Dual-mode communication meter reading abnormality analysis method and related device

Abstract

The present application provides a dual-mode communication meter reading anomaly analysis method and related devices, the method comprising: obtaining first power data of a target channel area and a target communication state of a dual-mode channel; performing data preprocessing and feature extraction on the first power data to obtain second power data; performing model training based on the second power data to obtain an anomaly detection model; analyzing meter reading information in the target channel area to obtain a first anomaly analysis result; if the power line channel and / or the wireless channel is in a blocked state, determining the target communication state as a second anomaly analysis result; determining a first target anomaly analysis result based on the first anomaly analysis result and the second anomaly analysis result; if both the power line channel and the wireless channel are in an unblocked state, performing a channel transmission test and a channel switching test to obtain a third anomaly analysis result, and determining a second target anomaly analysis result based on the first anomaly analysis result and the third anomaly analysis result.

Description

Technical Field

[0001] The present application relates to the field of dual-mode communication technology, and in particular to a dual-mode communication meter reading anomaly analysis method and related devices. Background Art

[0002] In the context of smart grids, accurate energy metering is crucial for the economic operation and management of power systems. Dual-mode communication technology, by combining power line communication and wireless communication, effectively improves data acquisition success rates and system stability. However, in practical applications, various abnormal faults may occur, which can affect the stability of the power system.

[0003] Therefore, how to quickly and accurately identify abnormal fault conditions in the dual-mode communication process needs to be solved urgently. Summary of the Invention

[0004] The embodiment of the present application provides a dual-mode communication meter reading anomaly analysis method and related devices, which can quickly and accurately identify abnormal fault conditions in the dual-mode communication process by comprehensively considering power data and communication status, thereby improving the stability and reliability of the power system.

[0005] In a first aspect, an embodiment of the present application provides a dual-mode communication meter reading anomaly analysis method, which is applied to an anomaly analysis module of a dual-mode communication system. The dual-mode communication system includes the anomaly analysis module, an electric meter measurement module, and a channel area, where a is an integer greater than 1. The method includes:

[0006] Acquiring first power data of a target channel area during a preset time period and a target communication state of the dual-mode channel; the target channel area is any one of the a channel areas; the dual-mode channel includes a power line channel and a wireless channel; and the communication state includes a smooth state and a blocked state;

[0007] performing data preprocessing and feature extraction on the first power data to obtain second power data;

[0008] Performing model training based on the second power data to obtain an anomaly detection model;

[0009] Acquire meter reading information of the electric meter metering module in the preset time period corresponding to the target channel area;

[0010] Analyzing the meter reading information according to the anomaly detection model to obtain a first anomaly analysis result;

[0011] If the communication state of the power line channel and / or the wireless channel in the target communication state is the blocked state, determining the target communication state as a second abnormality analysis result; determining a first target abnormality analysis result according to the first abnormality analysis result and the second abnormality analysis result;

[0012] If the communication states of the power line channel and the wireless channel in the target communication state are both the unblocked state, performing a channel transmission test on the power line channel and the wireless channel to obtain a first test result; and performing a channel switching test on the power line channel and the wireless channel to obtain a second test result;

[0013] A third abnormality analysis result is determined according to the first test result and the second test result, and a second target abnormality analysis result is determined according to the first abnormality analysis result and the third abnormality analysis result.

[0014] In a second aspect, an embodiment of the present application provides a dual-mode communication meter reading anomaly analysis device, which is applied to an anomaly analysis module of a dual-mode communication system. The dual-mode communication system includes the anomaly analysis module, an electric meter measurement module, and a channel area, where a is an integer greater than 1. The device includes an acquisition module, a processing module, a training module, an analysis module, a determination module, and a testing module, wherein:

[0015] The acquisition module is configured to acquire first power data of a target channel area during a preset time period and a target communication state of the dual-mode channel; the target channel area is any one of the a channel areas; the dual-mode channel includes a power line channel and a wireless channel; and the communication state includes a smooth state and a blocked state.

[0016] The processing module is configured to perform data preprocessing and feature extraction on the first power data to obtain second power data;

[0017] The training module is configured to perform model training based on the second power data to obtain an anomaly detection model;

[0018] The acquisition module is further configured to acquire meter reading information of the electric meter measurement module in the preset time period corresponding to the target channel area;

[0019] The analysis module is configured to analyze the meter reading information according to the anomaly detection model to obtain a first anomaly analysis result;

[0020] The determining module is configured to determine that the target communication state is a second abnormality analysis result if the communication state of the power line channel and / or the wireless channel in the target communication state is the blocked state; and determine a first target abnormality analysis result based on the first abnormality analysis result and the second abnormality analysis result;

[0021] The testing module is configured to, if the communication states of the power line channel and the wireless channel in the target communication state are both in the unblocked state, perform a channel transmission test on the power line channel and the wireless channel to obtain a first test result; and perform a channel switching test on the power line channel and the wireless channel to obtain a second test result;

[0022] The determination module is further configured to determine a third abnormality analysis result based on the first test result and the second test result, and to determine a second target abnormality analysis result based on the first abnormality analysis result and the third abnormality analysis result.

