Transient electric energy quality problem identification method and related equipment

Abstract

The invention relates to the technical field of electric energy quality, and provides a transient electric energy quality problem identification method and related equipment. The identification method comprises the following steps: based on all electric energy quality signals, acquiring low-level features and high-level features by using a lightweight network model; fusing the low-level features and the high-level features by using a feature fusion model to obtain final features; based on all the final features and all the high-level features, a prediction model is utilized to obtain an identification result; constructing a loss function according to all identification results, and optimizing the lightweight network model, the feature fusion model and the prediction model by using the loss function; and acquiring a current power quality signal of the to-be-identified power distribution network, and identifying the current power quality signal by using the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to obtain a final identification result of the to-be-identified power distribution network. According to the method, the comprehensiveness of transient power quality problem identification can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power quality, and in particular to a method for identifying transient power quality problems and related equipment. Background Art

[0002] With the continuous construction of new power systems, more and more distributed renewable energy sources are connected to the power system on a large scale. This process has also led to increasingly complex and random power quality problems in the power system. Frequent power quality problems will not only affect the user experience of power users, but also have an adverse effect on the stable operation and service life of electronic power equipment. In severe cases, it may even cause huge economic losses. Therefore, it is necessary to accurately identify the power quality problems occurring in the power system and provide information support to the upper control center in a timely manner to assess whether it is necessary to warn and control the power quality problems that occur, so as to avoid further major impacts and ensure the reliability and stability of the power supply of the overall power system.

[0003] Transient power quality problems refer to power quality problems that occur in a very short period of time. They are essentially transient manifestations of voltage quality and are often closely related to events such as internal system faults and lightning strikes.

[0004] The commonly used power quality problem identification system currently deploys edge devices in distribution stations and user areas to collect raw power quality data. However, due to the limited computing power and storage resources of edge devices, edge devices will transmit the collected power quality data to the control center or the cloud through 4G, Ethernet and other communication methods for further monitoring and analysis of power quality problems. With the large-scale involvement of distributed new energy, the number of monitored nodes will increase, and the raw data received by the control center or the cloud will also double. Faced with the massive growth of data, the control center and the cloud will be under tremendous pressure in terms of computing performance and storage performance. At the same time, this centralized processing method is inevitably affected by communication bandwidth and latency, resulting in the existing power quality problem identification system being difficult to meet the urgent needs of new power systems for real-time monitoring of power quality problems.

[0005] In response to the above problems, edge devices that are closer to the data source take on more computing tasks, and completing the identification and monitoring of power quality problems on-site after collecting the original data has become a new idea for the power quality problem identification system. In recent years, deep learning methods have gradually become a research hotspot in new power quality problem identification systems due to their powerful learning and mapping capabilities. Deploying high-performance deep learning models to edge devices for the identification and monitoring of power quality problems provides a new idea for meeting the dual needs of new power systems for accuracy and real-time identification of power quality problems, and has huge development potential.

[0006] The existing power quality problem identification method based on deep learning has a powerful and flexible feature extraction capability, which makes up for the shortcomings of redundant and slow manual feature extraction in traditional methods, and can effectively improve the recognition accuracy of various power quality problems. However, most of the deep learning methods currently used in the field of power quality problem identification only use deep learning algorithms to identify and classify data, and are used to distinguish which type of power quality problem the current data belongs to, and their functions are relatively simple. It can be seen that the current power quality problem identification method has the problem of poor comprehensiveness in identifying transient power quality problems. Summary of the invention

[0007] The present application provides a method for identifying transient power quality problems and related equipment, which can solve the problem of poor comprehensiveness in identifying transient power quality problems.

[0008] In a first aspect, an embodiment of the present application provides a method for identifying a transient power quality problem, the method comprising:

[0009] Acquire multiple power quality signal sequences of the target distribution network; each power quality signal sequence includes power quality signals at multiple historical moments, and the power quality signals are used to describe the power quality of the target distribution network;

[0010] Based on all power quality signals in each power quality signal sequence, a lightweight network model is used to obtain the underlying features and high-level features of each power quality signal sequence; the underlying features are used to describe the location information of all power quality signals in the power quality signal sequence, and the high-level features are used to obtain the semantic information of all power quality signals in the power quality signal sequence;

[0011] For each power quality signal sequence, the feature fusion model is used to fuse the bottom-level features and high-level features of the power quality signal sequence to obtain the final features;

[0012] Based on all final features and all high-level features, a prediction model is used to obtain the recognition results of each power quality signal sequence; the recognition results include a plurality of predicted transient power quality problems corresponding to the prediction center point, prediction length, confidence and prediction category, the prediction length is the number of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the prediction center point is the central power quality signal of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the confidence is used to describe the probability of the predicted transient power quality problem actually occurring, and the prediction category is the category of the predicted transient power quality problem;

[0013] A loss function is constructed based on all recognition results, and the lightweight network model, feature fusion model and prediction model are optimized using the loss function to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model; the loss function is used to describe the accuracy of all recognition results;

[0014] Acquire multiple current power quality signals of the distribution network to be identified, use the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to identify all current power quality signals, and obtain the final identification result of the distribution network to be identified; the final identification result includes the prediction center point, prediction length, confidence and prediction category corresponding to multiple final predicted transient power quality problems.

