Photovoltaic abnormal data identification method and device, electronic equipment and storage medium

Abstract

The invention discloses a photovoltaic abnormal data identification method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining photovoltaic data, determining a suspected abnormal point according to the absolute deviation of the median of a time local sequence of the photovoltaic data, and taking the time of the suspected abnormal point as a time sequence interval division reference, segmenting the photovoltaic data into a plurality of photovoltaic data subsequences in a time sequence; screening out suspected abnormal points on the photovoltaic data subsequences, and judging the suspected abnormal points in a threshold range as abnormal points; and carrying out time sequence interval cutting on the photovoltaic data subsequences where the abnormal points are located, constructing a boundary power constraint according to the power change of the photovoltaic data, and identifying the abnormal data from the cut photovoltaic data subsequences through the boundary power constraint. According to the invention, the photovoltaic data is divided into a plurality of subsequences, and independent analysis is carried out for different subsequences, so that the abnormal data monitoring speed and detection accuracy can be improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy power generation, and particularly to a method, device, electronic device and storage medium for identifying abnormal photovoltaic data. Background Art

[0002] As a clean and renewable form of energy utilization, photovoltaic power generation has become one of the important directions for global energy transformation due to its extensive resource sustainability, environmental protection and low-carbon characteristics, as well as reduced power generation costs. With the acceleration of energy structure adjustment in various regions, photovoltaic power generation has not only continued to develop in the field of large-scale centralized photovoltaics, but also has seen rapid development in local photovoltaic systems. Its wide application has played a key role in reducing greenhouse gas emissions such as fire protection, increasing solar energy dependence, and promoting the development of a green and low-carbon economy.

[0003] In recent years, the innovation and application of photovoltaic technology have continuously promoted the progress of the new energy industry, resulting in significant improvements in the power generation efficiency, intelligent operation and maintenance, and grid connection management of photovoltaic systems. In this context, the advantages of flexible layout, diverse access methods, and high energy utilization efficiency of photovoltaic systems have become one of the important directions for the development of photovoltaic power generation. The operation depends on accurate real-time data, and the quality of these data directly affects key links such as photovoltaic power generation performance evaluation, power prediction, and grid connection control. Especially for intelligent photovoltaic systems, accurate and reliable data is not only an important basis for optimizing the operation of photovoltaic systems and improving power generation efficiency, but also an important guarantee for supporting the large-scale access of photovoltaic power to the grid and enhancing the stability of the power system.

[0004] In the actual process of collecting photovoltaic power generation data, due to the influence of multiple factors such as system equipment status, environmental conditions, and measurement tasks, a large number of abnormal data often appear. For example, environmental factors such as cloud cover changes, temperature fluctuations, climate, lightning strikes, etc. can cause mirror fluctuations in the power generation power of photovoltaic modules. Equipment failures in the system, such as sensor failures, inverter failures, electrical connection failures, etc., can also lead to data anomalies. The existence of these abnormal data not only affects the performance evaluation, power prediction, and subsequent optimal scheduling analysis of photovoltaic power generation systems, but also makes data processing and predictive intelligent modeling more complex.

[0005] The existence of abnormal data poses challenges to the operation status evaluation, fault diagnosis, and optimal control of photovoltaic systems. Existing technologies usually classify abnormal data and then identify abnormal data according to judgment criteria. However, it is difficult for existing technologies to quickly identify abnormal data of weak faults. If abnormal data cannot be identified and cleared in time, it may lead to incorrect power prediction results, which in turn affects the grid dispatching strategy and even the safe and stable operation of the entire power system. Summary of the Invention

[0006] The present invention provides a method, device, electronic device, and storage medium for identifying abnormal photovoltaic data, which solves the problem of inability to timely identify abnormal data of weak faults.