[0023] In a third aspect, an embodiment of the present application provides an electronic device comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program comprises instructions for executing the steps of any method of the first aspect of the embodiment of the present application.

[0024] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the above-mentioned computer-readable storage medium stores a computer program for electronic data exchange, wherein the above-mentioned computer program enables a computer to execute part or all of the steps described in any method of the first aspect of the embodiment of the present application.

[0025] In a fifth aspect, embodiments of the present application provide a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a computer to execute some or all of the steps described in any method of the first aspect of the embodiments of the present application. The computer program product may be a software installation package.

[0026] By implementing the embodiments of the present application and comprehensively considering the power data and communication status, abnormal fault conditions in the dual-mode communication process can be quickly and accurately identified, thereby improving the stability and reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0028] Figure 1 This is a system architecture diagram of a dual-mode communication system provided by an embodiment of the present application;

[0029] Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0030] Figure 3 This is a flow chart of a dual-mode communication meter reading anomaly analysis method provided in an embodiment of the present application;

[0031] Figure 4 This is a block diagram of the functional modules of a dual-mode communication meter reading anomaly analysis device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0033] The terms "first," "second," and the like in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, not to describe a particular order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0034] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document indicates that the associated objects are in an "or" relationship. The "plurality" appearing in the embodiments of this application refers to two or more.

[0035] In the embodiments of the present application, "at least one item" or similar expressions refers to any combination of these items, including any combination of single items or plural items, and refers to one or more, and multiple refers to two or more. For example, at least one item (item) of a, b, or c can represent the following seven situations: a, b, c, a and b, a and c, b and c, a, b, and c. Among them, each of a, b, and c can be an element or a set containing one or more elements.

[0036] The "connection" appearing in the embodiments of the present application refers to various connection methods such as direct connection or indirect connection to achieve communication between devices, and the embodiments of the present application do not impose any limitations on this.

[0037] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0038] In the context of smart grids, accurate energy metering is crucial for the economic operation and management of power systems. Dual-mode communication technology, by combining power line communication and wireless communication, effectively improves data acquisition success rates and system stability. However, in practical applications, various abnormal faults may occur, which can affect power system stability. Therefore, how to quickly and accurately identify abnormal faults during dual-mode communication is an urgent problem to be solved.

[0039] In order to solve the above problems, the embodiment of the present application provides a dual-mode communication meter reading anomaly analysis method and related devices, which are applied to the anomaly analysis module of the dual-mode communication system, and the dual-mode communication system includes the anomaly analysis module, the meter metering module and a channel area, a is an integer greater than 1, and obtains the first power data of the preset time period of the target channel area and the target communication state of the dual-mode channel; the target channel area is any one of the a channel areas; the dual-mode channel includes a power line channel and a wireless channel; the communication state includes a smooth state and a blocked state; the first power data is preprocessed and feature extracted to obtain the second power data; the model is trained according to the second power data to obtain the anomaly detection model; the meter reading information of the meter metering module in the preset time period corresponding to the target channel area is obtained; according to the anomaly detection The model analyzes the meter reading information to obtain a first abnormality analysis result; if the communication status of the power line channel and / or the wireless channel in the target communication state is the blocked state, the target communication state is determined to be a second abnormality analysis result; a first target abnormality analysis result is determined based on the first abnormality analysis result and the second abnormality analysis result; if the communication status of the power line channel and the wireless channel in the target communication state is both the unblocked state, a channel transmission test is performed on the power line channel and the wireless channel to obtain a first test result; a channel switching test is performed on the power line channel and the wireless channel to obtain a second test result; a third abnormality analysis result is determined based on the first test result and the second test result, and a second target abnormality analysis result is determined based on the first abnormality analysis result and the third abnormality analysis result. By comprehensively considering the power data and communication status, abnormal fault conditions in the dual-mode communication process can be quickly and accurately identified, thereby improving the stability and reliability of the power system.

[0040] The following combination Figure 1 The system architecture of a dual-mode communication system for executing a dual-mode communication meter reading anomaly analysis method in an embodiment of the present application is described. Figure 1 This is a system architecture diagram of a dual-mode communication system provided in an embodiment of the present application. The dual-mode communication system includes an abnormality analysis module, an electricity meter measurement module and a channel areas, where a is an integer greater than 1.

[0041] The anomaly analysis module monitors the metering module and channel areas in the dual-mode communication system. If an anomaly is detected in the communication link of the metering module or channel area, the module analyzes the anomaly and issues an early warning signal if a fault is confirmed, prompting maintenance personnel to inspect and repair the fault. These faults may include meter data errors, communication interruptions, equipment failures, and other issues, which are not specifically defined here.

[0042] The dual-mode communication system includes a different channel area, each channel area has multiple dual-mode communication modules, and data transmission can be performed through the power line channel or wireless channel in each channel area.