[0015] Optionally, the lightweight network model includes a first extraction sub-model, a second extraction sub-model, a third extraction sub-model, a fourth extraction sub-model, a fifth extraction sub-model, a sixth extraction sub-model, a first one-dimensional convolution sub-model, and a second one-dimensional convolution sub-model connected in sequence;

[0016] The first extraction sub-model, the second extraction sub-model, the third extraction sub-model, the fourth extraction sub-model, the fifth extraction sub-model, and the sixth extraction sub-model each include an extraction sub-model;

[0017] The extraction sub-model includes a one-dimensional convolution unit and a pooling unit connected in sequence, the input end of the one-dimensional convolution unit is the input end of the extraction sub-model, and the output end of the pooling unit is the output end of the extraction sub-model;

[0018] The input end of the first extraction sub-model is the input end of the lightweight network model, the output end of the one-dimensional convolution unit in the fifth extraction sub-model is the first output end of the lightweight network model, and the output end of the second one-dimensional convolution sub-model is the second output end of the lightweight network model.

[0019] Optionally, the first one-dimensional convolution sub-model and the second one-dimensional convolution sub-model each include a one-dimensional convolution unit.

[0020] Optionally, the one-dimensional convolution unit includes a one-dimensional convolution layer, a batch normalization layer, an activation layer, and an attention mechanism layer connected in sequence;

[0021] The input end of the one-dimensional convolution layer is the input end of the one-dimensional convolution unit, and the output end of the attention mechanism layer is the output end of the one-dimensional convolution unit.

[0022] Optionally, the feature fusion model includes a one-dimensional convolution sub-model, an upsampling sub-model, and a feature splicing sub-model connected in sequence;

[0023] The input end of the one-dimensional convolution sub-model and the input end of the feature splicing sub-model are both input ends of the feature fusion model. The input end of the feature splicing sub-model is connected to the first output end of the lightweight network model, the input end of the one-dimensional convolution sub-model is connected to the second output end of the lightweight network model, and the output end of the feature splicing sub-model is the output end of the feature fusion model.

[0024] Optionally, the prediction model includes a first prediction sub-model, a second prediction sub-model and a splicing sub-model;

[0025] The input end of the splicing sub-model is connected to the output end of the first prediction sub-model and the output end of the second prediction sub-model respectively.

[0026] Optionally, the first prediction sub-model and the second prediction sub-model each include a prediction sub-model;

[0027] The prediction sub-model includes a one-dimensional convolution unit, a semantic convolution unit, a semantic prediction unit, a detection convolution unit, a detection prediction unit, a confidence prediction unit, and a splicing unit;

[0028] The input end of the one-dimensional convolution unit is the input end of the prediction sub-model, and the output end of the concatenation unit is the output end of the prediction sub-model;

[0029] The output end of the one-dimensional convolution unit is respectively connected to the input end of the semantic convolution unit and the input end of the detection convolution unit, the output end of the semantic convolution unit is connected to the input end of the semantic prediction unit, the output end of the detection convolution unit is respectively connected to the input end of the detection prediction unit and the input end of the confidence prediction unit, and the input end of the splicing unit is respectively connected to the output end of the semantic prediction unit, the output end of the detection prediction unit and the output end of the confidence prediction unit.

[0030] Optionally, the loss function is:

[0031] loss=loss_iou+loss_obj+loss_cls

[0032] Among them, loss represents the value of the loss function, loss_iou represents the position loss function, loss_obj represents the confidence loss function, and loss_cls represents the classification loss function:

[0033] loss_iou=1-IOU 2

[0034]

[0035] Among them, IOU represents the intersection-over-union ratio between all predicted center points and the true center points and the intersection-over-union ratio between all predicted lengths and the true lengths, y represents the set of true confidences of all predicted transient power quality problems, represents the set of confidences of all predicted transient power quality problems, c represents the set of true categories of transient power quality problems of all power quality signal sequences, Represents the set of prediction categories for all predicted transient power quality issues.

[0036] In a second aspect, an embodiment of the present application provides a device for identifying transient power quality problems, including:

[0037] An acquisition module is used to acquire multiple power quality signal sequences of a target distribution network; each power quality signal sequence includes power quality signals at multiple historical moments, and the power quality signals are used to describe the power quality of the target distribution network;

[0038] The feature acquisition module uses a lightweight network model to acquire the underlying features and high-level features of each power quality signal sequence based on all power quality signals in each power quality signal sequence; the underlying features are used to describe the location information of all power quality signals in the power quality signal sequence, and the high-level features are used to obtain the semantic information of all power quality signals in the power quality signal sequence;

[0039] The fusion module uses a feature fusion model to fuse the bottom-level features and high-level features of each power quality signal sequence to obtain the final features.

[0040] The recognition result acquisition module obtains the recognition result of each power quality signal sequence by using the prediction model based on all the final features and all the high-level features; the recognition result includes the prediction center point, prediction length, confidence and prediction category corresponding to multiple predicted transient power quality problems, the prediction length is the number of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the prediction center point is the central power quality signal of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the confidence is used to describe the probability of the predicted transient power quality problem actually occurring, and the prediction category is the category of the predicted transient power quality problem;

[0041] The optimization module constructs a loss function based on all recognition results, and uses the loss function to optimize the lightweight network model, feature fusion model and prediction model to obtain the optimized lightweight network model, optimized feature fusion model and optimized prediction model; the loss function is used to describe the accuracy of all recognition results;

[0042] The identification module obtains multiple current power quality signals of the distribution network to be identified, and uses the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to identify all current power quality signals to obtain the final identification result of the distribution network to be identified; the final identification result includes the prediction center point, prediction length, confidence and prediction category corresponding to multiple final predicted transient power quality problems.

