Fault detection method and system applied to photovoltaic energy storage equipment

Abstract

The invention discloses a fault detection method and system applied to photovoltaic energy storage equipment. The method comprises the following steps: firstly, arranging time sequences of operating parameters of each photovoltaic energy storage device to obtain a set of operating parameter time sequence input matrixes of the photovoltaic energy storage devices; secondly, performing feature extraction on each photovoltaic energy storage equipment operation parameter time sequence input matrix to obtain a set of photovoltaic energy storage equipment operation parameter time sequence correlation feature vectors; calculating the operation mode offset degree semantic measurement coefficient of the photovoltaic energy storage equipment operation parameter time sequence correlation feature vector of the analyzed photovoltaic energy storage equipment and the set of the operation parameter time sequence correlation feature vectors of all other photovoltaic energy storage equipment in the set of the photovoltaic energy storage equipment operation parameter time sequence correlation feature vectors; and determining whether the analyzed photovoltaic energy storage equipment has a fault or not based on the operation mode offset degree semantic measurement coefficient. Therefore, the safety and reliability of the equipment can be improved, and the maintenance cost and risk are reduced.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic energy storage, and more specifically, to a fault detection method and system applied to photovoltaic energy storage equipment. Background Art

[0002] Photovoltaic energy storage equipment is a device that uses solar energy to generate electricity and stores excess electricity. It can provide a stable power supply at night or on rainy days when power generation is not possible. It has the advantages of being clean, efficient, and renewable. However, due to the long-term operation of photovoltaic energy storage equipment and the influence of the external environment, there may be a risk of failure, such as battery overcharge, over-discharge, short circuit, overheating, etc. These failures will affect the performance and life of the equipment, and even cause serious accidents such as fire and explosion. Therefore, it is very necessary to perform fault detection on photovoltaic energy storage equipment.

[0003] However, since the operating failures of photovoltaic energy storage equipment can be reflected in many aspects from different operating parameters of the equipment, and the operating environment of photovoltaic energy storage equipment is usually more complex, traditional fault detection methods based on thresholds or rules are difficult to adapt to this complex operating environment, and it is also difficult to use the mutual correlation between different operating parameters to perform equipment fault detection, which is prone to false alarms or missed alarms.

[0004] Therefore, an optimized fault detection scheme for photovoltaic energy storage devices is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The present application provides a fault detection method and system for photovoltaic energy storage equipment, which can monitor the different operating parameters of each photovoltaic energy storage equipment in the photovoltaic energy storage equipment cluster, and use the correlation between multiple motion parameters of each device in the photovoltaic energy storage equipment cluster to comprehensively perform fault detection of the photovoltaic energy storage equipment analyzed in the cluster, thereby improving the fault detection and judgment accuracy of the photovoltaic energy storage equipment, thereby providing accurate fault diagnosis results.

[0006] According to one aspect of the present application, a fault detection method for a photovoltaic energy storage device is provided, which comprises:

[0007] Acquire a time series of operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, wherein the operating parameters include voltage value, current value, temperature value and power value;

[0008] Arranging the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster into photovoltaic energy storage device operating parameter timing input matrices according to the time dimension and the operating parameter sample dimension to obtain a set of photovoltaic energy storage device operating parameter timing input matrices;

[0009] Performing feature extraction on each photovoltaic energy storage device operating parameter timing input matrix in the set of photovoltaic energy storage device operating parameter timing input matrices by using an operating mode feature extractor based on a deep neural network model to obtain a set of photovoltaic energy storage device operating parameter timing associated feature vectors;

[0010] Extracting a photovoltaic energy storage device operating parameter timing-related feature vector of the analyzed photovoltaic energy storage device from the set of photovoltaic energy storage device operating parameter timing-related feature vectors as a query feature vector, and calculating an operating mode deviation semantic metric coefficient between the query feature vector and a set of all other photovoltaic energy storage device operating parameter timing-related feature vectors in the set of photovoltaic energy storage device operating parameter timing-related feature vectors; and

[0011] Based on the operation mode deviation semantic metric coefficient, it is determined whether the analyzed photovoltaic energy storage device has a fault.

[0012] According to another aspect of the present application, a fault detection system for photovoltaic energy storage equipment is provided, comprising:

[0013] A data acquisition module, used to acquire a time series of operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, wherein the operating parameters include voltage value, current value, temperature value and power value;

[0014] A matrixing module, for arranging the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster into a photovoltaic energy storage device operating parameter timing input matrix according to the time dimension and the operating parameter sample dimension to obtain a set of photovoltaic energy storage device operating parameter timing input matrices;

[0015] A feature extraction module, used to extract features from each photovoltaic energy storage device operating parameter timing input matrix in the set of photovoltaic energy storage device operating parameter timing input matrices through an operating mode feature extractor based on a deep neural network model to obtain a set of photovoltaic energy storage device operating parameter timing associated feature vectors;

[0016] A coefficient calculation module, used to extract the photovoltaic energy storage device operating parameter timing association feature vector of the analyzed photovoltaic energy storage device from the set of photovoltaic energy storage device operating parameter timing association feature vectors as a query feature vector, and calculate the operation mode deviation semantic measurement coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter timing association feature vectors in the set of photovoltaic energy storage device operating parameter timing association feature vectors; and

[0017] The fault analysis module is used to determine whether the analyzed photovoltaic energy storage device has a fault based on the operation mode deviation semantic measurement coefficient.