[0043] Among them, the electricity meter measurement module is used to collect, process and transmit electricity data. The electricity meter measurement module can collect various metering data of the electricity meter in real time, such as voltage, current, power, energy consumption, etc., which are not specifically limited here. In addition, the electricity meter measurement module can also work in conjunction with a channel area, so that electricity data can be smoothly transmitted between different channel areas through dual-mode channels. Among them, the dual-mode channel includes a power line channel and a wireless channel. Each channel area includes multiple dual-mode communication modules for managing data communications within the area, and the electricity meter measurement module maintains a stable connection with the communication links of these channel areas to ensure the timeliness and reliability of electricity data transmission.

[0044] It can be seen that by detecting power data and communication status through the abnormal analysis module, abnormal fault conditions in the dual-mode communication process can be identified quickly and accurately, thereby improving the stability and reliability of the power system.

[0045] The following combination Figure 2 The electronic device in the embodiment of the present application is described. Figure 2 is a structural diagram of an electronic device provided in an embodiment of the present application, such as Figure 2 As shown, the electronic device includes one or more processors, a memory, a communication interface and one or more programs, and the processor is communicatively connected with the memory and the communication interface via an internal communication bus.

[0046] Among them, the processor is mainly used for:

[0047] Acquiring first power data of a target channel area during a preset time period and a target communication state of a dual-mode channel; the target channel area is any one of a channel area; the dual-mode channel includes a power line channel and a wireless channel; and the communication state includes a smooth state and a blocked state;

[0048] performing data preprocessing and feature extraction on the first power data to obtain second power data;

[0049] Performing model training based on the second power data to obtain an anomaly detection model;

[0050] Obtaining meter reading information of an electric meter metering module in a preset time period corresponding to a target channel area;

[0051] Analyze the meter reading information according to the anomaly detection model to obtain a first anomaly analysis result;

[0052] If the communication state of the power line channel and / or the wireless channel in the target communication state is a blocked state, determining the target communication state as a second abnormality analysis result; determining the first target abnormality analysis result according to the first abnormality analysis result and the second abnormality analysis result;

[0053] If the communication states of the power line channel and the wireless channel in the target communication state are both unblocked, performing a channel transmission test on the power line channel and the wireless channel to obtain a first test result; performing a channel switching test on the power line channel and the wireless channel to obtain a second test result;

[0054] A third abnormality analysis result is determined according to the first test result and the second test result, and a second target abnormality analysis result is determined according to the first abnormality analysis result and the third abnormality analysis result.

[0055] The one or more programs are stored in the above-mentioned memory and are configured to be executed by the above-mentioned processor, and the one or more programs include instructions for executing any step in the above-mentioned method embodiment.

[0056] Among them, the processor can be, for example, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It can implement or execute the various exemplary logic blocks, units and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. The communication unit can be a communication interface, a transceiver, a transceiver circuit, etc., and the storage unit can be a memory.

[0057] The memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory may be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DR RAM).

[0058] It is understood that the electronic device may include more or fewer structural elements than those in the above structural block diagram, for example, including a power module, physical buttons, Wi-Fi module, speaker, Bluetooth module, sensor, display module, etc., which are not limited here. It is understood that the electronic device may be equipped with Figure 1 The system architecture described.

[0059] After understanding the software and hardware architecture of this application, Figure 3 A dual-mode communication meter reading abnormality analysis method in an embodiment of the present application is described. Figure 3 This is a flow chart of a dual-mode communication meter reading anomaly analysis method provided in an embodiment of the present application, which is applied to an anomaly analysis module of a dual-mode communication system. The dual-mode communication system includes the anomaly analysis module, an electric meter measurement module, and a channel area, where a is an integer greater than 1. The method specifically includes the following steps:

[0060] Step S301 : acquiring first power data of a target channel area in a preset time period and a target communication state of a dual-mode channel.

[0061] The target channel area is any one of the a channel areas; the dual-mode channel includes a power line channel and a wireless channel; and the communication state includes a smooth state and a blocked state.

[0062] Specifically, the preset time period includes but is not limited to one day, one month, one quarter, and one year, which are not specifically limited here. The first power data of the power system in the target channel area can be obtained, and the communication status of the dual-mode channel in the target channel area can be detected to obtain the target communication status. Among them, the attenuation of the signal during transmission on the power line channel, the packet loss rate and other indicators can be detected. If the packet loss rate is extremely low and the signal attenuation is within the normal range, it can be determined to be a smooth state. On the contrary, if the packet loss is serious and the signal cannot be transmitted normally, it is a blocked state. For wireless channels, its signal strength, connection stability, data transmission success rate, etc. can be checked. A strong signal and a high transmission success rate indicate a smooth state. Frequent disconnections and transmission failures indicate a blocked state.

[0063] Step S302: performing data preprocessing and feature extraction on the first power data to obtain second power data.

