New energy operation abnormity monitoring method and system

Abstract

The invention discloses a new energy operation abnormity monitoring method and system. The method comprises the following steps: firstly, obtaining historical data and real-time data in a new energy system; operation index signals in the new energy system are obtained according to the historical data, association rule mining of preset historical moment windows of all the operation index signals is executed, and historical operation abnormal coefficients are determined according to the operation index difference values of all adjacent moments; determining real-time normal probability distribution data corresponding to the new energy system; and finally, according to the real-time normal probability distribution data, executing abnormal condition monitoring on the current operation condition of the new energy system. According to the method, the potential abnormal mode in the operation of the new energy system can be identified more accurately, real-time data are introduced, the monitoring instantaneity is ensured, the system can be perceived at the initial stage of abnormity, and the monitoring comprehensiveness and accuracy are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of abnormality monitoring, and in particular to a method and system for monitoring abnormality in operation of new energy. Background Art

[0002] With the growing global demand for renewable energy, new energy systems such as solar energy and wind energy have become an important part of the energy structure. However, the operation of new energy systems is affected by many factors, such as weather conditions, equipment aging, maintenance status, etc. These factors may cause system operation abnormalities, thereby affecting the stable supply of energy and the overall efficiency of the system. Therefore, it is particularly important to conduct timely and accurate monitoring and early warning of abnormal operation of new energy systems.

[0003] Traditional monitoring methods for new energy systems mainly rely on manual inspections and regular maintenance, which is not only inefficient but also difficult to detect and handle potential abnormal problems in a timely manner. In recent years, with the rapid development of technologies such as big data and artificial intelligence, abnormal monitoring methods based on data analysis have gradually attracted attention. These methods collect historical data and real-time data from new energy systems, and use data mining, machine learning and other technical means to monitor the operating status of the system in real time and identify abnormalities, greatly improving the accuracy and timeliness of monitoring.

[0004] However, the existing abnormal monitoring methods for new energy systems still have some shortcomings. On the one hand, most methods only focus on the changes of a single operating indicator, ignoring the correlation and mutual influence between the indicators, resulting in low accuracy of abnormal identification. On the other hand, although some methods consider the comprehensive analysis of multiple indicators, they lack in-depth mining and effective use of historical data, making it difficult to accurately reflect the historical operating laws and abnormal characteristics of the system, which has become a problem that needs to be solved urgently. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for monitoring abnormal operation of new energy to address the deficiencies in the prior art. The method and system can more accurately identify potential abnormal modes in the operation of new energy systems, introduce real-time data, ensure the immediacy of monitoring, and enable the system to be detected at the early stage of abnormality, thereby improving the comprehensiveness and accuracy of monitoring.

[0006] An embodiment of the present application provides a method for monitoring abnormal operation of new energy sources, the method comprising: Obtain historical and real-time data in new energy systems; According to historical data, the operation index signal in the new energy system is obtained, and the association rule mining of the preset historical time window of each operation index signal is performed, and the historical operation abnormality coefficient is determined according to the difference of the operation index at all adjacent times; wherein the operation index signal includes energy conversion efficiency, power density and discrete electrical signal; Determining real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system; According to the real-time normal probability distribution data, abnormal situation monitoring is performed on the current operation situation of the new energy system.

[0007] Optionally, the acquiring of historical data and real-time data in the new energy system includes: Obtain historical time series data of the new energy system and preprocess the historical time series data; The preprocessed historical time series data is randomly divided into samples of different time intervals; each sample corresponds to the historical operation data at that moment; In response to the presence of abnormal values ​​in the currently obtained historical operation data, the abnormal values ​​are removed, the historical operation data is dimensionally expanded, and the operation characteristic information contained in each historical moment is added to generate historical data in the new energy system.

[0008] Optionally, the acquiring of historical data and real-time data in the new energy system includes: Taking the current time as a reference, the load data of the new energy system collected by the load collection device of the new energy system is recorded as the real-time data of the new energy system.

[0009] Optionally, the method of acquiring an operation index signal in the new energy system according to historical data, performing association rule mining of a preset historical time window for each operation index signal, and determining a historical operation abnormality coefficient according to the difference of the operation index at all adjacent times includes: Perform discrete operations on the electrical signals in the operation index signals, and use association rule mining algorithms to perform clustering operations on each discrete electrical signal; Calculate the absolute value of the difference between the discrete electrical signals corresponding to any adjacent moments, and take the sum of the absolute values ​​of all the differences as the historical operation abnormality coefficient.