[0018] Compared with the prior art, the fault detection method and system for photovoltaic energy storage equipment provided by the present application first arranges the time series of the operating parameters of each photovoltaic energy storage device to obtain a set of photovoltaic energy storage device operating parameter timing input matrices, then extracts features from each photovoltaic energy storage device operating parameter timing input matrix to obtain a set of photovoltaic energy storage device operating parameter timing associated feature vectors, then calculates the operation mode deviation degree semantic measurement coefficient of the photovoltaic energy storage device operating parameter timing associated feature vector of the photovoltaic energy storage device being analyzed and the set of all other photovoltaic energy storage device operating parameter timing associated feature vectors in the set of photovoltaic energy storage device operating parameter timing associated feature vectors, and finally, based on the operation mode deviation degree semantic measurement coefficient, determines whether the analyzed photovoltaic energy storage device has a fault. In this way, the safety and reliability of the equipment can be improved, and the maintenance cost and risk can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0020] Figure 1 The figure is a flow chart of a fault detection method applied to a photovoltaic energy storage device according to an embodiment of the present application.

[0021] Figure 2 Schematic diagram of the architecture of a fault detection method applied to a photovoltaic energy storage device according to an embodiment of the present application.

[0022] Figure 3 4 is a block diagram of a fault detection system applied to a photovoltaic energy storage device according to an embodiment of the present application.

[0023] Figure 4 This is an application scenario diagram of a fault detection method applied to a photovoltaic energy storage device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0025] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0026] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0027] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0028] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0029] In response to the above technical problems, in the technical solution of this application, a fault detection method applied to photovoltaic energy storage equipment is proposed, which can monitor the different operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, such as voltage, current, temperature and power, and use the correlation between multiple motion parameters of each device in the photovoltaic energy storage device cluster to comprehensively perform fault detection of the photovoltaic energy storage device analyzed in the cluster, so as to improve the fault detection and judgment accuracy of the photovoltaic energy storage device, thereby providing accurate fault diagnosis results. In this way, real-time monitoring and intelligent diagnosis of photovoltaic energy storage equipment can be achieved, so as to take maintenance measures in time, improve the safety and reliability of the equipment, and reduce maintenance costs and risks.

[0030] Figure 1 The figure is a flow chart of a fault detection method applied to a photovoltaic energy storage device according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of a fault detection method applied to a photovoltaic energy storage device according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to the fault detection method applied to photovoltaic energy storage equipment in the embodiment of the present application, the method comprises the following steps: S110, obtaining the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, wherein the operating parameters include voltage value, current value, temperature value and power value; S120, arranging the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster into a photovoltaic energy storage device operating parameter timing input matrix according to the time dimension and the operating parameter sample dimension to obtain a set of photovoltaic energy storage device operating parameter timing input matrices; S130, respectively extracting the set of photovoltaic energy storage device operating parameter timing input matrices through an operating mode feature extractor based on a deep neural network model; The photovoltaic energy storage device operating parameter timing input matrix in the combination is subjected to feature extraction to obtain a set of photovoltaic energy storage device operating parameter timing associated feature vectors; S140, extracting the photovoltaic energy storage device operating parameter timing associated feature vector of the analyzed photovoltaic energy storage device from the set of photovoltaic energy storage device operating parameter timing associated feature vectors as a query feature vector, and calculating the operating mode deviation degree semantic measurement coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter timing associated feature vectors in the set of photovoltaic energy storage device operating parameter timing associated feature vectors; and, S150, determining whether the analyzed photovoltaic energy storage device has a fault based on the operating mode deviation degree semantic measurement coefficient.

[0031] Specifically, in the technical solution of the present application, first, the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster is obtained, wherein the operating parameters include voltage value, current value, temperature value and power value. Next, considering that the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster include voltage value, current value, temperature value and power value, these operating parameter items not only have a dynamic change law of time series in the time dimension, but also have a coordinated correlation relationship of time series between these operating parameter items, which is of great significance for the operation status monitoring of photovoltaic energy storage devices and the overall operation mode analysis of photovoltaic energy storage device clusters. Based on this, in the technical solution of the present application, it is necessary to arrange the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster according to the time dimension and the operating parameter sample dimension as a photovoltaic energy storage device operating parameter timing input matrix to obtain a set of photovoltaic energy storage device operating parameter timing input matrices.

[0032] Then, each photovoltaic energy storage device operating parameter timing input matrix in the set of photovoltaic energy storage device operating parameter timing input matrices is subjected to feature mining in an operating mode feature extractor based on a convolutional neural network model, so as to respectively extract the timing coordinated correlation feature information between multiple operating parameter items of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, so as to reflect the operating mode of each photovoltaic energy storage device in the cluster, thereby obtaining a set of photovoltaic energy storage device operating parameter timing correlation feature vectors.