[0064] The step of performing data preprocessing and feature extraction on the first power data to obtain the second power data specifically includes:

[0065] Perform data cleaning and data standardization on the first power data to obtain reference first power data; perform feature extraction on the reference first power data to obtain power feature data; the power feature data includes p feature sub-data; the feature sub-data includes any one of the following: electrical quantity feature, time domain feature, frequency domain feature; p is an integer greater than 1; the p feature sub-data are randomly divided into C random subsets, and C decision trees are generated based on the C random subsets; C is a positive integer less than p; the relative importance of each feature sub-data in the p feature sub-data is calculated according to the C decision trees to obtain p relative importances; the p relative importances are sorted in descending order to obtain a first sorting result; obtain the feature sub-data corresponding to the first M relative importances in the first sorting result to obtain the second power data; M is a positive integer less than p.

[0066] Specifically, first, by performing data cleaning on the first power data, erroneous, duplicate, and incomplete power data can be removed to ensure the accuracy and completeness of the data. The first power data is then normalized, i.e., the first power data is scaled to the range of 0 to 1, thereby obtaining reference first power data. Feature extraction is performed on the reference first power data to obtain power feature data. The power feature data includes p feature sub-data, and the feature sub-data includes any of the following: electrical quantity features, time domain features, and frequency domain features. It should be noted that electrical quantity features include, but are not limited to, basic electrical quantity parameters such as voltage, current, active power, and reactive power in the reference first power data, which are used to intuitively reflect the basic operating status of the power system; time domain features include, but are not limited to, waveform characteristics, amplitude change trends, and periodic characteristics of the power signal corresponding to the reference first power data over time, which are used to analyze the dynamic changes of the power signal in the time dimension; and frequency domain features include, but are not limited to, spectrum distribution and frequency components corresponding to the reference first power data, which are used to reflect the frequency stability of the power system.

[0067] Next, the p feature sub-data are randomly divided into C random subsets, and C decision trees are generated based on the C random subsets, so that the generated C decision trees have diversity and avoid the overfitting problem of a single decision tree. Among them, a decision tree corresponding to each random subset can be generated according to a preset decision tree algorithm, thereby obtaining C decision trees. Then, the relative importance of each feature sub-data in the p feature sub-data is calculated based on the C decision trees to obtain p relative importances. The p relative importances are sorted in order from large to small to obtain a first sorting result. Finally, the feature sub-data corresponding to the first M relative importances in the first sorting result are obtained to obtain the second power data.

[0068] It can be seen that by selecting characteristic sub-data with higher relative importance as the second power data, the characteristic sub-data can be streamlined, focusing on the most critical features for power system analysis, reducing the data dimensions for subsequent processing, improving calculation efficiency, and avoiding unimportant characteristic data from interfering with subsequent analysis results.

[0069] The step of calculating the relative importance of each of the p feature sub-data according to the C decision trees to obtain p relative importances specifically includes:

[0070] The Gini index value of the reference feature sub-data at the node m of the decision tree i is calculated according to a preset Gini index calculation formula to obtain a reference Gini index value; the reference feature sub-data is any one feature sub-data among the p feature sub-data; the decision tree i is the i-th decision tree among the C decision trees; the decision tree i includes S nodes, and the node m is the m-th node among the S nodes; S is an integer greater than 1;

[0071] The Gini index calculation formula is as follows:

[0072]

[0073] Among them, G ini,i,m represents the reference Gini index value; p m,k represents the ratio between the number of messages of category k in node m of the decision tree i and the number of messages in node m, where each message corresponds to a reference feature sub-data and a message category;

[0074] Calculate the reference Gini index value according to a preset change calculation formula to obtain a reference change of the reference feature sub-data at the node m of the decision tree i;

[0075] The change calculation formula is as follows:

[0076] D vim,β,i,m =G ini,i,m -G imi,i,e -G imi,i,r

[0077] Among them, D vim,β,i,m represents the reference variation; β represents the reference feature sub-data; G imi,i,e G represents the Gini index value of the first new node after the node m of the decision tree i branches; imi,i,r represents the Gini index value of the second new node after the node m of the decision tree i branches;

[0078] Importance calculation is performed on the reference variation to obtain a reference relative importance; the reference relative importance is the relative importance corresponding to the reference feature sub-data in the p relative importances.

[0079] The step of calculating the importance of the reference variation to obtain the reference relative importance comprises the following steps:

[0080] Calculating the reference change amount according to a preset first importance calculation formula to obtain a reference first importance;

[0081] The first importance calculation formula is as follows:

[0082]

[0083] Among them, D vim,β Indicates the first importance of the reference;

[0084] Calculating the reference first importance according to a preset second importance calculation formula to obtain the reference relative importance;

[0085] The second importance calculation formula is as follows:

[0086]

[0087] in, Indicates the relative importance of the reference; D vim.n Indicates the first importance of the nth reference; represents the sum of the p reference first importances.

[0088] It can be seen that by calculating the relative importance of different feature data and screening out key feature data with higher relative importance for subsequent processing, it is convenient to improve the performance and accuracy of the anomaly detection model, thereby ensuring the safe and stable operation of the power system.