[0010] Optionally, the determining of the real-time normal probability distribution data corresponding to the new energy system includes: analyzing the relationship between the historical operation abnormality coefficient and the operation state of the new energy system to identify the abnormal mode of the new energy system; Determine the impact factors of abnormal modes of new energy systems on key indicators of new energy systems; According to the influencing factors and the real-time data, a real-time normal probability distribution model corresponding to the new energy system is established to determine the real-time normal probability distribution data corresponding to the new energy system.

[0011] Optionally, the real-time normal probability distribution model is dynamically updated using a sliding window technology to adapt to changes in the operating status of the new energy system over time.

[0012] Optionally, performing abnormal situation monitoring on the current operation situation of the new energy system according to the real-time normal probability distribution data includes: Calculate the abnormal degree of abnormal data in real-time normal probability distribution data; monitoring abnormal load data of the new energy system according to the abnormality degree of the abnormal data; Based on the relationship between the monitored abnormal load data and the preset load threshold, complete the abnormal monitoring of new energy operation.

[0013] Another embodiment of the present application provides a new energy operation abnormality monitoring system, the system comprising: Acquisition module, used to acquire historical data and real-time data in the new energy system; The first execution module is used to obtain the operation index signal in the new energy system according to the historical data, perform the association rule mining of the preset historical time window of each operation index signal, and determine the historical operation abnormality coefficient according to the difference of the operation index at all adjacent times; wherein the operation index signal includes energy conversion efficiency, power density and discrete electrical signal; A determination module, used for determining real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system; The second execution module is used to perform abnormal situation monitoring on the current operation status of the new energy system according to the real-time normal probability distribution data.

[0014] Yet another embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the method described above.

[0015] Another embodiment of the present application provides a storage medium, wherein a computer program is stored in the storage medium, wherein the computer program is configured to execute the method described above when it is run. Compared with the prior art, the present invention first obtains historical data and real-time data in the new energy system; obtains the operating index signal in the new energy system based on the historical data, performs association rule mining on the preset historical time window of each operating index signal, and determines the historical operating abnormality coefficient according to the difference in operating indexes at all adjacent times; determines the real-time normal probability distribution data corresponding to the new energy system; and finally, performs abnormal situation monitoring on the current operating status of the new energy system based on the real-time normal probability distribution data. It can more accurately identify potential abnormal patterns in the operation of the new energy system, introduce real-time data, ensure the immediacy of monitoring, and enable the system to be detected at the early stage of the abnormality, thereby improving the comprehensiveness and accuracy of monitoring. Description of the drawings Figure 1 A hardware structure block diagram of a computer terminal for a method for monitoring abnormal operation of new energy provided by an embodiment of the present invention; Figure 2 A schematic diagram of a flow chart of a method for monitoring abnormal operation of new energy provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a new energy operation abnormality monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, but should not be construed as limiting the present invention.

[0017] An embodiment of the present invention provides a method for monitoring abnormal operation of new energy, which can be applied to electronic devices such as computer terminals, specifically ordinary computers, tablets, etc.

[0018] The following describes it in detail by taking running on a computer terminal as an example. Figure 1 The hardware structure block diagram of a computer terminal of a new energy operation abnormality monitoring method provided by an embodiment of the present invention. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Optionally, the computer terminal may also include a transmission device 106 for communication functions and an input and output device 108. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. Figure 1More or fewer components as shown, or with Figure 1 Different configurations are shown. The memory 104 can be used to store software programs and modules of application software, such as the program instructions / modules corresponding to the new energy operation abnormality monitoring method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implementing the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0019] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0020] See also Figure 2 , Figure 2 A schematic flow chart of a method for monitoring abnormal operation of new energy provided by an embodiment of the present invention may include the following steps: S201: Acquire historical data and real-time data in the new energy system.

[0021] Specifically, the acquisition of historical data and real-time data in the new energy system may include: 1. Obtain historical time series data of the new energy system and preprocess the historical time series data.

[0022] 2. Randomly divide the preprocessed historical time series data into samples of different time intervals; each sample corresponds to the historical operation data at that moment.

[0023] 3. In response to the presence of abnormal values ​​in the currently obtained historical operation data, the abnormal values ​​are removed, the historical operation data is dimensionally expanded, and the operation characteristic information contained in each historical moment is added to generate historical data in the new energy system.