[0033] Correspondingly, in step S130, the deep neural network model is a convolutional neural network model. In other words, the operating mode feature extractor based on the deep neural network model is an operating mode feature extractor based on the convolutional neural network model.

[0034] Furthermore, in the process of fault detection of the analyzed photovoltaic energy storage device, it is considered that if the operating state of the analyzed photovoltaic energy storage device is normal, then its operating mode is consistent with that of other photovoltaic energy storage devices in the cluster, otherwise the semantic difference of the operating mode is high. Based on this, in order to be able to more accurately perform fault detection of the analyzed photovoltaic energy storage device, it is necessary to compare and analyze the operating mode of the analyzed photovoltaic energy storage device with the overall operating mode of each photovoltaic energy storage device in the cluster, so as to determine the semantic deviation of its operating mode with other devices. Specifically, in the technical solution of the present application, the photovoltaic energy storage device operating parameter timing association feature vector of the analyzed photovoltaic energy storage device is further extracted from the set of the photovoltaic energy storage device operating parameter timing association feature vector as a query feature vector, and the operating mode deviation semantic measurement coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter timing association feature vectors in the set of the photovoltaic energy storage device operating parameter timing association feature vector is calculated. By calculating the operation mode deviation semantic measurement coefficient, the operation mode difference between the analyzed photovoltaic energy storage device and other photovoltaic energy storage devices in the cluster can be measured. If the operation mode deviation semantic measurement coefficient is large, it means that there is a significant difference in the operation mode between the devices, which means that the analyzed device has a fault. On the contrary, if the operation mode deviation semantic measurement coefficient is small, it means that the operation modes between the devices are relatively similar, indicating that the operation status of the analyzed device is normal.

[0035] Accordingly, in step S140, the photovoltaic energy storage device operating parameter timing association feature vector of the analyzed photovoltaic energy storage device is extracted from the set of photovoltaic energy storage device operating parameter timing association feature vectors as the query feature vector, and the operation mode deviation semantic measurement coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter timing association feature vectors in the set of photovoltaic energy storage device operating parameter timing association feature vectors is calculated, including: calculating the operation mode deviation semantic measurement coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter timing association feature vectors in the set of photovoltaic energy storage device operating parameter timing association feature vectors using the following coefficient calculation formula; wherein the coefficient calculation formula is:

[0036]

[0037] Among them, v i is the query feature vector, v j is the time series associated feature vector of each photovoltaic energy storage device operating parameter in the set of the photovoltaic energy storage device operating parameter time series associated feature vectors except the query feature vector, ||·||1 represents the 1 norm of the feature vector, M is the number of photovoltaic energy storage device operating parameter time series associated feature vectors in the set of the photovoltaic energy storage device operating parameter time series associated feature vectors -1, V k is the set of time series associated feature vectors of the operating parameters of the photovoltaic energy storage device, D i Indicates the semantic measurement coefficient of the operating mode deviation.

[0038] Further, in the technical solution of the present application, each photovoltaic energy storage device operating parameter time series associated feature vector in the set of photovoltaic energy storage device operating parameter time series associated feature vectors expresses the local high-order semantic association features of the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster in the spatial domain woven by the time dimension and the cross dimension of the operating parameter samples. Therefore, when the photovoltaic energy storage device operating parameter time series associated feature vector of the analyzed photovoltaic energy storage device is extracted from the set of photovoltaic energy storage device operating parameter time series associated feature vectors as the query feature vector, and the operation mode deviation semantic metric coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter time series associated feature vectors in the set of photovoltaic energy storage device operating parameter time series associated feature vectors is calculated, the vector set composed of all other photovoltaic energy storage device operating parameter time series associated feature vectors will cause the significance of the associated feature distribution information of each photovoltaic energy storage device operating parameter time series associated feature vector due to the difference in the associated feature distribution of the source data subset of each photovoltaic energy storage device operating parameter time series associated feature vector, thereby affecting the calculation accuracy of the operation mode deviation semantic metric coefficient.

[0039] Based on this, in another preferred embodiment of the present application, a photovoltaic energy storage device operating parameter timing association feature vector of the analyzed photovoltaic energy storage device is extracted from the set of photovoltaic energy storage device operating parameter timing association feature vectors as a query feature vector, and an operating mode deviation semantic metric coefficient of the query feature vector and a set of all other photovoltaic energy storage device operating parameter timing association feature vectors in the set of photovoltaic energy storage device operating parameter timing association feature vectors is calculated, including:

[0040] Arranging a set of time-series associated characteristic vectors of the photovoltaic energy storage device operating parameters into a characteristic matrix to obtain an initial key matrix;

[0041] Input the query feature vector and the initial key matrix into a cross-domain attention gate query encoder based on key matrix local correlation enhancement to obtain a self-correlated cross-domain query matching feature vector of the device to be analyzed;

[0042] The self-associative cross-domain query matching feature vector of the device to be analyzed is input into a decoder to obtain the semantic measurement coefficient of the operating mode deviation degree.