[0007] According to one aspect of the present invention, there is provided a method for identifying abnormal photovoltaic data, including: obtaining photovoltaic data, determining suspected abnormal points according to the median absolute deviation of the time local sequence of the photovoltaic data, using the moment of the suspected abnormal points as the benchmark for dividing the time sequence interval, and splitting the photovoltaic data into multiple sub-sequences of photovoltaic data in the time sequence; screening out the suspected abnormal points where there are local maximum values in the adjacent points on the sub-sequence of photovoltaic data, performing threshold judgment on the screened suspected abnormal points, and determining the suspected abnormal points within the threshold range as abnormal points; cutting the time sequence interval of the sub-sequence of photovoltaic data where the abnormal points are located, constructing boundary power constraints according to the power change of the photovoltaic data, and identifying abnormal data through the boundary power constraints of the cut sub-sequence of photovoltaic data.

[0008] Furthermore, the calculation formula for the median absolute deviation of the time local sequence of the photovoltaic data is:

[0009]

[0010] where, is the suspected abnormal point, X i =[x1, x2,..., x n is the time local sequence, λ is a proportional parameter, m is the sliding window width, T is the sample size, k is the number of times the sliding window slides, and the value range of k is [0, T - 1].

[0011] Furthermore, the specific formula for performing threshold judgment on the screened suspected abnormal points is:

[0012] η = γ(max f(x i ) - min f(x i )) 0≤i≤T ;

[0013] where, f(x) is the photovoltaic data value corresponding to the time local sequence point x, and γ is an adjustable parameter.

[0014] Furthermore, the step of screening out the suspected abnormal points where there are local maximum values in the adjacent points on the sub-sequence of photovoltaic data specifically includes: calculating the difference values of the adjacent points around the suspected abnormal points on the sub-sequence of photovoltaic data; if the difference values less than the difference value of the suspected abnormal point time sequence are all greater than zero within the preset interval, and the difference values greater than the difference value of the suspected abnormal point time sequence are all less than zero within the preset interval, then it is determined that there are local maximum values in the adjacent points of the suspected abnormal point; screening out the suspected abnormal points with local maximum values within the preset interval.

[0015] Furthermore, the calculation formula for the boundary power constraint is:

[0016]

[0017] s min ≤s(f(x i ) - f(x j )) ≤ s max ;

[0018]

[0019] where s max is the upper threshold of the change amount of photovoltaic data, and s max represents the maximum constraint of the change of photovoltaic data; s min is the lower threshold of the change amount of photovoltaic data, and s min represents the minimum constraint of the change of photovoltaic data.

[0020] Furthermore, after identifying the abnormal data, calculate the constraint range of the abnormal data according to the interquartile range method, and eliminate the misjudged abnormal data according to the constraint range.

[0021] Furthermore, the constraint range is:

[0022] f gmin (x) ≤ f(x i ) ≤ f gmax (x);

[0023] where f gmin (x) is the photovoltaic data value obtained by the lower standardized interquartile range method, and f gmax (x) are the photovoltaic data values obtained by the upper standardized interquartile range method respectively.

[0024] According to another aspect of the present invention, there is provided a photovoltaic abnormal data identification device, comprising: a data acquisition module, which is used to acquire photovoltaic data, determine suspected abnormal points according to the median absolute deviation of the time local sequence of the photovoltaic data, and use the moments of the suspected abnormal points as the time series interval division benchmark to divide the photovoltaic data into multiple segments of photovoltaic data subsequences in time series; a data screening module, which is used to screen out the suspected abnormal points with local maximum values in the adjacent points on the photovoltaic data subsequences, perform threshold judgment on the screened suspected abnormal points, and judge the suspected abnormal points within the threshold range as abnormal points; a data identification module, which is used to cut the time series interval of the photovoltaic data subsequence where the abnormal point is located, construct a boundary power constraint according to the power change of the photovoltaic data, and identify the abnormal data through the boundary power constraint of the cut photovoltaic data subsequence.

[0025] According to another aspect of the present invention, there is provided an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor;

[0026] Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any photovoltaic abnormal data identification method in the embodiments of the present invention.

[0027] According to another aspect of the present invention, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute any photovoltaic abnormal data identification method in the embodiments of the present invention.

