description

Technical Field

[0001] This invention relates to the field of power big data analysis technology, and in particular to a distributed power data intelligent assessment and anomaly diagnosis system and method. Background Technology

[0002] In modern society, with the widespread application of distributed power systems (such as wind, solar, and ocean energy), a large number of sensors and IoT devices are being deployed on a massive scale in power systems, resulting in an exponential increase in daily electricity consumption data. Under these circumstances, accurate assessment and anomaly diagnosis of electricity consumption data have become crucial to ensuring the safe and stable operation of the system.

[0003] Regarding this research, application CN202310555147.6 provides an intelligent distributed fault diagnosis and type identification system for power transmission and distribution. This technical solution includes a power transmission and distribution line monitoring device and a fault management system. The monitoring device includes a power frequency and traveling wave acquisition module, a GPS clock, a signal processing module, an anti-interference module, a signal transmission module, and a communication module. The fault management system includes a data center, fault diagnosis, fault type identification, and fault report generation. This technical solution can quickly collect fault information and accurately diagnose and identify faults in multi-terminal power supply modes and situations with many branches.

[0004] Another application, CN202511458915.1, provides an intelligent diagnostic system for collecting abnormal electricity consumption information. This technical solution collects abnormal characteristic waveform data through distributed intelligent terminals, selects characteristic harmonic components in the power line frequency voltage to construct an inherent test signal, extracts the envelope shape similarity between the response signal and the test signal through envelope analysis as the first distortion evaluation index, and after statistical significance judgment, performs instantaneous frequency stability analysis on signals with significant distortion to obtain a second distortion evaluation index. Finally, based on the dual indexes, corresponding waveform compensation parameters are generated to achieve real-time compensation and correction. This effectively maintains the integrity of abnormal characteristic waveforms during transmission and significantly improves the accuracy and reliability of abnormal electricity consumption diagnosis.

[0005] However, the aforementioned technical solutions rely on fixed thresholds and rules, lack a multi-dimensional data fusion mechanism, and struggle to perceive and predict the power grid's operational status in real time. They also cannot adapt to complex changes under different operating conditions, leading to high false alarm and false negative rates. In such situations, it is difficult to quickly and accurately pinpoint the cause of anomalies. Summary of the Invention

[0006] In view of the problems existing in the field of power big data analysis technology, the present invention is proposed.

[0007] Therefore, one of the objectives of this invention is to provide a distributed power data intelligent assessment and anomaly diagnosis system and method, which realizes intelligent assessment and anomaly detection of distributed power generation through refined interval division, dynamic correlation analysis, multi-dimensional feature extraction and quantitative trend diagnosis, providing strong support for the stable operation of distributed energy systems.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] On one hand, the present invention provides a distributed intelligent power data assessment and anomaly diagnosis system, comprising:

[0010] The data acquisition module is used to acquire the power generation data of the distributed power source, distinguish the power generation data, including distinguishing it into high power generation range, medium power generation range and low power generation range, and acquire the power range corresponding to the high power generation range, medium power generation range and low power generation range.

[0011] The data processing module includes an acquisition unit and a processing unit;

[0012] The acquisition unit is used to acquire relevant data, including acquiring the power corresponding to the lowest power generation in the high power generation range and the medium power generation range, marking the power as a reference power, and acquiring current and voltage based on the reference power;

[0013] The relevant data also includes obtaining the power corresponding to the highest power generation in the medium power generation range, and marking the power as I power;

[0014] The processing unit is used to perform related processing, which includes:

[0015] Within the reference power corresponding to the medium power generation range, obtain the power difference between this reference power and the I power;

[0016] The power difference is broken down into an intermediate power, a minimum power, and a maximum power. Correlation analysis is then performed based on the intermediate power and the maximum power. The correlation analysis includes analyzing the correlation between the intermediate power and the current and voltage as the intermediate power changes toward the maximum power.

[0017] A feature extraction module, which responds to the association effect and is used to extract features based on the association effect, wherein the feature extraction includes front-layer feature extraction and back-layer feature extraction;

[0018] The preceding feature extraction is to obtain the changes in current and voltage when the intermediate power increases by 1%;

[0019] The data fusion and analysis module includes a computing unit, an analysis unit, and an evaluation and diagnosis unit.