[0043] In a third aspect, an embodiment of the present application provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for identifying transient power quality problems when executing the above-mentioned computer program.

[0044] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the method for identifying transient power quality problems described above is implemented.

[0045] The above solution of the present application has the following beneficial effects:

[0046] In some embodiments of the present application, multiple power quality signal sequences of the target distribution network are obtained, and then based on all power quality signals in each power quality signal sequence, the underlying features and high-level features of each power quality signal sequence are obtained using a lightweight network model, and then for each power quality signal sequence, the underlying features and high-level features of the power quality signal sequence are fused using a feature fusion model to obtain the final features, and then based on all the final features and all the high-level features, the recognition result of each power quality signal sequence is obtained using a prediction model, and then a loss function is constructed according to all the recognition results, and the loss function is used to optimize the lightweight network model, the feature fusion model and the prediction model to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model, and finally, multiple current power quality signals of the distribution network to be identified are obtained, and all current power quality signals are identified using the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to obtain the final recognition result of the distribution network to be identified. Among them, by obtaining the underlying features and the high-level features, the information of the power quality signal sequence at two levels can be expressed, the comprehensiveness of the information is increased, and the underlying features and the high-level features are merged to obtain the final features, so that the final features contain the information of the underlying features and the high-level features, and the information richness of the final features is improved. The recognition results are obtained based on the information-rich final features and high-level features, which can improve the accuracy of the recognition results. At the same time, the recognition results include four aspects of data: prediction center point, prediction length, confidence and prediction category, which improves the comprehensiveness of power quality problem identification.

[0047] Other beneficial effects of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 A flow chart of a method for identifying transient power quality problems provided in an embodiment of the present application;

[0050] Figure 2 A schematic diagram of the structure of a lightweight network model provided in one embodiment of the present application;

[0051] Figure 3 A schematic diagram of the structure of a one-dimensional convolution unit provided in one embodiment of the present application;

[0052] Figure 4 A schematic diagram of the structure of a feature fusion model provided in one embodiment of the present application;

[0053] Figure 5 A schematic diagram of the structure of a prediction model provided in one embodiment of the present application;

[0054] Figure 6 A loss function curve diagram provided for an embodiment of the present application;

[0055] Figure 7 A schematic diagram of the structure of a device for identifying transient power quality problems provided in an embodiment of the present application;

[0056] Figure 8 A schematic diagram of the structure of a terminal device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0057] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0058] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0059] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0060] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0061] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0062] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0063] In view of the problem of poor comprehensiveness of existing transient power quality problem identification, an embodiment of the present application provides a method for identifying transient power quality problems. The identification method obtains multiple power quality signal sequences of the target distribution network, and then uses a lightweight network model to obtain the underlying features and high-level features of each power quality signal sequence based on all power quality signals in each power quality signal sequence. Then, for each power quality signal sequence, the underlying features and high-level features of the power quality signal sequence are fused using a feature fusion model to obtain the final features. Then, based on all the final features and all the high-level features, a prediction model is used to obtain the identification result of each power quality signal sequence. Then, a loss function is constructed according to all the identification results, and the loss function is used to optimize the lightweight network model, the feature fusion model and the prediction model to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model. Finally, multiple current power quality signals of the distribution network to be identified are obtained, and all current power quality signals are identified using the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to obtain the final identification result of the distribution network to be identified. Among them, by obtaining the underlying features and the high-level features, the information of the power quality signal sequence at two levels can be expressed, the comprehensiveness of the information is increased, and the underlying features and the high-level features are merged to obtain the final features, so that the final features contain the information of the underlying features and the high-level features, and the information richness of the final features is improved. The recognition results are obtained based on the information-rich final features and high-level features, which can improve the accuracy of the recognition results. At the same time, the recognition results include four aspects of data: prediction center point, prediction length, confidence and prediction category, which improves the comprehensiveness of power quality problem identification.

[0064] Step 11: Acquire multiple power quality signal sequences of the target distribution network.

[0065] Each of the above power quality signal sequences includes power quality signals at multiple historical moments. The above power quality signals are used to describe the power quality of the target distribution network, such as voltage amplitude, voltage frequency or voltage waveform quality, etc. The information contained in all power quality signals is of the same type, such as signals describing voltage amplitude.

[0066] In some embodiments of the present application, all power quality signals of the target distribution network within a period of historical time can be obtained, and then all power quality signals are divided into multiple continuous power quality signal sequences. Then, for each power quality signal sequence, a starting power quality signal and an ending power quality signal of a transient power quality problem are randomly generated from all power quality signals in the power quality signal sequence, the starting power quality signal is the power quality signal corresponding to the moment when the transient power quality problem occurs, and the ending power quality signal is the power quality signal corresponding to the moment when the transient power quality problem ends, and the real category of the transient power quality problem is set (such as voltage sag, voltage swell, short interruption, etc.), and then the real center point and real length of the transient power quality problem are calculated based on the starting power quality signal and the ending power quality signal, which can be calculated by the formula:

[0067]

[0068] Calculate the true center point number x c and true length w.

[0069] Wherein, t1 represents the serial number of the starting power quality signal, t2 represents the serial number of the ending power quality signal, and K represents the number of power quality signals in the data set. The power quality signal corresponding to the serial number of the real center point is taken as the real center point.