[0043] In particular, in this preferred embodiment, the query feature vector and the initial key matrix are input into a cross-domain attention gate query encoder based on the local association enhancement of the key matrix to obtain a self-association cross-domain query matching feature vector of the device to be analyzed, including:

[0044] Calculating feature distribution difference energy coefficients between the query feature vector and each row vector in the initial key matrix to obtain a set of feature distribution difference energy coefficients;

[0045] Comparing each feature distribution difference energy coefficient in the set of feature distribution difference energy coefficients with a predetermined energy threshold, and in response to the feature distribution difference energy coefficient being less than the predetermined energy threshold, taking a mean vector between a row vector corresponding to the feature distribution difference energy coefficient and the query feature vector as an updated row vector to obtain an updated key matrix;

[0046] Processing the query feature vector using a value embedding matrix to obtain a value vector;

[0047] The query feature vector is used as a query vector, each row vector in the update key matrix is ​​used as a key vector, and the value vector is input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of cross-domain query attention vectors;

[0048] The position-based mean vector of the sequence of the cross-domain query attention vector vector is calculated to obtain the self-associated cross-domain query matching feature vector of the device to be analyzed.

[0049] This process can be expressed as:

[0050] K={k1,k2,...,k n}

[0051] M o ={k1; k2; ...; k n}

[0052]

[0053]

[0054]

[0055] v v =v q W q +b q

[0056]

[0057]

[0058] Wherein, K represents the set of time series associated characteristic vectors of the operating parameters of the photovoltaic energy storage equipment, k1, k2, k n represents the first, second and nth photovoltaic energy storage device operating parameter timing association feature vectors in the set of photovoltaic energy storage device operating parameter timing association feature vectors, n represents the total number of vectors in the set of photovoltaic energy storage device operating parameter timing association feature vectors, k i represents the ith photovoltaic energy storage device operating parameter timing association feature vector in the set of photovoltaic energy storage device operating parameter timing association feature vectors, {·} represents a set of vectors, M o represents the initial bond matrix, v q represents the query feature vector, represents positional subtraction, ⊙ represents positional multiplication, represents positional addition, [·,·,·] represents vector concatenation, conv1D represents a one-dimensional convolution operation, MaxPool represents maximum pooling, and f int Represents the characteristic correlation coefficient between calculated vectors, R i are the characteristic correlation coefficients, μ and σ corresponding to the time series correlation characteristic vector of the operating parameters of the i-th photovoltaic energy storage device 2 is represented by the characteristic mean and characteristic variance corresponding to the time series associated characteristic vector of the operating parameters of the i-th photovoltaic energy storage device, λ represents a predetermined hyperparameter, max(·) represents the calculated maximum value, e(R i) represents the i-th feature distribution difference energy coefficient in the set of feature distribution difference energy coefficients, θ represents a predetermined energy threshold, k i ′ represents the updated i-th row vector, M i represents the updated key matrix, W q represents the value embedding matrix, b q Represents the value bias vector, v v represents the value vector corresponding to the query feature vector, T represents the transpose of the vector, d represents the length of each row vector in the update key matrix, softmax represents the normalized exponential function, v pi represents the i-th cross-domain query attention vector in the sequence of the cross-domain query attention vector vector, v p Represents the self-association cross-domain query matching feature vector of the device to be analyzed.

[0059] In particular, the cross-domain attention gate query encoder based on the local association enhancement of the key matrix first calculates the feature distribution difference between the query feature vector and the row vector in the initial key matrix to obtain a set of difference energy coefficients. That is, the query feature vector is used as a soft anchor reference point and the feature distribution energy difference coefficient is used to measure the feature distribution difference between the query feature vector and each key vector in the initial key matrix. It should be understood that, considering that the query feature vector is a common soft anchor reference point, the set of feature distribution energy difference coefficients as a whole also contains local association information between key feature vectors. Next, the feature distribution difference energy coefficient is compared with a predetermined energy threshold. For the case where it is less than the threshold, the mean vector of the query feature vector and the corresponding row vector is used to update the row vector in the key matrix. Here, the update operation strengthens the timing association feature vector of the photovoltaic energy storage device operating parameters that is more similar to the query vector. Through the local association enhancement mechanism, the pertinence and matching quality of the key matrix are improved.

[0060] Furthermore, a query vector and a value vector are constructed based on the query feature vector, and each row vector in the update key matrix is ​​used as a key vector, and is input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of cross-domain query attention vector vectors. That is, a cross-domain attention gate query module based on a Transformer structure is used to combine the query vector, key vector and value vector to perform self-attention calculation of features to achieve cross-domain information integration. The effect of the cross-domain attention mechanism is to allow the model to flow and integrate information between different feature domains, thereby improving the richness of feature representation and the accuracy of matching. Finally, by calculating the mean of the cross-domain query attention vector sequence by position, the final self-associated cross-domain query matching feature vector of the device to be analyzed is obtained. The purpose is to comprehensively consider the information of all cross-domain query attention vectors in the sequence and generate a comprehensive feature representation, which helps to improve the model's overall understanding and response capabilities to complex queries.