[0089] Step S303: Perform model training based on the second power data to obtain an anomaly detection model.

[0090] The step of performing model training based on the second power data to obtain an anomaly detection model specifically includes:

[0091] Obtain a preset first model; the first model includes a filtering network and a classification network; obtain x model parameter combinations corresponding to the first model; each model parameter combination includes a filtering network parameter and a classification network parameter; x is an integer greater than 1; configure the first model according to the filtering network parameters and the classification network parameters of each model parameter combination in the x model parameter combinations to obtain x second models; perform feature extraction on the second power data through the filtering network of each second model in the x second models to obtain x target feature data; perform feature classification on the corresponding target feature data in the x target feature data through the classification network of each second model in the x second models to obtain x classification results; evaluate the x classification results according to a preset loss function to obtain x loss function values; each loss function value corresponds to a second model; determine that the second model corresponding to the minimum loss function value in the x loss function values ​​is the anomaly detection model.

[0092] Specifically, first, a preset first model is obtained. This first model is a one-dimensional convolutional neural network model. This first model includes a filtering network and a classification network. The filtering network includes convolutional layers and pooling layers, while the classification network includes fully connected layers and a Softmax layer. The Softmax layer is an activation function layer commonly used in multi-classification tasks. Its primary function is to convert the model output into a probability distribution, such that the predicted value for each class is between 0 and 1, and the sum of the probabilities of all classes is 1. Next, x model parameter combinations corresponding to the first model are obtained. Each model parameter combination includes filtering network parameters and classification network parameters. The filtering network parameters include, but are not limited to, the convolution kernel size, the number of convolution kernels, and the pooling size. The classification network parameters include, but are not limited to, the number of neurons and the activation function, which are not specifically defined here. The first model is then configured based on the filtering network parameters and classification network parameters of each of the x model parameter combinations to obtain x second models. The filtering network of each second model extracts features from the second power data to obtain x target feature data. The classification network of each second model then classifies the corresponding target feature data to obtain x classification results. The x classification results are evaluated using a preset loss function to obtain x loss function values. Each loss function value corresponds to a second model. The loss function includes, but is not limited to, a mean squared error loss function and a cross entropy loss function, which are not specifically limited here. Finally, the second model corresponding to the minimum loss function value among the x loss function values ​​is determined as the anomaly detection model.

[0093] It can be seen that by selecting the second model corresponding to the minimum loss function value as the anomaly detection model, the operating status of the power system can be detected more accurately, and abnormal situations can be discovered in time, providing strong guarantees for the stable operation, fault warning and rapid maintenance of the power system.

[0094] In one possible embodiment, the second power data is input as an input sample into the first model, i.e., a one-dimensional convolutional neural network model, wherein the first layer is a convolutional layer, and the one-dimensional convolution operation formula of the convolutional layer is as follows:

[0095]

[0096] in, represents the vector obtained after the j-th convolution calculation of layer l; P represents the number of input feature vectors; represents the d-th input feature vector of layer l; Represents related operations; represents the jth convolution kernel of layer l convolved with the dth input feature vector; Represents the j-th bias vector of layer l.

[0097] The Rectified Linear Unit (ReLU) is used as the activation function in the convolutional layer. This activation function has no gradient dissipation problem, has a faster convergence speed, and can improve the sparsity of the network and effectively prevent overfitting. The definition of this activation function is as follows:

[0098]

[0099] in, represents the e-th output after the j-th convolution operation of layer l; express The activation value of .

[0100] Next, the feature information of the second power data is dimensionality-reduced through the pooling layer, while retaining the most representative feature information to reduce computational cost and prevent overfitting. The pooling operation can be performed by using the maximum value of the local receptive field as the output, or by using the mean value of the local receptive field as the output through mean pooling, which is not specifically limited here. Among them, the l+1 layer is the pooling layer, and the calculation formula corresponding to its maximum pooling is as follows:

[0101]

[0102] in, represents the activation value of the hth neuron of the jth feature vector in layer l; w represents the width of the pooling area; Represents the pooling result corresponding to the g-th neuron of the j-th eigenvector in the l+1 layer.

[0103] The classification network consists of a fully connected layer and a Softmax layer. The fully connected layer flattens the output of the pooling layer into a one-dimensional feature vector, fully connects the end to the end, and applies the ReLU activation function to the output of the fully connected layer, enabling the neural network to learn nonlinear relationships and enhancing the network's expressive power. The Softmax layer is used for multi-classification problems and can convert the results into interpretable class probabilities. Assuming a class label y∈{1,2,…,R}, the conditional probability that a sample x belongs to class r is calculated as follows:

[0104]

[0105] Where p(y=r|x; Θ) represents the conditional probability that sample x belongs to category r; Θ represents the set of weight vectors, including weight vectors of all categories; r represents category r; θ r Represents the weight vector of category r; θ j represents the weight vector of category j; Represents vector θ rThe dot product of the input sample x is the linear discriminant score of the category; Represents vector θ j The dot product of the input sample x; express The exponential function is used to convert the score into a positive number to avoid negative probability values; express R represents the total number of categories.