[0070] Thus, the transient power quality problem described by the real center point, real length and real category of each power quality signal sequence is obtained.

[0071] Step 12: Based on all power quality signals in each power quality signal sequence, a lightweight network model is used to obtain the bottom-level features and high-level features of each power quality signal sequence.

[0072] The above-mentioned low-level features are used to describe the position information of all power quality signals in the power quality signal sequence (such as the relative position of each power quality signal in the power quality signal sequence), and the high-level features are used for the semantic information of all power quality signals in the power quality signal sequence (such as the information contained in each power quality signal).

[0073] like Figure 2 As shown, the above-mentioned lightweight network model includes a first extraction sub-model, a second extraction sub-model, a third extraction sub-model, a fourth extraction sub-model, a fifth extraction sub-model, a sixth extraction sub-model, a first one-dimensional convolution sub-model and a second one-dimensional convolution sub-model which are connected in sequence.

[0074] The first extraction sub-model, the second extraction sub-model, the third extraction sub-model, the fourth extraction sub-model, the fifth extraction sub-model, and the sixth extraction sub-model each include an extraction sub-model.

[0075] The extraction sub-model includes a one-dimensional convolution unit and a pooling unit connected in sequence, the input end of the one-dimensional convolution unit is the input end of the extraction sub-model, and the output end of the pooling unit is the output end of the extraction sub-model.

[0076] The input end of the first extraction sub-model is the input end of the lightweight network model, the output end of the one-dimensional convolution unit in the fifth extraction sub-model is the first output end of the lightweight network model, and the output end of the second one-dimensional convolution sub-model is the second output end of the lightweight network model.

[0077] It should be noted that the first output terminal of the lightweight network model outputs the bottom-level features, and the second output terminal outputs the high-level features. The first one-dimensional convolution sub-model and the second one-dimensional convolution sub-model each include a one-dimensional convolution unit. Figure 3 As shown, the one-dimensional convolution unit includes a one-dimensional convolution layer, a batch normalization layer, an activation layer, and an attention mechanism layer connected in sequence. The input end of the one-dimensional convolution layer is the input end of the one-dimensional convolution unit, and the output end of the attention mechanism layer is the output end of the one-dimensional convolution unit. The one-dimensional convolution layer is used to extract features, the batch normalization layer is used for normalization, the activation function layer (i.e., the activation layer) is used to introduce nonlinear features, and the attention mechanism layer is used for further feature extraction. The pooling unit is a pooling operation, which is used to reduce dimensionality and prevent overfitting.

[0078] In some embodiments of the present application, for all power quality signals corresponding to each power quality signal sequence, all power quality signals are evenly divided into multiple sets, the number of sets is the same as the number of input channels of the first extraction sub-model, the sets correspond to the input channels one by one, and then the sets are input into the corresponding input channels. The operation of the one-dimensional convolution layer in the one-dimensional convolution unit of the lightweight network can be expressed as:

[0079]

[0080] Among them, y n [i] represents the output of the nth output channel of the one-dimensional convolutional layer, n represents the number of output channels, m represents the mth input channel, k represents the length of the convolution kernel, s represents the convolution stride, and x represents the length of the convolution kernel. m represents the input of the mth input channel, w n,m represents the convolution kernel weight between the nth output channel and the mth input channel, i represents the i-th sequence number of the nth output channel, i∈l', l' represents the sequence length of the output channel, and the output of the final one-dimensional convolution layer is y=[n,l'], which is then input into the batch normalization layer for calculation. The expression is:

[0081]

[0082] Among them, y' represents the output of the batch normalization layer, γn represents the scaling parameter, μ n represents the mean of the nth channel, represents the variance of the nth channel, ε represents the adjustment parameter, β n Represents the offset parameter, which is then input into the activation function layer for calculation. The expression is:

[0083] y'=max(0,y')

[0084] Among them, y" represents the output of the activation function layer. This formula simply sets all negative values ​​in the input to 0, and the positive values ​​remain unchanged, thereby introducing nonlinear characteristics, and finally inputs it into the attention mechanism layer for calculation. The expression is:

[0085]

[0086] f n =σ(s n )

[0087]

[0088] in, represents the output of the attention mechanism layer, f n represents the channel attention coefficient of n channels, s n represents the channel attention weight of n channels, s j represents the channel attention weight of the jth channel, s j ∈s n , w i represents the convolution kernel weight, z n Represents the global feature vector of the nth channel, y' n,i Represents the output of the ith sequence number of the nth channel in y". The sequence length obtained by the above formula is n-k+1, which is filled with 0s evenly on the left and right to make its sequence length n.

[0089] It should be noted that in the above expression, the output shape of each layer is the same, that is, [n, l'], with a total of n*l' data. The output of each one-dimensional convolution unit is padded, and 0 is evenly padded on the left and right of the output sequence to make the output channel sequence length the same as the input, and the sequence dimension reduction is only performed through the subsequent pooling unit.

[0090] In some embodiments of the present application, in the lower layers or convolution layers with fewer channels, the feature expression is insufficient, and the effect of the attention mechanism on such features is not obvious. Therefore, in some embodiments, the attention mechanism layer can be added only to the one-dimensional convolution unit with the number of output channels greater than 64.