[0061] In this way, the calculation accuracy of the operation mode offset semantic measurement coefficient between the photovoltaic energy storage device operation parameter timing association feature vector of the analyzed photovoltaic energy storage device and all other photovoltaic energy storage device operation parameter timing association feature vectors in the set of photovoltaic energy storage device operation parameter timing association feature vectors is improved.

[0062] Then, based on the comparison between the semantic metric coefficient of the operating mode deviation and the predetermined threshold, it is determined whether the analyzed photovoltaic energy storage device has a fault. In particular, in a specific example of the present application, in response to the semantic metric coefficient of the operating mode deviation being greater than the predetermined threshold, it is determined that the analyzed photovoltaic energy storage device has a fault. In this way, the difference between the operating mode of the photovoltaic energy storage device cluster as a whole and the operating mode of the analyzed photovoltaic energy storage device can be used to perform fault detection of the analyzed photovoltaic energy storage device, thereby improving the fault detection and judgment accuracy of the photovoltaic energy storage device, thereby providing accurate fault diagnosis results.

[0063] Accordingly, in step S150, based on the operation mode deviation semantic measurement coefficient, determining whether the analyzed photovoltaic energy storage device has a fault includes: based on a comparison between the operation mode deviation semantic measurement coefficient and a predetermined threshold, determining whether the analyzed photovoltaic energy storage device has a fault.

[0064] Furthermore, in the technical solution of the present application, the fault detection method applied to photovoltaic energy storage equipment also includes a training step: for training an operation mode feature extractor based on a convolutional neural network model.

[0065] In one example, the training step includes: obtaining training data, the training data including the time series of training operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, and the true value of the semantic measurement coefficient of the operating mode deviation, wherein the training operating parameters include training voltage values, training current values, training temperature values, and training power values; arranging the time series of training operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster according to the time dimension and the operating parameter sample dimension as a training photovoltaic energy storage device operating parameter timing input matrix to obtain a set of training photovoltaic energy storage device operating parameter timing input matrices; performing feature extraction on each training photovoltaic energy storage device operating parameter timing input matrix in the set of training photovoltaic energy storage device operating parameter timing input matrices through the operating mode feature extractor based on the convolutional neural network model to obtain training photovoltaic energy storage device operating parameter timing input matrices; A set of time-series associated feature vectors; extracting the training photovoltaic energy storage device operating parameter time-series associated feature vectors of the analyzed photovoltaic energy storage device from the set of training photovoltaic energy storage device operating parameter time-series associated feature vectors as the training query feature vector, calculating the training operation mode offset semantic metric coefficient of the training query feature vector and the set of all other training photovoltaic energy storage device operating parameter time-series associated feature vectors in the set of training photovoltaic energy storage device operating parameter time-series associated feature vectors; calculating the cross entropy function value between the training operation mode offset semantic metric coefficient and the true value of the operation mode offset semantic metric coefficient; and using the cross entropy function value as the loss function value to train the operation mode feature extractor based on the convolutional neural network model, wherein, in each round of iteration of the training, the set of training photovoltaic energy storage device operating parameter time-series associated feature vectors is optimized.

[0066] In the technical solution of the present application, each training photovoltaic energy storage device operating parameter timing association feature vector in the set of training photovoltaic energy storage device operating parameter timing association feature vectors expresses the local high-order semantic association characteristics of the time series of training operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster in the spatial domain woven by the time dimension and the cross dimension of the operating parameter samples. Therefore, when extracting the training photovoltaic energy storage device operating parameter time series associated feature vector of the analyzed photovoltaic energy storage device from the set of training photovoltaic energy storage device operating parameter time series associated feature vectors as the training query feature vector, and calculating the training operation mode deviation semantic metric coefficient of the set of the training query feature vector and all other training photovoltaic energy storage device operating parameter time series associated feature vectors in the set of training photovoltaic energy storage device operating parameter time series associated feature vectors, the vector set composed of all other training photovoltaic energy storage device operating parameter time series associated feature vectors will cause the significance of the associated feature distribution information of each training photovoltaic energy storage device operating parameter time series associated feature vector due to the difference in the associated feature distribution of the source data subsets of each training photovoltaic energy storage device operating parameter time series associated feature vector, making it difficult to stably focus on the significant local distribution of features when the multiple training photovoltaic energy storage device operating parameter time series associated feature vectors are used as a set during the training process, thereby affecting the calculation accuracy of the training operation mode deviation semantic metric coefficient.

[0067] Based on this, the applicant of the present application optimizes the set of trained photovoltaic energy storage device operating parameter timing-related feature vectors during each model iteration training, for example, by back-propagating the difference loss function between the inferred training operating mode offset semantic measurement coefficient and the actual operating mode offset semantic measurement coefficient through the set of trained photovoltaic energy storage device operating parameter timing-related feature vectors.