[0020] The calculation unit is used to perform relevant calculations based on the changes, and the steps of the relevant calculations include:

[0021] Each 1% increase in the median power is taken as a calculation target, and 10 consecutive calculation targets constitute a calculation cycle;

[0022] During the calculation cycle, the shortest, average, and longest time required to calculate the current and voltage for each 1% increase.

[0023] In a preferred embodiment of the present invention, the analysis unit responds to the calculation unit and performs analysis based on the results of the relevant calculations, the analysis including trend analysis, as shown below:

[0024] The time required for each 1% increase in the median power is obtained. Taking a continuous 10% increase in the median power as an acquisition period, the average time required for each 1% increase in this process is obtained, and the average time is regarded as the reference average time.

[0025] The time trend required for each 1% increase in current and voltage is analyzed during the acquisition period.

[0026] In a preferred embodiment of the present invention, the evaluation and diagnosis unit responds to the analysis unit to evaluate and diagnose the power generation of the distributed power source. If, at a future time within a certain acquisition period, the power of the distributed power source increases by 1% with the reference average time, and the time required for each 1% increase in current and voltage is the same as the longest time, the system determines that the power generation is abnormal; otherwise, the power generation is not determined to be abnormal.

[0027] In a preferred embodiment of the present invention, the processing unit analyzes the correlation between the intermediate power and the current and voltage as the power changes towards the maximum power, and calculates the result based on the following formula:

[0028] ;

[0029] In the formula, Active power represents the actual electrical energy output by the distributed power source.

[0030] This represents the effective value of the voltage, indicating the magnitude of the output voltage of the distributed power source.

[0031] This represents the effective value of the current, indicating the magnitude of the output current of the distributed power source.

[0032] The power factor (dimensionless) is used to reflect the efficiency of electrical energy utilization.

[0033] ;

[0034] In the formula, Indicates the change in current;

[0035] Indicates the change in power;

[0036] Indicates the effective value of the voltage;

[0037] ;

[0038] In the formula, The droop factor represents the rate at which the voltage decreases as power increases.

[0039] In a preferred embodiment of the present invention, the calculation unit calculates the shortest time, average time, and longest time required for each 1% increase in current and voltage, based on the following formula:

[0040] ;

[0041] In the formula, Indicates the first Within the calculation cycle, the first Time increment when power increases by 1%;

[0042] Indicates the first Within the calculation cycle, the first The starting point when the secondary power increases by 1%;

[0043] Indicates the first Within the calculation cycle, the first The end time when the power increases by 1%;

[0044] ; ;

[0045] In the formula, Indicates the first Within the calculation cycle, the first The time required for a 1% change in current when the secondary power increases by 1%;

[0046] Indicates the first Within the calculation cycle, the first The time required for a 1% voltage change when the secondary power increases by 1%;

[0047] Indicates the first The change in current corresponding to a 1% increase in secondary power;

[0048] This represents the current corresponding to the median power value;

[0049] Indicates the first The change in voltage corresponding to a 1% increase in secondary power;

[0050] This represents the voltage corresponding to the median power value;

[0051] ; ;

[0052] In the formula, Indicates the first The shortest time required for a 1% change in current within a calculation cycle;

[0053] Indicates the first The shortest time required for a 1% voltage change within a calculation cycle;

[0054] This represents a function that takes the minimum value.

[0055] ; ;

[0056] In the formula, Indicates the first The average time required for a 1% change in current within a calculation cycle;

[0057] Indicates the first The average time required for a 1% voltage change within a calculation cycle;

[0058] This represents the summation function;

[0059] ; ;

[0060] In the formula, Indicates the first The longest time required for a 1% change in current within a calculation cycle;

[0061] Indicates the first The longest time required for a 1% voltage change within a calculation cycle;

[0062] This represents the function that takes the maximum value.

[0063] In a preferred embodiment of the present invention, relevant information is obtained based on the calculation results. This relevant information includes the fluctuation rate of time required to obtain the change in current (for every 1% increase) within the calculation period, calculated according to the following formula:

[0064] ;

[0065] In the formula, Indicates the first The standard deviation of the time required for a 1% change in current within a calculation cycle.