[0091] It is worth mentioning that by obtaining the underlying features and high-level features, the information of the power quality signal sequence at two levels can be expressed, increasing the comprehensiveness of the information. A lightweight network model mainly composed of one-dimensional convolution units and pooling units is constructed to achieve accurate feature extraction of the input data.

[0092] At the same time, the lightweight network model constructed in this application has a simple structure and effectively reduces computing and storage resources.

[0093] Step 13: for each power quality signal sequence, the feature fusion model is used to fuse the bottom-level features and high-level features of the power quality signal sequence to obtain the final features.

[0094] like Figure 4 As shown, the feature fusion model includes a one-dimensional convolution sub-model, an upsampling sub-model and a feature splicing sub-model connected in sequence. Figure 4 Input 1 corresponds to the low-level features, and input 2 corresponds to the high-level features.

[0095] The input end of the one-dimensional convolution sub-model and the input end of the feature splicing sub-model are both input ends of the feature fusion model. The input end of the feature splicing sub-model is connected to the first output end of the lightweight network model, the input end of the one-dimensional convolution sub-model is connected to the second output end of the lightweight network model, and the output end of the feature splicing sub-model is the output end of the feature fusion model.

[0096] In some embodiments of the present application, the input end of the one-dimensional convolution sub-model inputs high-level features, the input end of the feature splicing sub-model inputs low-level features, and the one-dimensional convolution sub-model includes a one-dimensional convolution unit.

[0097] It should be noted that the upsampling module is a nearest neighbor interpolation calculation, which is used to interpolate the channel sequence in the output of the one-dimensional convolution sub-model so that the sequence length of its channel sequence is the same as the sequence length of the underlying features, and then spliced ​​with the underlying features in the feature splicing sub-model to obtain the final features.

[0098] It is worth mentioning that the low-level features and high-level features are fused to obtain the final features, so that the final features contain information of both low-level features and high-level features, thereby improving the information richness of the final features.

[0099] Step 14: Based on all the final features and all the high-level features, a prediction model is used to obtain the recognition result of each power quality signal sequence.

[0100] The above-mentioned identification results include the predicted central point, predicted length, confidence and predicted category corresponding to multiple predicted transient power quality problems. The predicted length is the number of all power quality signals corresponding to the duration period of the predicted transient power quality problem. The predicted central point is the central power quality signal of all power quality signals corresponding to the duration period of the predicted transient power quality problem. The confidence is used to describe the probability of the predicted transient power quality problem actually occurring. The predicted category is the category of the predicted transient power quality problem (such as voltage swell, voltage sag or short interruption, etc.).

[0101] like Figure 5 As shown, the above prediction model includes a first prediction sub-model, a second prediction sub-model and a splicing sub-model. Figure 5 Input 1 is the input of the first prediction sub-model, and input 2 is the input of the second prediction sub-model.

[0102] The input end of the splicing sub-model is connected to the output end of the first prediction sub-model and the output end of the second prediction sub-model respectively.

[0103] The first prediction sub-model and the second prediction sub-model each include a prediction sub-model.

[0104] The prediction sub-model includes a one-dimensional convolutional unit (such as Figure 5 ), semantic convolutional units (such as Figure 5 Cls convolution in ), semantic prediction unit (such as Figure 5 Cls prediction in ), detection convolutional units (such as Figure 5 Reg convolution in), detection prediction unit (such as Figure 5 Reg prediction in ), confidence prediction unit (such as Figure 5 Obj prediction in ), splicing unit.

[0105] The input end of the one-dimensional convolution unit is the input end of the prediction sub-model, and the output end of the concatenation unit is the output end of the prediction sub-model.

[0106] The output end of the one-dimensional convolution unit is respectively connected to the input end of the semantic convolution unit and the input end of the detection convolution unit, the output end of the semantic convolution unit is connected to the input end of the semantic prediction unit, the output end of the detection convolution unit is respectively connected to the input end of the detection prediction unit and the input end of the confidence prediction unit, and the input end of the splicing unit is respectively connected to the output end of the semantic prediction unit, the output end of the detection prediction unit and the output end of the confidence prediction unit.

[0107] In some embodiments of the present application, the final features are input to the input end of the first prediction sub-model, and the high-level features are input to the input end of the second prediction sub-model. The semantic convolution unit and the detection convolution unit are both convolution operations. Taking the first prediction sub-model as an example, the detection prediction unit outputs multiple detection boxes (i.e., the prediction center points and prediction lengths of multiple predicted transient power quality problems, and the multiple detection boxes correspond to the multiple predicted transient power quality problems one by one), the semantic prediction unit outputs the prediction category corresponding to each detection box (such as voltage sag, voltage swell, short interruption, etc.), and the confidence prediction unit outputs the confidence of each detection box. The detection prediction unit is a target detection algorithm (such as the anchor-free algorithm). The semantic prediction unit (cls unit) and the confidence prediction unit (obj unit) are just simple one-dimensional convolutional layers, and there is only one layer. The inputs of these two prediction units correspond to the semantic convolutional unit and the detection convolutional unit respectively (the architecture and output size of these two convolutional units are the same). The difference between the two prediction units is that the output channel size of the semantic prediction unit is equal to the prediction category size, and the confidence prediction unit only outputs 1 channel. In addition, the length of each channel output by these two prediction units is the same (equal to the number of prediction boxes).