[0024] Specifically, the historical time series data of the new energy system can include the time series data collected from the sensors, databases, log files, etc. of the new energy system. These data contain the operating status information of the new energy system at different time points, such as power generation, voltage, current, temperature, etc.

[0025] Then the historical time series data is preprocessed, including removing duplicate data, filling missing values, correcting erroneous data, etc. The data is standardized, normalized or converted as needed for subsequent processing and analysis. As needed, the original data may need to be aggregated to reduce the amount of data and improve processing efficiency. The preprocessed historical time series data is randomly divided into different time interval samples in order to construct the data set required for machine learning or statistical analysis. Through random segmentation, continuous time series data can be converted into multiple independent time interval samples, each of which represents the operating status of the new energy system in a certain time period. The length of each time interval sample can be set according to specific needs, for example, different time intervals such as 1 hour, 1 day or 1 week can be selected. Each time interval sample corresponds to a specific moment, and the historical operating data of the moment is included in the sample. These data are used to describe the operating status of the new energy system at that moment, such as energy conversion efficiency, power density, etc.

[0026] In the process of data preprocessing or sample segmentation, some outliers may be found, which may be caused by sensor failure, data recording errors, etc. In order to ensure data quality, these outliers need to be removed. Then the historical operation data is dimensionally expanded, and the operation feature information contained in each historical moment is added. Dimension expansion refers to increasing the dimension of data by calculating or adding new features. In the historical data of the new energy system, in addition to the basic operation indicators, some new features such as rate of change, standard deviation, moving average, etc. can be calculated based on the historical data. At the same time, in order to ensure that the operating status of each historical moment can be fully described, it is necessary to add all the important operation feature information contained in the moment. These feature information can help us understand the operating status of the new energy system more comprehensively and improve the accuracy of anomaly detection. After the above steps are processed, the historical data in the new energy system can be generated. These data will be used in subsequent model training, anomaly detection and other tasks.

[0027] In an optional implementation manner, the acquiring of historical data and real-time data in the new energy system may include: Taking the current time as a reference, the load data of the new energy system collected by the load collection device of the new energy system is recorded as the real-time data of the new energy system.

[0028] Specifically, real-time data refers to data collected at or very close to the current moment, which is highly timely. This helps the operation and maintenance personnel of the new energy system to understand the operating status of the system in a timely manner and respond quickly to possible abnormal situations. Load collection devices usually have high accuracy and stability, which can ensure that the collected real-time data is accurate and reliable. This is of great significance for performance evaluation, fault diagnosis and predictive maintenance of new energy systems. Real-time data covers many aspects of new energy systems, including key indicators such as power generation, voltage, current, and power. The comprehensive analysis of these data can fully reflect the operating status and performance characteristics of the new energy system.

[0029] The acquisition of real-time data usually relies on the real-time monitoring system of the new energy system, which collects the operating status information of the new energy system in real time and transmits this information to the data center or cloud platform for further processing and analysis. The acquisition of real-time data may include directly reading sensor data, querying the latest records from the database, and using API interface calls.

[0030] Exemplarily, new energy systems are usually equipped with special load collection devices, which can monitor and record various operating parameters of the new energy system in real time, including but not limited to power generation, voltage, current, power, etc. The load collection device obtains various operating data of the new energy system in real time through hardware devices such as sensors. These data are then converted into digital signals and transmitted to the data processing center or cloud platform for further processing and analysis. Real-time data is transmitted to the data center via wired or wireless means and stored in a special database. The database system can ensure the integrity, security and accessibility of the data for subsequent data analysis and application.

[0031] Real-time data can be used for real-time monitoring of new energy systems, helping operation and maintenance personnel to promptly detect and handle system anomalies and ensure the stable operation of the system. By analyzing real-time data, the performance indicators of the new energy system, such as power generation efficiency and energy conversion efficiency, can be evaluated to provide data support for system optimization and upgrading. Based on the analysis of real-time data, potential failures and maintenance needs of the new energy system can be predicted, maintenance plans can be made in advance, and system failure rates and maintenance costs can be reduced. In smart grids, real-time data can be used for scheduling and optimization of new energy power generation, and dynamic adjustments can be made according to grid demand and the actual power generation capacity of the new energy system to improve energy utilization efficiency.

[0032] In summary, the real-time data of the new energy system is an important indicator reflecting the current operating status of the system. These data play an important role in the monitoring, performance evaluation, predictive maintenance and scheduling optimization of the new energy system.