[0066] In a preferred embodiment of the present invention, the subsequent feature extraction in the feature extraction module comprises the following steps:

[0067] Relevance extraction is performed in a time period of at least three calculation cycles, wherein the correlation extraction is based on the shortest time and includes:

[0068] Within the extraction period, the period corresponding to the shortest time required for each 1% increase in current and voltage is extracted, the power that occurred in this period is obtained, and a dataset is generated based on the power.

[0069] In a preferred embodiment of the present invention, when the power of the distributed power source at a future time is the same as any power in the dataset, if the time required for each 1% increase in current and voltage is greater than the shortest time, the system determines that the power generation of the distributed power source is abnormal; otherwise, if the time is the same as the shortest time, the power generation is not determined to be abnormal.

[0070] On the other hand, the present invention provides a method for applying to a distributed power data intelligent assessment and anomaly diagnosis system as described above, comprising the following steps:

[0071] Obtain the power generation data of the distributed power source, and classify the power generation data into high power generation range, medium power generation range and low power generation range, and obtain the power range corresponding to the high power generation range, medium power generation range and low power generation range.

[0072] Obtain the power corresponding to the lowest power generation in the high power generation range and the medium power generation range, mark the power as the reference power, and obtain the current and voltage based on the reference power;

[0073] The relevant data also includes obtaining the power corresponding to the highest power generation in the medium power generation range, and marking the power as I power;

[0074] Within the reference power corresponding to the medium power generation range, obtain the power difference between this reference power and the I power;

[0075] The power difference is broken down into an intermediate power, a minimum power, and a maximum power. Correlation analysis is then performed based on the intermediate power and the maximum power. The correlation analysis includes analyzing the correlation between the intermediate power and the current and voltage as the intermediate power changes toward the maximum power.

[0076] Feature extraction is performed based on the correlation effect, and the feature extraction includes front-layer feature extraction and back-layer feature extraction;

[0077] The preceding feature extraction is to obtain the changes in current and voltage when the intermediate power increases by 1%;

[0078] Based on the changes, relevant calculations are performed, and the steps of the relevant calculations include:

[0079] Each 1% increase in the median power is taken as a calculation target, and 10 consecutive calculation targets constitute a calculation cycle;

[0080] During the calculation cycle, the shortest, average, and longest time required to calculate the current and voltage for each 1% increase.

[0081] Beneficial effects:

[0082] 1. This invention divides the power generation data of distributed power sources into high, medium, and low power generation ranges and clarifies the power range corresponding to each range. The system can more accurately capture the power characteristics under different operating conditions, which is conducive to improving the pertinence and accuracy of data processing. Furthermore, the system uses the lowest power generation in the high and medium power generation ranges as the reference power and combines droop control theory to analyze the correlation between the intermediate power and the maximum power and the current and voltage. By quantifying the dynamic impact of power changes on current and voltage, the inherent electrical characteristics of the system are revealed, providing a scientific basis for anomaly diagnosis.

[0083] 2. The system of the present invention achieves in-depth mining of power data through front-layer feature extraction (current and voltage changes when the median power increases by 1%) and back-layer feature extraction (generated based on the power dataset with the shortest time). The front-layer feature extraction captures the instantaneous dynamic response, while the back-layer feature extraction focuses on the typical features under stable operating conditions. The combination of the two improves the comprehensiveness and reliability of anomaly diagnosis.

[0084] 3. This invention uses each 1% increase in the intermediate power value as a calculation target, and 10 consecutive targets as a calculation cycle to dynamically calculate the shortest, average, and longest time required for changes in current and voltage. Through quantitative analysis in the time dimension, the system can identify response delays or abnormal fluctuations.

[0085] 4. The diagnostic logic of this invention is strictly based on quantitative indicators, avoiding subjective judgment errors. For example, when the power matches the dataset but the time taken exceeds the minimum time, the system immediately triggers an abnormal alarm to ensure that the problem is detected in a timely manner. Attached Figure Description

[0086] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the modular structure of the distributed power data intelligent assessment and anomaly diagnosis system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the method flow according to an embodiment of the present invention; The diagram is labeled as follows: 110 - Data acquisition module; 120 - Data processing module; 1201 - Acquisition unit; 1202 - Processing unit; 130 - Data fusion and analysis module; 1301 - Calculation unit; 1302 - Analysis unit; 1303 - Evaluation and diagnosis unit. Detailed Implementation

[0087] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0088] Because existing technical solutions rely on fixed thresholds and rules and lack a multi-dimensional data fusion mechanism, they are difficult to perceive and predict the power grid's operating status in real time and cannot adapt to complex changes under different operating conditions, resulting in a high rate of false alarms and missed alarms.