[0068] Accordingly, in one example, in each round of iteration of the training, the set of the training photovoltaic energy storage device operating parameter timing associated feature vectors is optimized using the following optimization formula to obtain an optimized set of training photovoltaic energy storage device operating parameter timing associated feature vectors; wherein the optimization formula is:

[0069]

[0070] Wherein, V1 is the cascade feature vector obtained by cascading the set of the training photovoltaic energy storage equipment operating parameter time series associated feature vectors, v 1i is the eigenvalue of the ith position of the concatenated eigenvector, and are the squares of the 1-norm and 2-norm of the concatenated feature vector, respectively, L is the length of the concatenated feature vector V1, and ω is a weight hyperparameter, log[·] represents a logarithmic function with base 2, v 1i′ It is the eigenvalue of the i-th position of the optimized cascade feature vector obtained by cascading a set of the optimized training photovoltaic energy storage equipment operating parameter time series associated feature vectors.

[0071] Next, the optimized cascade feature vectors are further expanded to obtain a set of optimized training photovoltaic energy storage equipment operating parameter time series associated feature vectors.

[0072] Specifically, by geometrically registering the high-dimensional feature manifold shape based on the scale and structural parameters of the cascade feature vector V1, it is possible to focus on the features with rich feature semantic information in the feature set composed of the feature values ​​of the cascade feature vector V1, that is, the distinguishable stable interest features based on the dissimilarity of local context information representation during model training, thereby realizing the significant labeling of the feature information of the cascade feature vector V1 in the model training process, and improving the calculation accuracy of the semantic measurement coefficient of the operating mode offset between the photovoltaic energy storage device operating parameter time series associated feature vector of the analyzed photovoltaic energy storage device and the operating parameter time series associated feature vector of all other photovoltaic energy storage devices in the set of photovoltaic energy storage device operating parameter time series associated feature vectors. In this way, the difference between the operating mode of the photovoltaic energy storage device cluster as a whole and the operating mode of the analyzed photovoltaic energy storage device can be used to perform fault detection of the analyzed photovoltaic energy storage device, so as to improve the fault detection and judgment accuracy of the photovoltaic energy storage device. In this way, real-time monitoring and intelligent diagnosis of photovoltaic energy storage devices can be realized, so as to take maintenance measures in time, improve the safety and reliability of the equipment, and reduce maintenance costs and risks.

[0073] In summary, the fault detection method applied to photovoltaic energy storage equipment based on the embodiment of the present application is explained, which can improve the safety and reliability of the equipment and reduce maintenance costs and risks.

[0074] Figure 3 FIG. 1 is a block diagram of a fault detection system 100 applied to a photovoltaic energy storage device according to an embodiment of the present application. Figure 3As shown, according to the embodiment of the present application, the fault detection system 100 applied to the photovoltaic energy storage device includes: a data acquisition module 110, which is used to obtain the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, wherein the operating parameters include voltage value, current value, temperature value and power value; a matrix module 120, which is used to arrange the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster according to the time dimension and the operating parameter sample dimension into a photovoltaic energy storage device operating parameter timing input matrix to obtain a set of photovoltaic energy storage device operating parameter timing input matrices; a feature extraction module 130, which is used to extract the photovoltaic energy storage device operating parameter timing input matrix respectively through an operating mode feature extractor based on a deep neural network model. The operating parameter timing input matrix of each photovoltaic energy storage device in the set of matrices is subjected to feature extraction to obtain a set of photovoltaic energy storage device operating parameter timing associated feature vectors; a coefficient calculation module 140 is used to extract the photovoltaic energy storage device operating parameter timing associated feature vector of the analyzed photovoltaic energy storage device from the set of photovoltaic energy storage device operating parameter timing associated feature vectors as a query feature vector, and calculate the operating mode deviation degree semantic measurement coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter timing associated feature vectors in the set of photovoltaic energy storage device operating parameter timing associated feature vectors; and a fault analysis module 150 is used to determine whether the analyzed photovoltaic energy storage device has a fault based on the operating mode deviation degree semantic measurement coefficient.

[0075] In one example, in the above-mentioned fault detection system 100 applied to photovoltaic energy storage equipment, the deep neural network model is a convolutional neural network model.

[0076] In one example, in the above-mentioned fault detection system 100 applied to photovoltaic energy storage equipment, the coefficient calculation module 140 is used to: arrange the set of time-series correlation feature vectors of the operating parameters of the photovoltaic energy storage equipment into a feature matrix to obtain an initial key matrix; input the query feature vector and the initial key matrix into a cross-domain attention gate query encoder based on the local correlation enhancement of the key matrix to obtain a self-correlated cross-domain query matching feature vector of the device to be analyzed; input the self-correlated cross-domain query matching feature vector of the device to be analyzed into a decoder to obtain the operating mode offset semantic measurement coefficient.

[0077] In one example, in the fault detection system 100 applied to photovoltaic energy storage equipment, the fault analysis module 150 is used to determine whether the analyzed photovoltaic energy storage equipment has a fault based on a comparison between the operation mode deviation semantic metric coefficient and a predetermined threshold.