[0033] In short, through the above steps, historical data and real-time data in the new energy system can be effectively obtained, providing strong support for subsequent abnormal monitoring, fault prediction and other tasks.

[0034] S202: Acquire operating index signals in the new energy system based on historical data, perform association rule mining on preset historical time windows of each operating index signal, and determine the historical operating abnormality coefficient based on the operating index difference of all adjacent moments; wherein, the operating index signal includes energy conversion efficiency, power density and discrete electrical signals.

[0035] First, collect operating indicator data such as energy conversion efficiency, power density, and discrete electrical signals from the historical database of the new energy system. These data may contain information such as timestamps and specific values. Clean the data to remove noise, missing values, or outliers. For missing values, interpolation methods (such as linear interpolation, mean value filling, etc.) can be used to supplement them; for outliers, they need to be identified and processed (such as deleted or replaced with reasonable values). Convert operating indicator data of different dimensions into a unified format for subsequent analysis. For each operating indicator, calculate the difference between adjacent moments. These differences reflect the trend of the indicator over time and help identify abnormal behavior. In addition to the difference, other features such as moving average, standard deviation, etc. can be constructed according to business needs to more comprehensively describe the operating status of the system.

[0036] A historical moment window size is preset to analyze the correlation between the operating indicator signals within the window. The selection of the window size should be determined according to the system characteristics and analysis requirements. Use data mining algorithms (such as Apriori algorithm, FP-Growth algorithm, etc.) to mine the association rules between the operating indicator signals within the historical moment window. These rules can help identify which operating indicators have strong correlations and how they jointly affect the operating status of the system. Based on the difference of the operating indicators of all adjacent moments, combined with the preset threshold or anomaly detection algorithm (such as statistical-based methods, machine learning models, etc.), identify the time points of operating anomalies. For each identified abnormal time point, calculate its historical operating anomaly coefficient. This coefficient can be defined based on the severity of the anomaly (such as the difference size, duration, etc.) to quantify the impact of the anomaly. Comprehensively analyze the mined association rules and calculated anomaly coefficients to understand the operating status and potential problems of the new energy system. Formulate corresponding operation and maintenance strategies and optimization measures based on the analysis results. For example, perform preventive maintenance for frequently occurring anomaly types; adjust system operating parameters according to association rules to improve efficiency, etc. Integrate the above process into the daily operation and maintenance of the new energy system, continuously monitor the operating indicator signals and update the analysis results. Continuously optimize anomaly detection models and operation and maintenance strategies based on new monitoring data and analysis results to improve system stability and efficiency.

[0037] Through the above steps, historical data can be effectively used to explore the correlation between operating indicator signals in the new energy system and identify potential abnormal behaviors, providing strong support for the operation, maintenance and optimization of the system.

[0038] Exemplarily, the method of acquiring the operation index signal in the new energy system according to the historical data, performing association rule mining of the preset historical time window of each operation index signal, and determining the historical operation abnormality coefficient according to the difference of the operation index at all adjacent times may include: Step 1: Perform discrete operations on the electrical signals in the operation index signal, and use the association rule mining algorithm to perform clustering operations on each discrete electrical signal. Step 2: Calculate the absolute value of the difference between the corresponding discrete electrical signals at any adjacent time, and use the sum of the absolute values ​​of all differences as the historical operation abnormality coefficient.

[0039] Specifically, first, the electrical signal in the operation index signal is discretized. Discretization is the process of mapping the continuously changing electrical signal value to a finite number of discrete values. This can be achieved by dividing the interval, using the threshold method or automatic discretization based on the clustering algorithm. The purpose of discretization is to simplify the data, reduce noise, and make subsequent processing more efficient.

[0040] The association rule mining algorithm is used to perform clustering operations on each discrete electrical signal, with the goal of identifying electrical signal patterns or frequent item sets. For example, when using association rule mining to analyze electrical signals in a similar way to clustering, the electrical signals can be first converted into item sets (for example, each discrete value is considered as an item), and then association rule mining can be applied to discover frequent item sets, which can indirectly reflect the pattern of the electrical signal. Alternatively, a clustering algorithm can be used to group discrete electrical signals, with each group representing an electrical signal pattern.