[0089] Based on this, this invention proposes a distributed power data intelligent assessment and anomaly diagnosis system and method. Through refined interval division, dynamic correlation analysis, multi-dimensional feature extraction, and quantitative trend diagnosis, it achieves intelligent assessment and anomaly detection of distributed power generation. Its core advantage lies in the deep integration of electrical theory (such as vertical control) with data-driven methods, ensuring both the scientific rigor of the diagnosis and improving real-time performance and operability, thus providing strong technical support for the stable operation of distributed energy systems.

[0090] The present solution will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0091] Reference Figures 1 to 2 As one embodiment of the present invention, this embodiment provides a distributed power data intelligent assessment and anomaly diagnosis system, including:

[0092] The data acquisition module 110 is used to acquire the power generation data of the distributed power source (the distributed power source in the target area), distinguish the power generation data, including distinguishing it into high power generation range, medium power generation range and low power generation range, and acquire the power range corresponding to the high power generation range, medium power generation range and low power generation range.

[0093] It should be noted that the division of high, medium, and low power generation ranges needs to be comprehensively judged based on the specific power source type, installed capacity, environmental conditions, and actual application scenarios, and should be flexibly set without a unified fixed numerical range.

[0094] Furthermore, in reality, power is the core indicator of power generation. Therefore, obtaining the power ranges corresponding to high, medium, and low power generation ranges is of practical significance. Obviously, this is known to those skilled in the art.

[0095] In addition, the power generated by distributed power sources is related to current and voltage. The current and voltage will be obtained according to the technical solution of this application. The three factors together determine the energy output characteristics of the power generation system.

[0096] The system divides the power generation data of distributed power sources into three intervals: high, medium, and low, and clearly defines the power range corresponding to each interval. By distinguishing the power characteristics of different intervals, the system can analyze the power data under different operating conditions in a targeted manner.

[0097] This avoids the errors caused by the coarse division of intervals in traditional methods, improves the accuracy of data processing, and allows for differentiated analysis strategies for different power generation intervals, thereby enhancing the system's adaptability to complex operating conditions.

[0098] Data processing module 120, which includes acquisition unit 1201 and processing unit 1202;

[0099] The acquisition unit 1201 is used to acquire relevant data, including acquiring the power corresponding to the lowest power generation in the high power generation range and the medium power generation range, marking the power as the reference power, and acquiring the current and voltage based on the reference power;

[0100] The relevant data also includes obtaining the power corresponding to the highest power generation in the medium power generation range, and marking the power as I power;

[0101] Processing unit 1202 is used to perform related processing, which includes:

[0102] Within the reference power corresponding to the medium power generation range, obtain the power difference between this reference power and power I;

[0103] The power difference is broken down into median power, minimum power, and maximum power. Correlation analysis is then performed based on the median power and maximum power, including the correlation effect of the median power changing toward the maximum power on current and voltage.

[0104] This embodiment uses the lowest power generation in the high and medium power generation ranges as the reference power, and combines the droop control theory to analyze the correlation between current and voltage when the power changes from the intermediate value to the maximum value.

[0105] This method quantifies correlations through electrical theory, reveals the inherent characteristics of the system, provides theoretical support for anomaly diagnosis, and can reflect the impact of power changes on current and voltage in real time, thereby improving the sensitivity of diagnosis.

[0106] The feature extraction module responds to the correlation effect and is used to extract features based on the correlation effect. Feature extraction includes pre-layer feature extraction and post-layer feature extraction.

[0107] The preceding feature extraction is used to obtain the changes in current and voltage when the median power increases by 1%;

[0108] For every 1% increase in the intermediate power, the changes in current and voltage are recorded to capture the instantaneous dynamic response.

[0109] The data fusion analysis module 130 includes a calculation unit 1301, an analysis unit 1302, and an evaluation and diagnosis unit 1303.