[0106] Finally, the loss function value of the model is calculated by the cross entropy loss function, which is as follows:

[0107]

[0108] Among them, U represents the total number of samples; I{y u =r} represents the indicator function, when the sample x u The true label y u When I{y u =r}=1, otherwise I{y u =r}=0.

[0109] In order to minimize the loss function value of the model, it is necessary to optimize the weights of the neural network, and the optimizer uses the backpropagation algorithm to complete this. The corresponding calculation formula is as follows:

[0110]

[0111] Among them, θ * represents the optimal model parameters corresponding to the second model, that is, the model parameters corresponding to the anomaly detection model, which include the filtering network parameters and the classification network parameters; L(f(x;θ),y) represents the loss function; f(x;θ) represents the output value of the model; y represents the target value of the model.

[0112] It can be seen that by training and optimizing the anomaly detection model, the accuracy and reliability of anomaly analysis of meter reading information can be improved, which makes it easier to combine historical data and expert knowledge to analyze the causes of anomalies and provide corresponding solutions and suggestions.

[0113] Step S304: acquiring meter reading information of the electric meter measuring module in the preset time period corresponding to the target channel area.

[0114] Specifically, the meter reading information includes but is not limited to electricity consumption data, power data, voltage and current data, power factor, and timestamp information, which are not specifically limited here.

[0115] Step S305: Analyze the meter reading information according to the anomaly detection model to obtain a first anomaly analysis result.

[0116] Specifically, the meter reading information can be input into the anomaly detection model, and the data in the meter reading information can be analyzed and detected by the anomaly detection model to obtain a corresponding first anomaly analysis result. If the meter reading information matches the normal pattern of power data, the first anomaly analysis result is determined to be that the meter reading information is in a normal state. If the meter reading information does not match the normal pattern, the first anomaly analysis result is determined to be that the meter reading information is in an abnormal state. The corresponding anomaly type is determined based on the abnormal state. For example, if the power value in the meter reading information is far above the normal power value range, the anomaly type is determined to be a power overload anomaly; if the power value is far below the normal power value range, the anomaly type is determined to be a power underload anomaly. The anomaly types include, but are not limited to, power anomalies, voltage anomalies, or power consumption anomalies, which are not specifically limited here.

[0117] Step S306: If the communication state of the power line channel and / or the wireless channel in the target communication state is the blocked state, determine the target communication state as the second abnormality analysis result; and determine the first target abnormality analysis result based on the first abnormality analysis result and the second abnormality analysis result.

[0118] Specifically, if the communication state of the power line channel and / or wireless channel in the target communication state is blocked, the target communication state is determined to be a second abnormality analysis result, and the first abnormality analysis result and the second abnormality analysis result are integrated to obtain a first target abnormality analysis result. This first target abnormality analysis result can provide maintenance personnel with comprehensive abnormality analysis information, facilitating rapid determination of the complexity and urgency of the abnormality, thereby enabling timely implementation of appropriate treatment measures.

[0119] In one possible embodiment, a test signal can be modulated and demodulated to a frequency band corresponding to the power line channel, then sent to the power line channel. Meter reading is performed based on a routing table in the metering module. If the meter reading is successful, the communication state of the power line channel is determined to be unblocked. If the meter reading fails, the communication state of the power line channel is determined to be blocked. Alternatively, the communication channel can be switched to a wireless channel, and a test signal can be sent via the wireless channel. If the test signal is successfully transmitted, the communication state of the wireless channel is determined to be unblocked. If the test signal fails to be transmitted, the communication state of the wireless channel is determined to be blocked.

[0120] It can be seen that by conducting targeted tests on the power line channel and the wireless channel respectively, the communication status of the dual-mode communication system can be fully and accurately grasped, and the reliability and performance of the dual-mode communication system can be improved.

[0121] Step S307: If the communication states of the power line channel and the wireless channel in the target communication state are both the unblocked states, a channel transmission test is performed on the power line channel and the wireless channel to obtain a first test result; and a channel switching test is performed on the power line channel and the wireless channel to obtain a second test result.

[0122] The target channel area includes b dual-mode communication modules, where b is an integer greater than 1; the first test result includes b reference first test results; each reference first test result corresponds to a dual-mode communication module; and performing a channel transmission test on the power line channel and the wireless channel to obtain the first test result specifically comprises the following steps:

[0123] The first test message is transmitted to the target node through the power line channel and the wireless channel respectively through the first dual-mode communication module, so that the target node receives the first test message through the power line channel and generates a first response message to be transmitted to the first dual-mode communication module, and receives the first test message through the wireless channel and generates a second response message to be transmitted to the first dual-mode communication module; the first dual-mode communication module is any one of the b dual-mode communication modules; the sending time of the first test message, the first receiving time of the first response message and the second receiving time of the second response message are obtained; the first receiving time and the second receiving time of the second response message are determined. The difference between the sending time is the first transmission duration; the difference between the second receiving time and the sending time is determined to be the second transmission duration; the first test message is compared with the first response message and the second response message respectively to obtain a first bit error rate and a second bit error rate; if the first transmission duration is less than a preset first transmission duration threshold, the second transmission duration is less than a preset second transmission duration threshold, and the first bit error rate and the second bit error rate are both less than a preset bit error rate threshold, then the first state of the first dual-mode communication module is determined to be a normal state; otherwise, the first state is determined to be an abnormal state; the first state is determined to be a reference first test result corresponding to the first dual-mode communication module.