[0108] The first prediction sub-model outputs the prediction center point, prediction length, confidence and prediction category corresponding to multiple predicted transient power quality problems obtained according to the final features, and the second prediction sub-model outputs the prediction center point, prediction length, confidence and prediction category corresponding to multiple predicted transient power quality problems obtained according to the high-level features. Then, the splicing sub-model is used to splice the output of the first prediction sub-model and the output of the second prediction sub-model to obtain the recognition result of the power quality signal sequence, and the recognition result includes the output of the first prediction sub-model and the output of the second prediction sub-model.

[0109] It is worth mentioning that the recognition results obtained based on the information-rich final features and high-level features can improve the accuracy of the recognition results. At the same time, the recognition results include four aspects of data: prediction center point, prediction length, confidence and prediction category, which improves the comprehensiveness of power quality problem identification.

[0110] Step 15, construct a loss function based on all recognition results, and use the loss function to optimize the lightweight network model, feature fusion model and prediction model to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model.

[0111] The above loss function is used to describe the accuracy of all recognition results.

[0112] Specifically, the loss function is:

[0113] loss=loss_iou+loss_obj+loss_cls

[0114] Among them, loss represents the value of the loss function, loss_iou represents the position loss function, loss_obj represents the confidence loss function, and loss_cls represents the classification loss function:

[0115] loss_iou=1-IOU 2

[0116]

[0117] Among them, IOU represents the intersection-over-union ratio between all predicted center points and the true center points and the intersection-over-union ratio between all predicted lengths and the true lengths, y represents the set of true confidences of all predicted transient power quality problems, represents the set of confidences of all predicted transient power quality problems, c represents the set of true categories of transient power quality problems of all power quality signal sequences, Represents the set of prediction categories for all predicted transient power quality issues.

[0118] Exemplarily, the true confidence level of the above-mentioned prediction of transient power quality problems may be obtained by manual methods or the like.

[0119] In some embodiments of the present application, an optimization algorithm such as Adam can be used to optimize the lightweight network model, feature fusion model and prediction model using a loss function to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model. The optimization termination condition may be that the number of optimizations reaches a preset number, or the value of the loss function is less than a preset loss value.

[0120] It is worth mentioning that by constructing a loss function, the accuracy of the recognition results can be described. Using the loss function to optimize the lightweight network model, feature fusion model and prediction model can improve the performance of the lightweight network model, feature fusion model and prediction model, thereby improving the accuracy of the recognition results.

[0121] The above loss function is illustrated below with reference to a specific example.

[0122] like Figure 6 As shown, Figure 6 The horizontal axis represents the number of iterations, the vertical axis represents the loss value, and the curves are the value curve of the confidence loss function (obj_loss), the value curve of the classification loss function (cls_loss), the value curve of the position loss function (reg_loss), and the value curve of the total loss function (total_loss).

[0123] Step 16, obtain multiple current power quality signals of the distribution network to be identified, use the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to identify all current power quality signals, and obtain the final identification result of the distribution network to be identified.

[0124] The above-mentioned final identification result includes the prediction center point, prediction length, confidence and prediction category corresponding to the multiple final predicted transient power quality problems. The above-mentioned distribution network to be identified is a distribution network that needs to identify transient power quality problems, and the above-mentioned multiple current power quality signals are all power quality signals of the current time period of the distribution network to be identified. If the current time is 10 o'clock, the current time period can be 9:30 to 10 o'clock.

[0125] Specifically, based on all current power quality signals, the optimized lightweight network model is used to obtain the underlying features and high-level features, and then the optimized feature fusion model is used to fuse the underlying features and high-level features to obtain the final features. Based on the final features and high-level features, the prediction model is used to obtain the final recognition results.

[0126] Exemplarily, after obtaining the final identification result, the power quality of the distribution network to be identified in the current time period can be analyzed according to the prediction center point, prediction length, confidence level and prediction category corresponding to all the final predicted transient power quality problems given therein, and the distribution network to be identified can be maintained in time according to the number, prediction category, prediction center point, prediction length and confidence level of the final predicted transient power quality problems to improve the power quality of the distribution network to be identified. If the number of voltage sags among all the final predicted transient power quality problems is greater than that of other types, it means that the distribution network to be identified is prone to voltage sags, and distribution network maintenance is required to address the voltage sag problem.

[0127] It is worth mentioning that by obtaining the underlying features and high-level features, the information of the power quality signal sequence at two levels can be expressed, the comprehensiveness of the information can be increased, and the underlying features and high-level features can be integrated to obtain the final features, so that the final features contain information of the underlying features and high-level features, thereby improving the information richness of the final features. The recognition results are obtained based on the information-rich final features and high-level features, which can improve the accuracy of the recognition results. At the same time, the recognition results include four aspects of data: prediction center point, prediction length, confidence and prediction category, which improves the comprehensiveness of power quality problem identification.

[0128] The method for identifying transient power quality problems of the present application is exemplarily described below with reference to a specific example.

[0129] Precision, recall, F1 score, average precision (AP), start power quality signal error (Begin_loss), end power quality signal error (Over_loss), and length error (W_loss) are selected as evaluation indicators, and four transient power quality problems, namely voltage sag, voltage swell, voltage interruption, and transient impulse, are selected for classification and positioning accuracy evaluation. The method of the present application calculates and obtains the evaluation indicators after maximum value suppression with a confidence threshold of 0.6 and an intersection-over-combination threshold of 0.5 as shown in Table 1.