[0110] The calculation unit 1301 is used to perform relevant calculations based on changes. The steps of the relevant calculations include:

[0111] Each 1% increase in the median power is used as a calculation target, and 10 consecutive calculation targets constitute a calculation cycle.

[0112] Within the calculation cycle, calculate the shortest, average, and longest time required for each 1% increase in current and voltage;

[0113] This embodiment quantifies response characteristics through the time dimension, avoiding subjective judgment errors and enabling early detection of potential fault trends, thus supporting preventative maintenance.

[0114] The subsequent feature extraction steps are as follows:

[0115] Relevant extraction is performed in periods of at least three computation cycles, based on the shortest extraction time, and includes:

[0116] Within the extraction period, the period corresponding to the shortest time required for each 1% increase in current and voltage is extracted, the power that appears in this period is obtained, and a dataset is generated based on the power.

[0117] Using at least three calculation cycles as time periods, extract power data that requires the shortest time to change both current and voltage, and generate a typical feature dataset.

[0118] The front layer focuses on instantaneous changes, while the back layer focuses on stable operating conditions, thus improving the comprehensiveness of the diagnosis.

[0119] When the power of a distributed source is the same as any power in the dataset at a future time, if the time required for each 1% increase in current and voltage is greater than the shortest time, the system determines that the power generation of the distributed source is abnormal; otherwise, if the time required is the same as the shortest time, the power generation is not determined to be abnormal.

[0120] By filtering typical features under stable operating conditions from the dataset, noise interference can be reduced and reliability improved.

[0121] Analysis unit 1302 is a response calculation unit used to perform analysis based on the results of relevant calculations. The analysis includes trend analysis, as shown below:

[0122] The time required to obtain the median power for each 1% increase is calculated, and the average time required for each 1% increase in the median power is obtained during a continuous 10% increase period. This average time is then used as the reference average time.

[0123] Analyze the time trend required for each 1% increase in current and voltage during the acquisition period.

[0124] The assessment and diagnosis unit 1303 response analysis unit is used to assess and diagnose the power generation of distributed power sources. When the power of distributed power sources increases by 1% with reference average time in a future acquisition period, if the time required for each 1% increase in current and voltage is the same as the longest time, the system determines that the power generation is abnormal; otherwise, it does not determine that the power generation is abnormal.

[0125] In reality, there is a design range for the power, current, and voltage of a distributed power system. If, within a certain time period, the changes in current and voltage are slower than the changes in power, it indicates that the control parameters for the distributed power system need to be optimized or that there is a hardware fault (such as sensor failure or capacitor aging). In this case, determining abnormal power generation is of practical significance.

[0126] In the processing unit, the correlation between the intermediate power and the current and voltage is analyzed as the power changes towards the maximum power. The results are calculated based on the following formula:

[0127] ;

[0128] In the formula, Active power represents the actual electrical energy output by the distributed power source.

[0129] This represents the effective value of the voltage, indicating the magnitude of the output voltage of the distributed power source.

[0130] This represents the effective value of the current, indicating the magnitude of the output current of the distributed power source.

[0131] The power factor (dimensionless) is used to reflect the efficiency of electrical energy utilization.

[0132] This formula is used to calculate the current or voltage value at any power point, or to verify the deviation between measured data and theoretical values;

[0133] ;

[0134] In the formula, It represents the change in current (i.e., the difference in current between the maximum power and the intermediate power).

[0135] It represents the change in power (i.e., the power difference between the maximum power and the intermediate power).

[0136] Indicates the effective value of the voltage (current);

[0137] This formula quantifies the speed of current response as power increases, and is used to evaluate the dynamic performance of a control system.

[0138] ;

[0139] In the formula, The droop factor represents the rate at which the voltage decreases as power increases (determined by the control system parameters of the distributed power source in real-world scenarios).

[0140] This formula is used to analyze the voltage stability of distributed power sources as power increases, and to determine whether voltage collapse is caused by an excessively large droop factor.