[0041] After discretization (and possible clustering), calculate the difference between the corresponding discrete electrical signals at any adjacent time (since it has been discretized, the difference here is actually a state change). Take the absolute value of these differences to eliminate the influence of positive and negative signs. The sum of the absolute values ​​of the differences between the discrete electrical signals at all adjacent times is taken as the historical operation anomaly coefficient. This coefficient reflects the degree of fluctuation or instability of the electrical signal during the historical time period. If the anomaly coefficient is high, it may mean that the system has experienced major changes or potential problems during this period.

[0042] S203: Determine real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system.

[0043] Specifically, determining the real-time normal probability distribution data corresponding to the new energy system may include: a. Analyze the relationship between the historical operation abnormality coefficient and the operating status of the new energy system, and identify the abnormal mode of the new energy system.

[0044] b. Determine the impact factors of abnormal modes of new energy systems on key indicators of new energy systems.

[0045] c. Based on the influencing factors and real-time data, a real-time normal probability distribution model corresponding to the new energy system is established to determine the real-time normal probability distribution data corresponding to the new energy system.

[0046] Specifically, first collect the long-term historical operation data of the new energy system, including operation indicators such as energy conversion efficiency, power density, electrical signals, etc. and the corresponding historical operation abnormality coefficients. Statistical methods such as Pearson correlation coefficient and Spearman rank correlation coefficient are used to analyze the correlation between the historical operation abnormality coefficient and each operation status indicator of the new energy system. Based on the results of the correlation analysis, combined with expert knowledge and system characteristics, identify the abnormal mode of the new energy system. Among them, the abnormal mode may be manifested as a specific indicator combination abnormality, a sudden change in time series, etc. For the identified abnormal modes, further analyze their specific impact on the key indicators of the new energy system (such as power generation, system efficiency, maintenance cost, etc.), and quantify the intensity and direction of these influencing factors through regression analysis, causal reasoning and other methods. In the quantification process, multiple influencing factors may be found, but not all factors are equally important for the establishment of a real-time normal probability distribution model. Therefore, factor screening is required to retain those factors that contribute significantly to the model prediction performance. Clean, transform and standardize the real-time data (including key indicators and possible other related variables) to meet the requirements of the model input. According to the characteristics of the data and the analysis objectives, select a suitable statistical or machine learning model to establish a real-time normal probability distribution model. It should be noted that more general probability distribution models, such as Gaussian mixture models, generalized linear models, etc., can also be used to approximate the normal distribution through transformation or parameter adjustment.

[0047] The selected model is trained using historical data and the selected influencing factors as training data. During the training process, the model parameters need to be adjusted to optimize the model performance, such as minimizing the prediction error, maximizing the likelihood function, etc. The model is verified using a data set that is not involved in the training to evaluate its prediction performance and generalization ability. If the model performance does not meet the requirements, return to the previous steps to adjust the model or reselect the model. Once the model is established and verified, it can be deployed to the real-time monitoring system of the new energy system. As new data is continuously generated, the model needs to be updated regularly to reflect the latest status and changes of the system. Using the trained model and real-time data input, the key indicators of the new energy system are predicted in real time. The probability distribution parameters (such as mean, variance, etc.) of the key indicators are extracted or calculated from the prediction results. These parameters represent the normal probability distribution data of the new energy system in real time. These data can be used for system status assessment, risk assessment, decision support, etc.

[0048] It should be noted that the real-time normal probability distribution model is dynamically updated using a sliding window technology to adapt to changes in the operating status of the new energy system over time.

[0049] Specifically, the real-time normal probability distribution model is dynamically updated using sliding window technology, which is an effective way to adapt to the time-varying operating status of the new energy system. The sliding window technology continuously moves the time window and updates the model based on the data in the window, thereby accurately capturing and predicting the real-time status of the new energy system.

[0050] According to the characteristics and analysis requirements of the new energy system, set a suitable time window size. The selection of the window size should be able to balance the sufficiency and real-time nature of the data, and should contain enough information to reflect the state of the system while maintaining a rapid response to system changes. Determine the time interval (step size) for each sliding of the window. The selection of the step size should be determined based on the speed of system changes and the frequency of data updates to ensure that the model can capture the latest state of the system in a timely manner. During the operation of the new energy system, key indicator data such as energy conversion efficiency, power density, and electrical signals are collected in real time. The collected real-time data is cleaned, converted, and standardized to meet the requirements of the model input. At the same time, according to the setting of the sliding window, a data subset within the window is extracted from the real-time data stream. Feature extraction is performed on the data subset within the window, and statistics (such as mean, standard deviation), frequency domain features, or other indicators that can reflect the state of the system are calculated. Using the data features within the window and possible historical information, the real-time normal probability distribution model is trained or updated, which can be achieved by retraining the entire model or only updating certain parameters of the model. Based on the updated model, the current state of the new energy system is predicted and the corresponding normal probability distribution data is output. These data can be used for system status assessment, anomaly detection, decision support, etc. According to the set sliding step, the time window is moved forward, and the above process of data collection, preprocessing, model update and result output is repeated. In this way, the model can continuously update its status according to the latest data, thereby realizing dynamic tracking and prediction of the real-time status of the new energy system.