[0124] In one possible embodiment, the transmission duration and bit error rate of dual-mode communication modules in a channel area can be analyzed, and the dual-mode communication modules can be classified according to their transmission duration and bit error rate using a preset cluster analysis algorithm, thereby obtaining multiple module sets. For module sets with higher bit error rates or longer transmission times, the testing frequency of these module sets can be increased to collect more data samples for more accurate performance assessment. For module sets with lower bit error rates or lower transmission times, the testing frequency of these module sets can be reduced to conserve testing resources.

[0125] The second test result includes a plurality of reference second test results; each reference second test result corresponds to a dual-mode communication module; and the channel switching test is performed on the power line channel and the wireless channel to obtain the second test result, specifically comprising the following steps:

[0126] Determine that the first dual-mode communication module is in a first working mode; the first working mode is that the power line channel is turned on and the wireless channel is turned off; transmit a preset second test message through the power line channel through the first dual-mode communication module, and record the third reception time of the third response message; switch the channel of the first dual-mode communication module so that the first dual-mode communication module is in a second working mode; the second working mode is that the power line channel is turned off and the wireless channel is turned on; transmit the second test message through the wireless channel through the first dual-mode communication module, and record the fourth reception time of the fourth response message; obtain a reference transmission duration of the wireless channel; determine the difference between the fourth reception time and the third reception time as the first transmission duration; determine the difference between the first transmission duration and the reference transmission duration as the first switching delay; if the first switching delay is less than the preset switching delay threshold, determine that the second state of the first dual-mode communication module is a normal state; otherwise, determine that the second state is an abnormal state; determine that the second state is the reference second test result corresponding to the first dual-mode communication module.

[0127] In one possible embodiment, the first dual-mode communication module can also be set to the second working mode, that is, the power line channel is closed and the wireless channel is turned on. Then, the second switching delay of the first dual-mode communication module from the second working mode to the first working mode is calculated. It should be noted that the working mode switching is repeated multiple times on the first dual-communication mode module to obtain multiple second switching delays, and the average delay is determined based on the average value of the multiple second switching delays, and the channel switching performance of the first dual-mode communication model is evaluated based on the average delay. Among them, the success rate of the first dual-mode communication model in multiple working mode switching processes can also be counted to evaluate the stability and reliability of the first dual-mode communication model.

[0128] In one possible embodiment, each of the b dual-mode communication modules can be tested. For example, a preset heartbeat packet mechanism is used to determine whether the dual-mode communication module is online. That is, each dual-mode communication module sends a heartbeat packet to the concentrator or main station of the dual-mode communication system every 1 minute. The concentrator or main station monitors the reception of the heartbeat packet. If the concentrator or main station does not receive the heartbeat packet of a dual-mode communication module for more than 3 consecutive minutes, it is considered that the dual-mode communication module is not online, that is, the dual-mode communication module is in an abnormal state. The log records of the dual-mode communication module can also be obtained and analyzed to obtain abnormal conditions that occur during the operation of the dual-mode communication module. The log records include various events such as startup, restart, error, and communication anomaly. Among them, the dual-mode communication module can be updated with firmware, the corresponding abnormal problems can be repaired, and the stability and correctness of the dual-mode communication module can be improved.

[0129] It can be seen that testing and updating the dual-mode communication module can greatly improve the reliability, stability and performance of the dual-mode communication module, ensure the normal communication function of the dual-mode communication system, and reduce various risks and losses caused by dual-mode communication module failures.

[0130] Step S308: determining a third abnormality analysis result according to the first test result and the second test result, and determining a second target abnormality analysis result according to the first abnormality analysis result and the third abnormality analysis result.

[0131] Specifically, the first test result and the second test result are integrated to obtain a third anomaly analysis result. The first test result represents the channel transmission test results for the power line channel and the wireless channel, and the second test result represents the channel switching test results for the power line channel and the wireless channel. The first and third anomaly analysis results are then integrated to obtain a second target anomaly analysis result.

[0132] It can be seen that through the above method, power data and communication status can be comprehensively considered, and abnormal fault conditions in the dual-mode communication process can be quickly and accurately identified, thereby improving the stability and reliability of the power system.

[0133] The above mainly introduces the solution of the embodiment of the present application from the perspective of the execution process of the method side. It is understandable that, in order to realize the above functions, the electronic device includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiment provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner 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.