[0141] In the calculation unit, the shortest, average, and longest time required for each 1% increase in current and voltage are calculated using the following formulas:

[0142] ;

[0143] In the formula, Indicates the first Within the calculation cycle, the first Time increment when power increases by 1%;

[0144] Indicates the first Within the calculation cycle, the first The starting point when the secondary power increases by 1%;

[0145] Indicates the first Within the calculation cycle, the first The end time when the power increases by 1%;

[0146] This formula calculates the duration of each 1% power increase, which serves as the basis for subsequent statistics;

[0147] ; ;

[0148] In the formula, Indicates the first Within the calculation cycle, the first The time required for a 1% change in current when the secondary power increases by 1%;

[0149] Indicates the first Within the calculation cycle, the first The time required for a 1% voltage change when the secondary power increases by 1%;

[0150] Indicates the first The change in current corresponding to a 1% increase in secondary power;

[0151] This represents the current corresponding to the median power value;

[0152] Indicates the first The change in voltage corresponding to a 1% increase in secondary power;

[0153] This represents the voltage corresponding to the median power value;

[0154] This formula is used to standardize the time increment to eliminate the influence of differences in the absolute values ​​of current / voltage across different power ranges;

[0155] ; ;

[0156] In the formula, Indicates the first The shortest time required for a 1% change in current within a calculation cycle;

[0157] Indicates the first The shortest time required for a 1% voltage change within a calculation cycle;

[0158] This represents a function that takes the minimum value.

[0159] This formula is used to identify the scenario where the system responds the fastest (such as when the hardware performance is optimal or the control strategy is optimal).

[0160] ; ;

[0161] In the formula, Indicates the first The average time required for a 1% change in current within a calculation cycle;

[0162] Indicates the first The average time required for a 1% voltage change within a calculation cycle;

[0163] This represents the summation function;

[0164] This formula is used to evaluate the typical response speed of distributed power sources under normal operating conditions, and is used for comparison with theoretical values ​​or historical data.

[0165] ; ;

[0166] In the formula, Indicates the first The longest time required for a 1% change in current within a calculation cycle;

[0167] Indicates the first The longest time required for a 1% voltage change within a calculation cycle;

[0168] This represents the function that takes the maximum value.

[0169] This formula detects system response delays or anomalies (such as slowed response due to hardware aging or communication failures).

[0170] Based on the calculation results, relevant information is obtained, including the fluctuation rate of time required to obtain the change in current (for every 1% increase) within the calculation period, calculated using the following formula:

[0171] ;

[0172] In the formula, Indicates the first The standard deviation of the time required for a 1% change in current within a calculation cycle.

[0173] The standard deviation reflects the degree of fluctuation in time; the larger the value, the worse the system stability.

[0174] This formula can quantify the consistency of distributed power source response and help diagnose intermittent faults (such as poor contact).

[0175] Obviously, those skilled in the art can also obtain the fluctuation rate required for the change in voltage (for every 1% increase) based on the calculation method of this formula.

[0176] As can be seen from the above, this application achieves intelligent assessment and anomaly detection of distributed power generation through interval division, droop control, multi-dimensional feature extraction, dynamic time calculation and quantitative diagnosis. Its technical effects cover accuracy, real-time and adaptability, providing efficient and reliable support for the stable operation of distributed energy systems.

[0177] This embodiment, in conjunction with the aforementioned distributed power data intelligent assessment and anomaly diagnosis system, also proposes a working method for this system, as follows:

[0178] S10: Obtain the power generation data of the distributed power source, distinguish the power generation data, including distinguishing it into high power generation range, medium power generation range and low power generation range, and obtain the power range corresponding to the high power generation range, medium power generation range and low power generation range.

[0179] S20: Obtain the power corresponding to the lowest power generation in the high power generation range and the medium power generation range, mark the power as the reference power, and obtain the current and voltage based on the reference power;

[0180] S30: The relevant data also includes obtaining the power corresponding to the highest power generation in the medium power generation range, and marking the power as I power;

[0181] S40: In the reference power corresponding to the medium power generation range, obtain the power difference between this reference power and the I power;

[0182] S50: Decompose the power difference into median power, minimum power and maximum power, and perform correlation analysis based on the median power and maximum power. The correlation analysis includes analyzing the correlation effect of the median power on current and voltage as it changes toward the maximum power.

[0183] S60: Feature extraction is performed based on the correlation effect. Feature extraction includes front-layer feature extraction and back-layer feature extraction.