[0078] Here, those skilled in the art will appreciate that the specific functions and operations of each module in the fault detection system 100 for photovoltaic energy storage equipment have been described in detail above. Figure 1 to Figure 2 It has been described in detail in the description of the fault detection method applied to photovoltaic energy storage equipment, and therefore, its repeated description will be omitted.

[0079] As described above, the fault detection system 100 applied to photovoltaic energy storage equipment according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a fault detection algorithm applied to photovoltaic energy storage equipment. In one example, the fault detection system 100 applied to photovoltaic energy storage equipment according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the fault detection system 100 applied to photovoltaic energy storage equipment can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the fault detection system 100 applied to photovoltaic energy storage equipment can also be one of the many hardware modules of the wireless terminal.

[0080] Alternatively, in another example, the fault detection system 100 applied to the photovoltaic energy storage device and the wireless terminal may also be separate devices, and the fault detection system 100 applied to the photovoltaic energy storage device may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0081] Figure 4 FIG. 1 is an application scenario diagram of a fault detection method for photovoltaic energy storage equipment according to an embodiment of the present application. Figure 4 As shown, in this application scenario, first, the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster is obtained (for example, Figure 4 D), wherein the operating parameters include voltage value, current value, temperature value and power value, and then the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster is input to a server (for example, Figure 4 In S) as shown in , the server is able to use the fault detection algorithm applied to photovoltaic energy storage devices to process the time series of operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster to obtain an operating mode deviation semantic measurement coefficient, and then, based on the comparison between the operating mode deviation semantic measurement coefficient and a predetermined threshold, determine whether the analyzed photovoltaic energy storage device has a fault.

[0082] According to another aspect of the present application, a non-volatile computer-readable storage medium is provided, on which computer-readable instructions are stored. When the instructions are executed by a computer, the above-mentioned method can be executed.

[0083] The program part of the technology can be considered as a "product" or "manufactured product" in the form of executable code and / or related data, which is participated in or realized by computer-readable media. Tangible and permanent storage media can include any memory or storage used by any computer, processor, or similar device or related module. For example, various semiconductor memories, tape drives, disk drives or any similar devices that can provide storage functions for software.

[0084] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0085] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0086] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application are described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A fault detection method for photovoltaic energy storage equipment, characterized in that: include: Acquire a time series of operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, wherein the operating parameters include voltage value, current value, temperature value and power value; Arranging the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster into photovoltaic energy storage device operating parameter timing input matrices according to the time dimension and the operating parameter sample dimension to obtain a set of photovoltaic energy storage device operating parameter timing input matrices; Performing feature extraction on each photovoltaic energy storage device operating parameter timing input matrix in the set of photovoltaic energy storage device operating parameter timing input matrices by using an operating mode feature extractor based on a deep neural network model to obtain a set of photovoltaic energy storage device operating parameter timing associated feature vectors; Extracting a photovoltaic energy storage device operating parameter timing-related feature vector of the analyzed photovoltaic energy storage device from the set of photovoltaic energy storage device operating parameter timing-related feature vectors as a query feature vector, and calculating an operating mode deviation semantic metric coefficient between the query feature vector and a set of all other photovoltaic energy storage device operating parameter timing-related feature vectors in the set of photovoltaic energy storage device operating parameter timing-related feature vectors; and Based on the operation mode deviation semantic metric coefficient, it is determined whether the analyzed photovoltaic energy storage device has a fault.

2. The fault detection method for photovoltaic energy storage equipment according to claim 1, characterized in that: The deep neural network model is a convolutional neural network model.

3. The fault detection method for photovoltaic energy storage equipment according to claim 2, characterized in that: Extracting a photovoltaic energy storage device operating parameter time series associated feature vector of the analyzed photovoltaic energy storage device from the set of photovoltaic energy storage device operating parameter time series associated feature vectors as a query feature vector, and calculating an operating mode deviation semantic metric coefficient between the query feature vector and a set of all other photovoltaic energy storage device operating parameter time series associated feature vectors in the set of photovoltaic energy storage device operating parameter time series associated feature vectors, including: Arranging a set of time-series associated characteristic vectors of the photovoltaic energy storage device operating parameters into a characteristic matrix to obtain an initial key matrix; Input the query feature vector and the initial key matrix into a cross-domain attention gate query encoder based on key matrix local correlation enhancement to obtain a self-correlated cross-domain query matching feature vector of the device to be analyzed; The self-associative cross-domain query matching feature vector of the device to be analyzed is input into a decoder to obtain the semantic measurement coefficient of the operating mode deviation degree.