[0028] According to the technology of the present invention, by screening suspected abnormal points through the absolute deviation method based on the time local sequence, and combining local feature analysis and threshold judgment, the abnormal data of weak faults can be accurately identified, misjudgment and missed judgment can be reduced, and the overall quality of photovoltaic data can be improved. By adopting the guiding interval division method, the photovoltaic data is divided into multiple subsequences, and independent analysis is carried out for different subsequences, which can effectively reduce the overall data fluctuation while improving the abnormal data monitoring speed, and enhance the stability of abnormal data acquisition and modification. The accuracy of weak fault abnormal detection is improved. Combining the boundary power constraint, by constructing a reasonable power change range, abnormal data can be accurately identified and repaired, preventing data deviation caused by measurement tasks, equipment failures or external interferences from affecting the overall operation of the photovoltaic system.

[0029] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The accompanying drawings are used to better understand the present solution and do not constitute a limitation to the present invention. Among them:

[0031] Figure 1 is a flowchart of the method for identifying photovoltaic abnormal data provided by an embodiment of the present invention;

[0032] Figure 2 is a schematic structural diagram of the device for identifying photovoltaic abnormal data provided by an embodiment of the present invention;

[0033] Figure 3 is a schematic diagram of the electronic device and storage medium of an embodiment of the present invention.

[0034] In the figure, 100 is the device for identifying photovoltaic abnormal data; 11 is the data acquisition module; 12 is the data screening module; 13 is the data identification module; 200 is the electronic device; 201 is the calculation unit; 202 is the ROM; 203 is the RAM; 204 is the bus; 205 is the I / O interface; 206 is the input unit; 207 is the output unit; 208 is the storage unit; 209 is the communication unit. Detailed implementation manners

[0035] The following describes exemplary embodiments of the present invention with reference to the accompanying drawings. Various details of the embodiments of the present invention are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the present invention. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0036] As Figure 1 shown, an embodiment of the present invention discloses a method for identifying photovoltaic abnormal data, including: S1, acquiring photovoltaic data, determining suspected abnormal points according to the median absolute deviation of the time local sequence of the photovoltaic data, using the moment of the suspected abnormal points as the benchmark for dividing the time sequence interval, and splitting the photovoltaic data into multiple segments of photovoltaic data subsequences in time sequence; S2, screening out the suspected abnormal points with local maximum values of adjacent points on the photovoltaic data subsequences, performing threshold judgment on the screened suspected abnormal points, and determining the suspected abnormal points within the threshold range as abnormal points; S3, cutting the time sequence interval of the photovoltaic data subsequence where the abnormal points are located, constructing boundary power constraints according to the power change of the photovoltaic data, and identifying the abnormal data through the boundary power constraints of the cut photovoltaic data subsequence.

[0037] The photovoltaic abnormal data identification method of the present application can accurately identify abnormal data, reduce misjudgment and missed judgment, and improve the overall quality of photovoltaic data by screening suspected abnormal points through the deviation absolute deviation method based on the time local sequence and combining local feature analysis and threshold judgment. By adopting the guiding interval division method, the photovoltaic data is divided into multiple subsequences, and independent analysis is carried out for different subsequences, which can effectively reduce the overall data fluctuation, enhance the stability of abnormal data acquisition and modification, and improve the accuracy of weak fault abnormal detection. Combining the modified boundary power constraint, by constructing a reasonable power change range, abnormal data can be accurately identified and repaired to prevent data deviation caused by measurement tasks, equipment failures or external interferences from affecting the overall operation of the photovoltaic system.

[0038] The present invention realizes the identification of photovoltaic abnormal data through an algorithm. First, the binary abnormal point detection algorithm is used to disperse a long time series into multiple subsequences through abnormal data points for better identification of abnormal data. When identifying abnormal data points, it is necessary to judge the peak value and give a threshold to judge abnormal data points and non-abnormal data points; finally, the segmented self-sequences use power-boundary constraints to further identify mutation abnormal data and continuous same-value abnormal data. The present invention accurately locates abnormal data points, timely discovers potential fault hazards in the photovoltaic system, ensures the stable operation of the system, greatly reduces the probability of system failures caused by unrecognized abnormal data, reduces unnecessary downtime maintenance time, improves the efficiency of photovoltaic power generation, and brings economic benefits to distributed photovoltaics.