[0184] The preceding feature extraction is used to obtain the changes in current and voltage when the median power increases by 1%;

[0185] S70: Perform relevant calculations based on the changes. The steps for these calculations include:

[0186] Each 1% increase in the median power is used as a calculation target, and 10 consecutive calculation targets constitute a calculation cycle.

[0187] During the calculation cycle, calculate the shortest, average, and longest time required for each 1% increase in current and voltage.

[0188] In summary, this invention divides power generation intervals and clarifies power ranges, quantifies the dynamic impact of power changes on current and voltage using droop control theory, achieves deep data mining through multi-dimensional feature extraction and dynamic time calculation, and finally realizes accurate and real-time anomaly diagnosis based on quantitative index comparison. The system has advantages such as strong adaptability, high scalability, and strong engineering applicability, which can significantly improve the stability and economy of distributed power generation operation and provide efficient technical support for the intelligent management of energy systems.

[0189] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A distributed intelligent power data assessment and anomaly diagnosis system, characterized in that, include: The data acquisition module is used to acquire the power generation data of the distributed power source, distinguish the power generation data, including distinguishing it into high power generation range, medium power generation range and low power generation range, and acquire the power range corresponding to the high power generation range, medium power generation range and low power generation range. The data processing module includes an acquisition unit and a processing unit; The acquisition unit is used to acquire relevant data, including acquiring the power corresponding to the lowest power generation in the high power generation range and the medium power generation range, marking the power as a reference power, and acquiring current and voltage based on the reference power; The relevant data also includes obtaining the power corresponding to the highest power generation in the medium power generation range, and marking the power as I power; The processing unit is used to perform related processing, which includes: Within the reference power corresponding to the medium power generation range, obtain the power difference between this reference power and the I power; The power difference is broken down into an intermediate power, a minimum power, and a maximum power. Correlation analysis is then performed based on the intermediate power and the maximum power. The correlation analysis includes analyzing the correlation between the intermediate power and the current and voltage as the intermediate power changes toward the maximum power. A feature extraction module, which responds to the association effect and is used to extract features based on the association effect, wherein the feature extraction includes front-layer feature extraction and back-layer feature extraction; The preceding feature extraction is to obtain the changes in current and voltage when the intermediate power increases by 1%; The data fusion and analysis module includes a computing unit, an analysis unit, and an evaluation and diagnosis unit. The calculation unit is used to perform relevant calculations based on the changes, and the steps of the relevant calculations include: Each 1% increase in the median power is taken as a calculation target, and 10 consecutive calculation targets constitute a calculation cycle; During the calculation cycle, the shortest, average, and longest time required to calculate the current and voltage for each 1% increase.

2. The distributed power data intelligent assessment and anomaly diagnosis system as described in claim 1, characterized in that, The analysis unit responds to the calculation unit by performing analysis based on the results of the relevant calculations. The analysis includes trend analysis, as shown below: The time required for each 1% increase in the median power is obtained. Taking a continuous 10% increase in the median power as an acquisition period, the average time required for each 1% increase in this process is obtained, and the average time is regarded as the reference average time. The time trend required for each 1% increase in current and voltage is analyzed during the acquisition period.

3. The distributed power data intelligent assessment and anomaly diagnosis system as described in claim 2, characterized in that, The assessment and diagnosis unit responds to the analysis unit to assess and diagnose the power generation of the distributed power source. If, at a future time within a certain acquisition period, the power of the distributed power source increases by 1% with the reference average time, and the time required for each 1% increase in current and voltage is the same as the longest time, the system determines that the power generation is abnormal; otherwise, it does not determine that the power generation is abnormal.

4. The distributed power data intelligent assessment and anomaly diagnosis system as described in claim 1, characterized in that, In the processing unit, the correlation between the intermediate power and the current and voltage is analyzed as the intermediate power changes towards the maximum power, and the results are calculated based on the following formula: ； In the formula, Active power represents the actual electrical energy output by the distributed power source. This represents the effective value of the voltage, indicating the magnitude of the output voltage of the distributed power source. This represents the effective value of the current, indicating the magnitude of the output current of the distributed power source. The power factor (dimensionless) is used to reflect the efficiency of electrical energy utilization. ； In the formula, Indicates the change in current; Indicates the change in power; Indicates the effective value of the voltage; ； In the formula, The droop factor represents the rate at which the voltage decreases as power increases.