4. The fault detection method for photovoltaic energy storage equipment according to claim 3 is characterized in that: Inputting the query feature vector and the initial key matrix into a cross-domain attention gate query encoder based on key matrix local correlation enhancement to obtain a self-correlated cross-domain query matching feature vector of the device to be analyzed, including: Calculating feature distribution difference energy coefficients between the query feature vector and each row vector in the initial key matrix to obtain a set of feature distribution difference energy coefficients; Comparing each feature distribution difference energy coefficient in the set of feature distribution difference energy coefficients with a predetermined energy threshold, and in response to the feature distribution difference energy coefficient being less than the predetermined energy threshold, taking a mean vector between a row vector corresponding to the feature distribution difference energy coefficient and the query feature vector as an updated row vector to obtain an updated key matrix; Processing the query feature vector using a value embedding matrix to obtain a value vector; The query feature vector is used as a query vector, each row vector in the update key matrix is ​​used as a key vector, and the value vector is input into a cross-domain attention gate query module based on a Transformer structure to obtain a sequence of cross-domain query attention vectors; The position-based mean vector of the sequence of the cross-domain query attention vector vector is calculated to obtain the self-associated cross-domain query matching feature vector of the device to be analyzed.

5. The fault detection method for photovoltaic energy storage equipment according to claim 4, characterized in that: Determining whether the analyzed photovoltaic energy storage device has a fault based on the operation mode deviation semantic measurement coefficient includes: Based on the comparison between the operation mode deviation semantic metric coefficient and a predetermined threshold, it is determined whether the analyzed photovoltaic energy storage device has a fault.

6. The fault detection method for photovoltaic energy storage equipment according to claim 5, characterized in that: The method also includes a training step for training an operation mode feature extractor based on a convolutional neural network model.

7. The fault detection method for photovoltaic energy storage equipment according to claim 6, characterized in that: The training step comprises: Acquire training data, wherein the training data includes a time series of training operation parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, and a true value of a semantic measurement coefficient of an operation mode deviation, wherein the training operation parameters include a training voltage value, a training current value, a training temperature value, and a training power value; Arranging the time series of the training operation parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster into a training photovoltaic energy storage device operation parameter timing input matrix according to the time dimension and the operation parameter sample dimension to obtain a set of training photovoltaic energy storage device operation parameter timing input matrices; The operation mode feature extractor based on the convolutional neural network model performs feature extraction on each training photovoltaic energy storage device operation parameter timing input matrix in the set of training photovoltaic energy storage device operation parameter timing input matrices to obtain a set of training photovoltaic energy storage device operation parameter timing associated feature vectors; Extracting a training photovoltaic energy storage device operating parameter timing association feature vector of the analyzed photovoltaic energy storage device from the set of training photovoltaic energy storage device operating parameter timing association feature vectors as a training query feature vector, and calculating a training operating mode deviation semantic metric coefficient between the training query feature vector and a set of all other training photovoltaic energy storage device operating parameter timing association feature vectors in the set of training photovoltaic energy storage device operating parameter timing association feature vectors; Calculating a cross entropy function value between the training operation mode deviation semantic metric coefficient and a true value of the operation mode deviation semantic metric coefficient; and The operation mode feature extractor based on the convolutional neural network model is trained using the cross entropy function value as the loss function value, wherein in each round of iteration of the training, the set of timing-related feature vectors of the operation parameters of the training photovoltaic energy storage device is optimized.

8. A fault detection system for photovoltaic energy storage equipment, characterized in that: include: A data acquisition module, used to acquire a time series of operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster, wherein the operating parameters include voltage value, current value, temperature value and power value; A matrixing module, for arranging the time series of the operating parameters of each photovoltaic energy storage device in the photovoltaic energy storage device cluster into a photovoltaic energy storage device operating parameter timing input matrix according to the time dimension and the operating parameter sample dimension to obtain a set of photovoltaic energy storage device operating parameter timing input matrices; A feature extraction module, used to extract features from each photovoltaic energy storage device operating parameter timing input matrix in the set of photovoltaic energy storage device operating parameter timing input matrices through an operating mode feature extractor based on a deep neural network model to obtain a set of photovoltaic energy storage device operating parameter timing associated feature vectors; A coefficient calculation module, used to extract the photovoltaic energy storage device operating parameter timing association feature vector of the analyzed photovoltaic energy storage device from the set of photovoltaic energy storage device operating parameter timing association feature vectors as a query feature vector, and calculate the operation mode deviation semantic measurement coefficient of the query feature vector and the set of all other photovoltaic energy storage device operating parameter timing association feature vectors in the set of photovoltaic energy storage device operating parameter timing association feature vectors; and The fault analysis module is used to determine whether the analyzed photovoltaic energy storage device has a fault based on the operation mode deviation semantic measurement coefficient.

9. The fault detection system for photovoltaic energy storage equipment according to claim 10, characterized in that: The deep neural network model is a convolutional neural network model.

10. The fault detection system for photovoltaic energy storage equipment according to claim 9, characterized in that: The coefficient calculation module is used for: Arranging a set of time-series associated characteristic vectors of the photovoltaic energy storage device operating parameters into a characteristic matrix to obtain an initial key matrix; Input the query feature vector and the initial key matrix into a cross-domain attention gate query encoder based on key matrix local correlation enhancement to obtain a self-correlated cross-domain query matching feature vector of the device to be analyzed; The self-associative cross-domain query matching feature vector of the device to be analyzed is input into a decoder to obtain the semantic measurement coefficient of the operating mode deviation degree.

Images