[0051] In summary, the real-time normal probability distribution model is dynamically updated using sliding window technology, which can effectively adapt to the changes in the operating status of the new energy system over time and provide strong support for the real-time monitoring and prediction of the system.

[0052] S204: Perform abnormal situation monitoring on the current operation status of the new energy system according to the real-time normal probability distribution data.

[0053] Specifically, performing abnormal situation monitoring on the current operation of the new energy system according to the real-time normal probability distribution data may include: (1) Calculate the degree of abnormality of abnormal data in real-time normal probability distribution data; (ii) monitoring abnormal load data of the new energy system according to the degree of abnormality of the abnormal data; (III) Complete the abnormal operation monitoring of new energy based on the relationship between the monitored abnormal load data and the preset load threshold.

[0054] First, we need to identify the data types that constitute abnormal data. For example, in a normal probability distribution, data that is far from the mean and located at the tail of the distribution can be considered abnormal. This can be determined by calculating the distance of the data point from the mean, such as multiples of the standard deviation. For each data point identified as abnormal, its degree of abnormality can be further calculated. This can be expressed by calculating the deviation of the data point from the mean and converting it into a percentage or multiples of the standard deviation. The higher the degree of abnormality, the more the data point deviates from the normal state.

[0055] Correlating abnormal data in real-time normal probability distribution with load data of new energy systems involves matching various parameters monitored by the system with abnormal data to identify which load data may be affected by the abnormality. By analyzing the load data corresponding to the abnormal data, it is possible to identify which load data exhibit abnormal characteristics, which can include sudden increases or decreases in load, increased volatility, and mismatches with normal patterns.

[0056] Before or during the operation of the new energy system, a series of load thresholds are set according to the system design specifications, historical data analysis and expert experience. These thresholds are used to define the normal range and abnormal range of load data. The monitored abnormal load data is compared with the preset load threshold. If the abnormal load data exceeds the set threshold, or there is a significant difference with the threshold, it can be judged that the new energy system has an abnormal operation. Once an abnormal operation is detected, the system should immediately trigger the alarm mechanism and notify relevant personnel to handle it. At the same time, according to the nature and severity of the abnormality, the corresponding emergency response measures can be automatically or manually executed to reduce the impact of the abnormality on the system.

[0057] For example, in a normal probability distribution, data outside the mean plus or minus a few standard deviations can be identified as abnormal. For example, a Z-score is used to quantify the degree of abnormality of each data point, where the Z-score calculation formula is: The sample mean of the points, Represents the sample standard deviation of the data points.

[0058] Set a threshold, such as ,When the absolute value of the Z score exceeds this threshold, the data point is considered abnormal.,After determining the abnormal data points, it is possible to further analyze whether these abnormal data,correspond to the abnormal load of the new energy system.

[0059] For example, suppose there is a time series load data ,in, It is the time index. At the same time, the Z score data of the corresponding time point is obtained. , for each time point ,if , then check the corresponding load data load data If the load data If certain abnormal conditions are also met, such as a sudden increase or decrease exceeding a certain proportion, the load data will be marked as abnormal.

[0060] Finally, based on the relationship between the monitored abnormal load data and the preset load threshold, it is determined whether the new energy system is in an abnormal operating state. For example, one or more load thresholds (such as maximum load threshold, minimum load threshold, load change rate threshold, etc.) are set. If the monitored abnormal load data exceeds any preset load threshold, it is considered that the new energy system is currently in an abnormal operating state.

[0061] For example, the load change rate is calculated as follows: if 3 and the rate of change , it is judged as abnormal load; if the abnormal load data Maximum load threshold or If the minimum load threshold is exceeded, the new energy system is judged to be in an abnormal operating state.