[0039] The data identification method based on power-boundary constraints of binary abnormal point detection of the present invention considers an abnormal data identification method for evaluating the operation performance of photovoltaic sensors for distributed photovoltaics, especially for weak fault data in photovoltaic data, to improve the overall quality of photovoltaic data measured by photovoltaic power stations, so as to provide more accurate data support for subsequent photovoltaic system performance analysis and optimization. The present invention effectively solves the problem of abnormal data existing in the actual data acquisition process of photovoltaic power stations. The method aims to accurately identify suspected abnormal points in the data by introducing abnormal point detection technology, thereby improving the authenticity and reliability of the data. Combining the boundary power constraint can more precisely constrain the reasonable range of the data and prevent abnormal values caused by measurement errors or sensor failures from affecting the overall performance evaluation of the photovoltaic power station. Through this method, not only the data quality of the photovoltaic power station is improved, but also reliable data support is provided for subsequent photovoltaic system operation optimization and fault diagnosis, thus promoting the intelligent management and efficient operation of distributed photovoltaic systems.

[0040] By using the median absolute deviation method of the time-local sequence to preliminarily screen for suspected abnormal points, and combining the local peak characteristics and threshold judgment mechanism of the obtained points, the identification of abnormal data is made more accurate, reducing the possibility of misjudgment and missed judgment. By dividing the photovoltaic data into intervals, the data is segmented into multiple subsequences, and abnormal points are further analyzed within the range of the subsequences, so that the detection of abnormal data is not affected by the overall data fluctuation, improving the reliability of the system. The segmented processing method is adopted to divide the photovoltaic data into multiple subsequences, and abnormal data screening and data analysis are carried out segment by segment. Compared with directly detecting globally, it can reduce the computational complexity, improve the order efficiency, and is more suitable for large-scale photovoltaic data processing.

[0041] Specifically, the suspected abnormal points of the photovoltaic data are mutation points, offset points, and noise points. Among them, the mutation points are caused by the drastic change of solar radiation intensity, instantaneous equipment failure, and power grid fluctuation. The offset points are caused by sensor drift resulting in measurement value offset and long-term equipment failure. The noise points are caused by sensor accuracy problems or interference, error codes or abnormal signal interference during data transmission.

[0042] In an alternative embodiment of the present invention, the formula for the median absolute deviation of the time-local sequence of the photovoltaic data is:

[0043]

[0044] Where is the suspected abnormal point, X i =[x1, x2,..., x n is the time-local sequence, λ is the proportional parameter, and the general range of λ is [0.2 - 0.5]. In this embodiment, the value of λ is λ = 0.3, m is the sliding window width, the value of m is related to the sample size T, T is the sample size, k is the number of times the sliding window slides, and the value range of k is [0, T - 1].

[0045] Through the calculation method of the median absolute deviation based on the time-local sequence, the accurate identification of abnormal points in the photovoltaic data is realized. Using the statistical index of median absolute deviation for anomaly detection increases the robustness of the mean and can effectively reduce the limit stability. Using the median absolute deviation as the statistical index for anomaly detection can effectively reduce the influence of extreme values on data analysis and improve the stability of abnormal data detection compared with the mean and standard deviation methods.

[0046] Calculate the time-local sequence through a sliding window method, enabling anomaly detection to better capture the characteristic changes of photovoltaic data at different time scales, thereby improving the ability to identify local anomalies and adapting to the dynamic change characteristics of photovoltaic power generation data. By calculating the median absolute deviation and providing some corresponding local extreme points as suspected anomaly points, the screening process of anomaly points becomes more accurate, reducing misjudgments and improving the accuracy of photovoltaic data anomaly detection.