[0130]

[0131] Table 1

[0132] It can be seen that the method of the present application has a good classification effect on the four types of transient power quality problems, not only with high accuracy but also with high recall rate. This shows that the method of the present application can not only find the transient power quality problems that may exist in the power quality signal correctly, but also find them completely.

[0133] In addition, the traditional method is to find the starting and ending positions of transient power quality problems by judging whether the effective value of a half-cycle is within the threshold range. The positioning accuracy of this method for the starting and ending positions is half a cycle, that is, 10ms. As shown in Table 1, the error of the embodiment of the present invention in the starting and ending positions does not exceed 2.7ms at most, especially the positioning accuracy of transient impacts with very long duration can reach 0.05ms; the detection error of the duration does not exceed 3.5ms at most, which is far less than the required 20ms measurement error. This proves that the method of the present application has excellent positioning performance for transient power quality problems.

[0134] The following is an exemplary description of the device for identifying transient power quality problems provided in the present application.

[0135] like Figure 7 As shown, an embodiment of the present application provides a device for identifying transient power quality problems. The device 700 for identifying transient power quality problems includes:

[0136] An acquisition module 701 acquires multiple power quality signal sequences of a target distribution network; each power quality signal sequence includes power quality signals at multiple historical moments, and the power quality signals are used to describe the power quality of the target distribution network;

[0137] The feature acquisition module 702 acquires the bottom-level features and high-level features of each power quality signal sequence based on all power quality signals in each power quality signal sequence using a lightweight network model; the bottom-level features are used to describe the location information of all power quality signals in the power quality signal sequence, and the high-level features are used to obtain the semantic information of all power quality signals in the power quality signal sequence;

[0138] The fusion module 703 uses a feature fusion model to fuse the bottom-level features and high-level features of each power quality signal sequence to obtain a final feature;

[0139] The recognition result acquisition module 704 obtains the recognition result of each power quality signal sequence by using the prediction model based on all the final features and all the high-level features; the recognition result includes the prediction center point, prediction length, confidence and prediction category corresponding to multiple predicted transient power quality problems, the prediction length is the number of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the prediction center point is the central power quality signal of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the confidence is used to describe the probability of the predicted transient power quality problem actually occurring, and the prediction category is the category of the predicted transient power quality problem;

[0140] The optimization module 705 constructs a loss function according to all recognition results, and uses the loss function to optimize the lightweight network model, the feature fusion model and the prediction model to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model; the loss function is used to describe the accuracy of all recognition results;

[0141] The identification module 706 obtains multiple current power quality signals of the distribution network to be identified, and uses the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to identify all current power quality signals to obtain the final identification result of the distribution network to be identified; the final identification result includes the prediction center point, prediction length, confidence and prediction category corresponding to multiple final predicted transient power quality problems.

[0142] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0143] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0144] like Figure 8 As shown, an embodiment of the present application provides a terminal device. The terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 8 Only one processor is shown in the figure), a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 implements the steps of any of the above-mentioned method embodiments when executing the computer program D102.

[0145] Specifically, when the processor D100 executes the computer program D102, it obtains multiple power quality signal sequences of the target distribution network, and then obtains the underlying features and high-level features of each power quality signal sequence based on all power quality signals in each power quality signal sequence using a lightweight network model, and then fuses the underlying features and high-level features of the power quality signal sequence using a feature fusion model for each power quality signal sequence to obtain the final features, and then obtains the recognition result of each power quality signal sequence based on all the final features and all the high-level features using a prediction model, and then constructs a loss function based on all the recognition results, and uses the loss function to optimize the lightweight network model, the feature fusion model and the prediction model to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model, and finally obtains multiple current power quality signals of the distribution network to be identified, and uses the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to identify all current power quality signals to obtain the final recognition result of the distribution network to be identified. Among them, by obtaining the underlying features and the high-level features, the information of the power quality signal sequence at two levels can be expressed, the comprehensiveness of the information is increased, and the underlying features and the high-level features are merged to obtain the final features, so that the final features contain the information of the underlying features and the high-level features, and the information richness of the final features is improved. The recognition results are obtained based on the information-rich final features and high-level features, which can improve the accuracy of the recognition results. At the same time, the recognition results include four aspects of data: prediction center point, prediction length, confidence and prediction category, which improves the comprehensiveness of power quality problem identification.

[0146] The processor D100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0147] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may also be an external storage device of the terminal device D10, such as a plug-in hard disk, a smart memory card (SMC, SmartMedia Card), a secure digital (SD, Secure Digital) card, a flash card (Flash Card), etc. equipped on the terminal device D10. Further, the memory D101 may also include both an internal storage unit of the terminal device D10 and an external storage device. The memory D101 is used to store an operating system, an application program, a boot loader (BootLoader), data and other programs, such as the program code of the computer program, etc. The memory D101 may also be used to temporarily store data that has been output or is to be output.