[0062] It can be seen that the present invention first obtains historical data and real-time data in the new energy system; obtains the operating index signal in the new energy system based on the historical data, performs association rule mining on the preset historical time window of each operating index signal, and determines the historical operation abnormality coefficient according to the difference of the operating index at all adjacent times; determines the real-time normal probability distribution data corresponding to the new energy system; and finally, performs abnormal situation monitoring on the current operation of the new energy system based on the real-time normal probability distribution data. It can more accurately identify potential abnormal modes in the operation of the new energy system, introduce real-time data, ensure the immediacy of monitoring, and enable the system to be detected at the early stage of abnormality, thereby improving the comprehensiveness and accuracy of monitoring.

[0063] Another embodiment of the present application provides a new energy operation abnormality monitoring system, such as Figure 3 The structural diagram of a new energy operation abnormality monitoring system shown in FIG. 1 includes: An acquisition module 301 is used to acquire historical data and real-time data in the new energy system; The first execution module 302 is used to obtain the operation index signal in the new energy system according to the historical data, perform association rule mining of the preset historical time window of each operation index signal, and determine the historical operation abnormality coefficient according to the difference of the operation index at all adjacent times; wherein the operation index signal includes energy conversion efficiency, power density and discrete electrical signal; A determination module 303 is used to determine real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system; The second execution module 304 is used to perform abnormal situation monitoring on the current operation status of the new energy system according to the real-time normal probability distribution data.

[0064] Compared with the prior art, the present invention first obtains historical data and real-time data in the new energy system; obtains the operating index signal in the new energy system based on the historical data, performs association rule mining for the preset historical time window of each operating index signal, and determines the historical operation abnormality coefficient based on the difference of the operating index at all adjacent times; determines the real-time normal probability distribution data corresponding to the new energy system; and finally, performs abnormal situation monitoring on the current operation of the new energy system based on the real-time normal probability distribution data. It can more accurately identify the potential abnormal mode in the operation of the new energy system, introduce real-time data, ensure the immediacy of monitoring, enable the system to be detected at the early stage of abnormality, and improve the comprehensiveness and accuracy of monitoring.

[0065] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0066] Specifically, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0067] Specifically, in this embodiment, the processor may be configured to perform the following steps through a computer program: S201: Acquire historical data and real-time data in the new energy system; S202: acquiring operation index signals in the new energy system according to historical data, performing association rule mining of preset historical time windows for each operation index signal, and determining a historical operation abnormality coefficient according to the difference of operation indexes at all adjacent times; wherein the operation index signals include energy conversion efficiency, power density, and discrete electrical signals; S203: Determine real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system; S204: Perform abnormal situation monitoring on the current operation status of the new energy system according to the real-time normal probability distribution data.

[0068] Compared with the prior art, the present invention first obtains historical data and real-time data in the new energy system; obtains the operating index signal in the new energy system based on the historical data, performs association rule mining for the preset historical time window of each operating index signal, and determines the historical operation abnormality coefficient based on the difference of the operating index at all adjacent times; determines the real-time normal probability distribution data corresponding to the new energy system; and finally, performs abnormal situation monitoring on the current operation of the new energy system based on the real-time normal probability distribution data. It can more accurately identify the potential abnormal mode in the operation of the new energy system, introduce real-time data, ensure the immediacy of monitoring, enable the system to be detected at the early stage of abnormality, and improve the comprehensiveness and accuracy of monitoring.

[0069] An embodiment of the present invention further provides a storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0070] Specifically, in this embodiment, the above storage medium may be configured to store a computer program for performing the following steps: S201: Acquire historical data and real-time data in the new energy system; S202: acquiring operation index signals in the new energy system according to historical data, performing association rule mining of preset historical time windows for each operation index signal, and determining a historical operation abnormality coefficient according to the difference of operation indexes at all adjacent times; wherein the operation index signals include energy conversion efficiency, power density, and discrete electrical signals; S203: Determine real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system; S204: Perform abnormal situation monitoring on the current operation status of the new energy system according to the real-time normal probability distribution data.

[0071] Specifically, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0072] Compared with the prior art, the present invention first obtains historical data and real-time data in the new energy system; obtains the operating index signal in the new energy system based on the historical data, performs association rule mining for the preset historical time window of each operating index signal, and determines the historical operation abnormality coefficient based on the difference of the operating index at all adjacent times; determines the real-time normal probability distribution data corresponding to the new energy system; and finally, performs abnormal situation monitoring on the current operation of the new energy system based on the real-time normal probability distribution data. It can more accurately identify the potential abnormal mode in the operation of the new energy system, introduce real-time data, ensure the immediacy of monitoring, enable the system to be detected at the early stage of abnormality, and improve the comprehensiveness and accuracy of monitoring.