5. The distributed power data intelligent assessment and anomaly diagnosis system as described in claim 1, characterized in that, In the calculation unit, the shortest, average, and longest time required for each 1% increase in current and voltage are calculated using the following formula: ； In the formula, Indicates the first Within the calculation cycle, the first Time increment when power increases by 1%; Indicates the first Within the calculation cycle, the first The starting point when the secondary power increases by 1%; Indicates the first Within the calculation cycle, the first The end time when the power increases by 1%; ； ； In the formula, Indicates the first Within the calculation cycle, the first The time required for a 1% change in current when the secondary power increases by 1%; Indicates the first Within the calculation cycle, the first The time required for a 1% voltage change when the secondary power increases by 1%; Indicates the first The change in current corresponding to a 1% increase in secondary power; This represents the current corresponding to the median power value; Indicates the first The change in voltage corresponding to a 1% increase in secondary power; This represents the voltage corresponding to the median power value; ； ； In the formula, Indicates the first The shortest time required for a 1% change in current within a calculation cycle; Indicates the first The shortest time required for a 1% voltage change within a calculation cycle; This represents a function that takes the minimum value. ； ； In the formula, Indicates the first The average time required for a 1% change in current within a calculation cycle; Indicates the first The average time required for a 1% voltage change within a calculation cycle; This represents the summation function; ； ； In the formula, Indicates the first The longest time required for a 1% change in current within a calculation cycle; Indicates the first The longest time required for a 1% voltage change within a calculation cycle; This represents the function that takes the maximum value.

6. The distributed power data intelligent assessment and anomaly diagnosis system as described in claim 5, characterized in that, Based on the calculation results, relevant information is obtained, including the fluctuation rate of time required to obtain the change in current (for every 1% increase) within the calculation period, calculated according to the following formula: ； In the formula, Indicates the first The standard deviation of the time required for a 1% change in current within a calculation cycle.

7. The distributed power data intelligent assessment and anomaly diagnosis system as described in claim 1, characterized in that, In the feature extraction module, the subsequent feature extraction steps are as follows: Relevance extraction is performed in a time period of at least three calculation cycles, wherein the correlation extraction is based on the shortest time and includes: Within the extraction period, the period corresponding to the shortest time required for each 1% increase in current and voltage is extracted, the power that occurred in this period is obtained, and a dataset is generated based on the power.

8. The distributed power data intelligent assessment and anomaly diagnosis system as described in claim 7, characterized in that, When the power of the distributed power source in the future is the same as any power in the dataset, if the time required for each 1% increase in current and voltage is greater than the shortest time, the system determines that the power generation of the distributed power source is abnormal; otherwise, if the time is the same as the shortest time, the power generation is not determined to be abnormal.

9. A method applied to a distributed power data intelligent assessment and anomaly diagnosis system as described in claim 1, characterized in that, Includes the following steps: Obtain the power generation data of the distributed power source, and classify the power generation data into high power generation range, medium power generation range and low power generation range, and obtain the power range corresponding to the high power generation range, medium power generation range and low power generation range. Obtain the power corresponding to the lowest power generation in the high power generation range and the medium power generation range, mark the power as the reference power, and obtain the current and voltage based on the reference power; The relevant data also includes obtaining the power corresponding to the highest power generation in the medium power generation range, and marking the power as I power; Within the reference power corresponding to the medium power generation range, obtain the power difference between this reference power and the I power; The power difference is broken down into an intermediate power, a minimum power, and a maximum power. Correlation analysis is then performed based on the intermediate power and the maximum power. The correlation analysis includes analyzing the correlation between the intermediate power and the current and voltage as the intermediate power changes toward the maximum power. Feature extraction is performed based on the correlation effect, and the feature extraction includes front-layer feature extraction and back-layer feature extraction; The preceding feature extraction is to obtain the changes in current and voltage when the intermediate power increases by 1%; Based on the changes, relevant calculations are performed, and the steps of the relevant calculations include: Each 1% increase in the median power is taken as a calculation target, and 10 consecutive calculation targets constitute a calculation cycle; Within the calculation cycle, the shortest, average, and longest time required to calculate each 1% increase in current and voltage.

Images