[0047] In an alternative embodiment of the present invention, the specific formula for threshold judgment of the screened suspected anomaly points is:

[0048] η = γ(maxf(x i ) - minf(x i )) 0≤i≤T ;

[0049] Where f(x) is the photovoltaic data value corresponding to the time-local sequence point x, γ is an adjustable parameter, and γ can be adjusted to obtain a relatively optimal threshold. In this embodiment, γ = 0.5.

[0050] By introducing the adjustable parameter γ, the threshold range can be dynamically adjusted according to different application scenarios, improving the compatibility and flexibility of the method. By setting the threshold in combination with the local data change range through the adjustable parameter γ, the detection standard can be optimized according to the data characteristics of different photovoltaic power stations, meeting the anomaly detection requirements under different environmental and equipment conditions.

[0051] In an alternative embodiment of the present invention, on the photovoltaic data subsequence, screen out the suspected anomaly points where there is a local maximum in the neighboring points, specifically: calculate the difference value of the neighboring points around the suspected anomaly point on the photovoltaic data subsequence; if the difference values less than the suspected anomaly point time series are all greater than zero within the preset interval, and the difference values greater than the suspected anomaly point time series are all less than zero within the preset interval, then it is judged that there is a local maximum in the neighboring points of the suspected anomaly point; screen out the suspected anomaly points where there is a local maximum in the neighboring points. The specific calculation formula of the difference value is as follows:

[0052]

[0053] Where df(i,j) is the difference between two points, df(i,j) = 1, 0, -1 respectively represent that the statistic shows an upward, stable, and downward trend, and f(x) is the photovoltaic data value corresponding to point x.

[0054] By calculating the difference values between the suspected abnormal points and the control abnormal points, only the points that meet specific conditions within the preset interval are identified and screened as local abnormal points, thereby reducing misjudgments and improving the accuracy of abnormal data screening. It can accurately capture local mutation points in photovoltaic data, such as spike anomalies caused by sensor failures, external interferences, or data transmission interruptions, thus avoiding the influence of such abnormal values on the accuracy of photovoltaic power prediction and system analysis. Since photovoltaic data is affected by environmental factors such as cloud changes and intermittent shading and shows short-term fluctuations, by combining differential value analysis, the accuracy of the power data of the photovoltaic system is ensured.

[0055] In an alternative embodiment of the present invention, the calculation formula for the boundary power constraint is:

[0056]

[0057] s min ≤s(f(x i ) - f(x j )) ≤ s max ;

[0058]

[0059] Where s max is the upper threshold of the change amount of photovoltaic data, and s max represents the maximum constraint of the change of photovoltaic data; s min is the lower threshold of the change amount of photovoltaic data, and s min represents the minimum constraint of the change of photovoltaic data.

[0060] By calculating the maximum and minimum constraint values of the change of photovoltaic data, the change range of the data is always within a reasonable range, thereby reducing the influence of abnormal fluctuations on the operation of the photovoltaic system and ensuring the stability and reliability of the data. After setting the boundary constraint of the power change, abnormal data points beyond the reasonable range can be identified, and abnormal data caused by sensor readings, environmental changes, or equipment failures can be effectively removed, improving the quality of photovoltaic data. By setting reasonable upper and lower limits for the data change, the prediction model can be prevented from being interfered by abnormal data during the training process, thereby improving the accuracy of photovoltaic power prediction and providing more reliable data support for intelligent scheduling and optimal control. By restricting the change range of photovoltaic data, the grid-connected power curve becomes smoother, reducing the unstable influence on the power grid caused by power fluctuations, thereby optimizing the grid dispatching and improving the safety and operation efficiency of the power system.

[0061] In an alternative embodiment of the present invention, after identifying abnormal data, the constraint range of the abnormal data is calculated according to the interquartile range method, and the misjudged abnormal data is removed according to the constraint range. The constraint range is:

[0062] fgmin f(x) ≤ f(x i ) ≤ f gmax (x);

[0063] Wherein, f gmin (x) is the photovoltaic data value obtained by the lower standardized interquartile range method, and f gmax (x) are the photovoltaic data values obtained by the upper standardized interquartile range method, respectively.