[0073] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, some steps can be performed in other sequences or simultaneously. For example, the actions and modules involved are not necessarily required by the present invention.

[0074] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0075] In the several embodiments provided by the present invention, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0076] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0078] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the above-mentioned methods of each embodiment of the present invention. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0079] The embodiments of the present invention are described in detail above. Specific examples are used herein 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 idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for monitoring abnormal operation of new energy, characterized in that: The method comprises: Obtain historical and real-time data in new energy systems; According to historical data, the operation index signal in the new energy system is obtained, and the association rule mining of the preset historical time window of each operation index signal is performed, and the historical operation abnormality coefficient is determined according to the difference of the operation index at all adjacent times; wherein the operation index signal includes energy conversion efficiency, power density and discrete electrical signal; Determining real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system; According to the real-time normal probability distribution data, abnormal situation monitoring is performed on the current operation situation of the new energy system.

2. The method according to claim 1, characterized in that: The acquisition of historical data and real-time data in the new energy system includes: Obtain historical time series data of the new energy system and preprocess the historical time series data; The preprocessed historical time series data is randomly divided into samples of different time intervals; each sample corresponds to the historical operation data at that moment; In response to the presence of abnormal values ​​in the currently obtained historical operation data, the abnormal values ​​are removed, the historical operation data is dimensionally expanded, and the operation characteristic information contained in each historical moment is added to generate historical data in the new energy system.

3. The method according to claim 2, characterized in that The acquisition of historical data and real-time data in the new energy system includes: Taking the current time as a reference, the load data of the new energy system collected by the load collection device of the new energy system is recorded as the real-time data of the new energy system.

4. The method according to claim 3, characterized in that The method of obtaining the operation index signal in the new energy system according to the historical data, performing association rule mining of the preset historical time window of each operation index signal, and determining the historical operation abnormality coefficient according to the difference of the operation index at all adjacent times includes: Perform discrete operations on the electrical signals in the operation index signals, and use association rule mining algorithms to perform clustering operations on each discrete electrical signal; Calculate the absolute value of the difference between the discrete electrical signals corresponding to any adjacent moments, and take the sum of the absolute values ​​of all the differences as the historical operation abnormality coefficient.

5. The method according to claim 4, characterized in that The determining of real-time normal probability distribution data corresponding to the new energy system includes: Analyze the relationship between historical operation abnormality coefficient and the operation status of new energy system, and identify the abnormal mode of new energy system; Determine the impact factors of abnormal modes of new energy systems on key indicators of new energy systems; According to the influencing factors and the real-time data, a real-time normal probability distribution model corresponding to the new energy system is established to determine the real-time normal probability distribution data corresponding to the new energy system.

6. The method according to claim 5, characterized in that The real-time normal probability distribution model is dynamically updated using a sliding window technology to adapt to changes in the operating status of the new energy system over time.

7. The method according to claim 6, characterized in that The performing abnormal situation monitoring on the current operation situation of the new energy system according to the real-time normal probability distribution data includes: Calculate the abnormal degree of abnormal data in real-time normal probability distribution data; monitoring abnormal load data of the new energy system according to the abnormality degree of the abnormal data; Based on the relationship between the monitored abnormal load data and the preset load threshold, complete the abnormal monitoring of new energy operation.

8. A new energy operation abnormality monitoring system, characterized in that: The system comprises: Acquisition module, used to acquire historical data and real-time data in the new energy system; The first execution module is used to obtain the operation index signal in the new energy system according to the historical data, perform the association rule mining of the preset historical time window of each operation index signal, and determine the historical operation abnormality coefficient according to the difference of the operation index at all adjacent times; wherein the operation index signal includes energy conversion efficiency, power density and discrete electrical signal; A determination module, used for determining real-time normal probability distribution data corresponding to the new energy system, wherein the normal probability distribution data is obtained based on the historical operation abnormality coefficient and the real-time data of the new energy system; The second execution module is used to perform abnormal situation monitoring on the current operation status of the new energy system according to the real-time normal probability distribution data.

9. An electronic device, comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 7 when executed.

Images