[0148] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0149] An embodiment of the present application provides a computer program product. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0150] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device that can carry the computer program code to the transient power quality problem identification method device / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0151] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0152] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0153] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for identifying transient power quality problems, characterized in that: include: Acquire multiple power quality signal sequences of the target distribution network; each power quality signal sequence includes power quality signals at multiple historical moments, and the power quality signals are used to describe the power quality of the target distribution network; Based on all power quality signals in each power quality signal sequence, a lightweight network model is used to obtain low-level features and high-level features of each power quality signal sequence; the low-level features are used to describe the location information of all power quality signals in the power quality signal sequence, and the high-level features are used to obtain the semantic information of all power quality signals in the power quality signal sequence; For each power quality signal sequence, the feature fusion model is used to fuse the bottom-level features and the high-level features of the power quality signal sequence to obtain the final features; Based on all final features and all high-level features, a prediction model is used to obtain recognition results for each power quality signal sequence; the recognition results include a plurality of prediction center points, prediction lengths, confidences, and prediction categories corresponding to the predicted transient power quality problems, the prediction length being the number of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the prediction center point being the central power quality signal of all power quality signals corresponding to the duration period of the predicted transient power quality problem, the confidence being used to describe the probability of the predicted transient power quality problem actually occurring, and the prediction category being the category of the predicted transient power quality problem; Constructing a loss function according to all recognition results, and optimizing the lightweight network model, the feature fusion model and the prediction model using the loss function to obtain an optimized lightweight network model, an optimized feature fusion model and an optimized prediction model; the loss function is used to describe the accuracy of all recognition results; Acquire multiple current power quality signals of the distribution network to be identified, use the optimized lightweight network model, the optimized feature fusion model and the optimized prediction model to identify all current power quality signals, and obtain the final identification result of the distribution network to be identified; the final identification result includes the prediction center point, prediction length, confidence and prediction category corresponding to multiple final predicted transient power quality problems.

2. The identification method according to claim 1, characterized in that: The lightweight network model includes a first extraction sub-model, a second extraction sub-model, a third extraction sub-model, a fourth extraction sub-model, a fifth extraction sub-model, a sixth extraction sub-model, a first one-dimensional convolution sub-model and a second one-dimensional convolution sub-model connected in sequence; The first extraction sub-model, the second extraction sub-model, the third extraction sub-model, the fourth extraction sub-model, the fifth extraction sub-model, and the sixth extraction sub-model each include an extraction sub-model; The extraction sub-model comprises a one-dimensional convolution unit and a pooling unit connected in sequence, the input end of the one-dimensional convolution unit is the input end of the extraction sub-model, and the output end of the pooling unit is the output end of the extraction sub-model; The input end of the first extraction sub-model is the input end of the lightweight network model, the output end of the one-dimensional convolution unit in the fifth extraction sub-model is the first output end of the lightweight network model, and the output end of the second one-dimensional convolution sub-model is the second output end of the lightweight network model.

3. The identification method according to claim 2, characterized in that: The first one-dimensional convolution sub-model and the second one-dimensional convolution sub-model each include a one-dimensional convolution unit.

4. The identification method according to claim 3, characterized in that: The one-dimensional convolution unit includes a one-dimensional convolution layer, a batch normalization layer, an activation layer, and an attention mechanism layer connected in sequence; The input end of the one-dimensional convolution layer is the input end of the one-dimensional convolution unit, and the output end of the attention mechanism layer is the output end of the one-dimensional convolution unit.

5. The identification method according to claim 4, characterized in that: The feature fusion model includes a one-dimensional convolution sub-model, an upsampling sub-model and a feature splicing sub-model connected in sequence; The input end of the one-dimensional convolution sub-model and the input end of the feature splicing sub-model are both input ends of the feature fusion model, the input end of the feature splicing sub-model is connected to the first output end of the lightweight network model, the input end of the one-dimensional convolution sub-model is connected to the second output end of the lightweight network model, and the output end of the feature splicing sub-model is the output end of the feature fusion model.

6. The identification method according to claim 1, characterized in that: The prediction model includes a first prediction sub-model, a second prediction sub-model and a splicing sub-model; The input end of the splicing sub-model is connected to the output end of the first prediction sub-model and the output end of the second prediction sub-model respectively.

7. The identification method according to claim 6, characterized in that: The first prediction sub-model and the second prediction sub-model each include a prediction sub-model; The prediction sub-model includes a one-dimensional convolution unit, a semantic convolution unit, a semantic prediction unit, a detection convolution unit, a detection prediction unit, a confidence prediction unit, and a splicing unit; The input end of the one-dimensional convolution unit is the input end of the prediction sub-model, and the output end of the splicing unit is the output end of the prediction sub-model; The output end of the one-dimensional convolution unit is respectively connected to the input end of the semantic convolution unit and the input end of the detection convolution unit, the output end of the semantic convolution unit is connected to the input end of the semantic prediction unit, the output end of the detection convolution unit is respectively connected to the input end of the detection prediction unit and the input end of the confidence prediction unit, and the input end of the splicing unit is respectively connected to the output end of the semantic prediction unit, the output end of the detection prediction unit and the output end of the confidence prediction unit.

8. The identification method according to claim 1, characterized in that: The loss function is: loss=loss_iou+loss_obj+loss_cls Among them, loss represents the value of the loss function, loss_iou represents the position loss function, loss_obj represents the confidence loss function, and loss_cls represents the classification loss function: loss_iou=1-IOU 2 Among them, IOU represents the intersection-over-union ratio between all predicted center points and the true center points and the intersection-over-union ratio between all predicted lengths and the true lengths, y represents the set of true confidences of all predicted transient power quality problems, represents the set of confidences of all predicted transient power quality problems, c represents the set of true categories of transient power quality problems of all power quality signal sequences, Represents the set of prediction categories for all predicted transient power quality issues.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for identifying transient power quality problems according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for identifying transient power quality problems according to any one of claims 1 to 8 is implemented.

Images