[0064] By calculating the reasonable fluctuation range of data through the interquartile range method, it is possible to effectively distinguish the true abnormal data from the extreme values in the normal data, reduce the misjudgment situation, and improve the accuracy of abnormal data identification. Directly eliminating all suspected abnormal points may lead to misjudgment. By using the interquartile range method to set a reasonable constraint range, some data that are actually normal but deviate from the mean can be retained, preventing the miselimination of normal data and improving the overall quality and credibility of the data. By clearly discriminating and removing incorrect data, the data used to train the photovoltaic power prediction model becomes more accurate, reducing the interference of abnormal data on the prediction model, thereby improving the accuracy and stability of photovoltaic power prediction. This method can adaptively adjust the screening criteria for abnormal data in the case of large data changes or complex system operating environments, ensuring the effectiveness of abnormal detection, and is applicable to photovoltaic systems in different operating environments. At the same time, it ultimately ensures the data accuracy of grid connection management, makes the grid-connected power of photovoltaic more stable, and improves the security and reliability of the power grid.

[0065] As Figure 2 shown, the photovoltaic abnormal data identification device 100 may include:

[0066] A data acquisition module 11, which is used to acquire photovoltaic data, determine suspected abnormal points according to the median absolute deviation of the time local sequence of the photovoltaic data, and use the moments of the suspected abnormal points as the benchmark for dividing the time series interval to divide the photovoltaic data into multiple segments of photovoltaic data subsequences in time series;

[0067] A data screening module 12, which is used to screen out the suspected abnormal points with local maximum values of adjacent points on the photovoltaic data subsequence, and perform a threshold judgment on the screened suspected abnormal points, and judge the suspected abnormal points within the threshold range as abnormal points;

[0068] A data identification module 13, which is used to cut the time series interval of the photovoltaic data subsequence where the abnormal point is located, perform abnormal data identification and analysis on the cut subsequence, and identify abnormal data according to the boundary power constraint constructed based on the power change.

[0069] For the specific functions and example descriptions of the modules and sub-modules of the device according to the embodiments of the present invention, reference may be made to the relevant descriptions of the corresponding steps in the above method embodiments, which will not be elaborated herein.

[0070] In the technical solution of the present invention, the acquisition, storage, and application of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0071] According to an embodiment of the present invention, the present invention also provides an electronic device, a readable storage medium, and a computer program product.

[0072] Figure 3 FIG. shows a schematic block diagram of an exemplary electronic device 200 that can be used to implement embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital assistant, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0073] As Figure 3 shown, the device 200 includes a computing unit 201, which can execute various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 202 or a computer program loaded from a storage unit 208 into a random access memory (RAM) 203. In the RAM 203, various programs and data required for the operation of the device 200 can also be stored. The computing unit 201, the ROM 202, and the RAM 203 are connected to each other via a bus 204. An input / output (I / O) interface 205 is also connected to the bus 204.

[0074] A plurality of components in the device 200 are connected to the I / O interface 205, including: an input unit 206, such as a keyboard, a mouse, etc.; an output unit 207, such as various types of displays, speakers, etc.; a storage unit 208, such as a disk, an optical disc, etc.; and a communication unit 209, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 209 allows the device 200 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0075] The computing unit 201 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 201 executes the various methods and processes described above, such as a method for identifying photovoltaic abnormal data. For example, in some embodiments, a method for identifying photovoltaic abnormal data can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as the storage unit 208. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 200 via the ROM 202 and / or the communication unit 209. When the computer program is loaded into the RAM 203 and executed by the computing unit 201, one or more steps of a method for identifying photovoltaic abnormal data described above can be executed. Alternatively, in other embodiments, the computing unit 201 can be configured to execute a method for identifying photovoltaic abnormal data in any other suitable manner (e.g., by means of firmware).

[0076] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-a-chip systems (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0077] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0078] In the context of the present invention, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0079] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0080] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0081] A computer system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, can also be a server of a distributed system, or a server incorporating a blockchain.

[0082] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and no limitations are imposed herein.

[0083] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for identifying abnormal photovoltaic data, characterized in that, Including: Obtain photovoltaic data, determine suspected abnormal points according to the median absolute deviation of the time local sequence of the photovoltaic data, use the moment of the suspected abnormal points as the benchmark for dividing the time sequence interval, and divide the photovoltaic data into multiple segments of photovoltaic data subsequences in time sequence; On the photovoltaic data subsequences, screen out the suspected abnormal points where the adjacent points have local maximum values, perform threshold judgment on the screened suspected abnormal points, and judge the suspected abnormal points within the threshold range as abnormal points; Cut the time sequence interval of the photovoltaic data subsequence where the abnormal point is located, construct a boundary power constraint according to the power change of the photovoltaic data, and identify the abnormal data through the boundary power constraint for the cut photovoltaic data subsequence.

2. The method according to claim 1, wherein The calculation formula for the median absolute deviation of the time local sequence of the photovoltaic data is: Among them, is the suspected abnormal point, X i = [x1, x2,..., x n is the time local sequence, λ is the proportional parameter, m is the sliding window width, T is the sample size, k is the number of times the sliding window slides, and the value range of k is [0, T - 1].

3. The method according to claim 2, wherein The specific formula for performing threshold judgment on the screened suspected abnormal points is: η = γ(max f(x i ) - min f(x i )) 0≤i≤T ; Where f(x) is the photovoltaic data value corresponding to the time local sequence point x, and γ is an adjustable parameter.

4. The method according to claim 1, characterized in that, The screening out of the suspected abnormal points where the adjacent points have local maximum values on the photovoltaic data subsequences is specifically: Calculate the difference values of the adjacent points around the suspected abnormal point on the photovoltaic data subsequence; If the difference values less than the time sequence of the suspected abnormal point within the preset interval are all greater than zero, and the difference values greater than the time sequence of the suspected abnormal point within the preset interval are all less than zero, then it is judged that there is a local maximum value in the adjacent points of the suspected abnormal point; Screen out the suspected abnormal points with local maximum values within the preset interval.

5. The method according to claim 1, wherein The calculation formula for the boundary power constraint is: s min ≤ s(f(x i ) - f(x j )) ≤ s max ; Among them, s max is the upper threshold of the photovoltaic data change amount, and s max represents the maximum constraint of the photovoltaic data change; s min is the lower threshold of the photovoltaic data change amount, and s min represents the minimum constraint of the photovoltaic data change.

6. The method according to claim 1, wherein After identifying the abnormal data, calculate the constraint range of the abnormal data according to the interquartile range method, and eliminate the misjudged abnormal data according to the constraint range.

7. The method according to claim 6, characterized in that, The constraint range is: f gmin (x) ≤ f(x i ) ≤ f gmax (x); Among them, f gmin (x) is the photovoltaic data value obtained by the lower standardized interquartile range method, and f gmax (x) are the photovoltaic data values obtained by the upper standardized interquartile range method respectively.

8. A photovoltaic abnormal data identification device, characterized in that, Including: A data acquisition module, which is used to obtain photovoltaic data, determine suspected abnormal points according to the median absolute deviation of the time local sequence of the photovoltaic data, use the moment of the suspected abnormal points as the benchmark for dividing the time sequence interval, and divide the photovoltaic data into multiple segments of photovoltaic data subsequences in time sequence; A data screening module, which is used to screen out the suspected abnormal points where the adjacent points have local maximum values on the photovoltaic data subsequences, perform threshold judgment on the screened suspected abnormal points, and judge the suspected abnormal points within the threshold range as abnormal points; A data identification module, which is used to cut the time sequence interval of the photovoltaic data subsequence where the abnormal point is located, construct a boundary power constraint according to the power change of the photovoltaic data, and identify the abnormal data through the boundary power constraint for the cut photovoltaic data subsequence.

9. An electronic device, characterized in that, Including: At least one processor, and a memory communicatively connected to the at least one processor; Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method according to any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to make the computer execute the method according to any one of claims 1-7.

Images