[0134] The embodiment of the present application can divide the functional units of the electronic device according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0135] In the case of dividing each functional module into corresponding functional modules, Figure 4 This is a functional module block diagram of a dual-mode communication meter reading anomaly analysis device provided in an embodiment of the present application, which is applied to an anomaly analysis module of a dual-mode communication system. The dual-mode communication system includes the anomaly analysis module, an electricity meter measurement module, and a channel area, where a is an integer greater than 1. The dual-mode communication meter reading anomaly analysis device 400 includes an acquisition module 410, a processing module 420, a training module 430, an analysis module 440, a determination module 450, and a testing module 460, wherein:

[0136] The acquisition module 410 is configured to acquire first power data of a target channel region during a preset time period and a target communication state of the dual-mode channel; the target channel region is any one of the a channel regions; the dual-mode channel includes a power line channel and a wireless channel; and the communication state includes a smooth state and a blocked state.

[0137] The processing module 420 is configured to perform data preprocessing and feature extraction on the first power data to obtain second power data;

[0138] The training module 430 is configured to perform model training based on the second power data to obtain an anomaly detection model;

[0139] The acquisition module 410 is further configured to acquire meter reading information of the electric meter measurement module in the preset time period corresponding to the target channel area;

[0140] The analysis module 440 is configured to analyze the meter reading information according to the anomaly detection model to obtain a first anomaly analysis result;

[0141] The determining module 450 is configured to determine that the target communication state is a second abnormality analysis result if the communication state of the power line channel and / or the wireless channel in the target communication state is the blocked state; and determine a first target abnormality analysis result based on the first abnormality analysis result and the second abnormality analysis result;

[0142] The testing module 460 is configured to, if the communication states of the power line channel and the wireless channel in the target communication state are both the unblocked state, perform a channel transmission test on the power line channel and the wireless channel to obtain a first test result; and perform a channel switching test on the power line channel and the wireless channel to obtain a second test result;

[0143] The determining module 450 is further configured to determine a third abnormality analysis result based on the first test result and the second test result, and to determine a second target abnormality analysis result based on the first abnormality analysis result and the third abnormality analysis result.

[0144] Optionally, in the aspect of performing data preprocessing and feature extraction on the first power data to obtain the second power data, the processing module 420 specifically includes:

[0145] performing data cleaning and data standardization on the first power data to obtain reference first power data;

[0146] Performing feature extraction on the reference first power data to obtain power feature data; the power feature data includes p feature sub-data; the feature sub-data includes any one of the following: electrical quantity feature, time domain feature, frequency domain feature; p is an integer greater than 1;

[0147] Randomly divide the p feature sub-data into C random subsets, and generate C decision trees based on the C random subsets; C is a positive integer less than p;

[0148] Calculate the relative importance of each feature sub-data in the p feature sub-data according to the C decision trees to obtain p relative importances;

[0149] Sort the p relative importances in descending order to obtain a first sorting result;

[0150] The characteristic sub-data corresponding to the first M relative importances in the first sorting result are obtained to obtain the second power data; M is a positive integer less than p.

[0151] Optionally, in calculating the relative importance of each of the p feature sub-data according to the C decision trees to obtain the p relative importances, the processing module 420 specifically further includes:

[0152] The Gini index value of the reference feature sub-data at the node m of the decision tree i is calculated according to a preset Gini index calculation formula to obtain a reference Gini index value; the reference feature sub-data is any one feature sub-data among the p feature sub-data; the decision tree i is the i-th decision tree among the C decision trees; the decision tree i includes S nodes, and the node m is the m-th node among the S nodes; S is an integer greater than 1;

[0153] The Gini index calculation formula is as follows:

[0154]

[0155] Among them, G ini,i,m represents the reference Gini index value; p m,k represents the ratio between the number of messages of category k in node m of the decision tree i and the number of messages in node m, where each message corresponds to a reference feature sub-data and a message category;

[0156] Calculate the reference Gini index value according to a preset change calculation formula to obtain a reference change of the reference feature sub-data at the node m of the decision tree i;

[0157] The change calculation formula is as follows:

[0158] D vim,β,i,m =G ini,i,m -G imi,i,e -G imi,i,r

[0159] Among them, D vim,β,i,m represents the reference variation; β represents the reference feature sub-data; G imi,i,e G represents the Gini index value of the first new node after the node m of the decision tree i branches; imi,i,r represents the Gini index value of the second new node after the node m of the decision tree i branches;

[0160] Importance calculation is performed on the reference variation to obtain a reference relative importance; the reference relative importance is the relative importance corresponding to the reference feature sub-data in the p relative importances.

[0161] Optionally, in calculating the importance of the reference variation to obtain the reference relative importance, the processing module 420 further includes:

[0162] Calculating the reference change amount according to a preset first importance calculation formula to obtain a reference first importance;

[0163] The first importance calculation formula is as follows:

[0164]

[0165] Among them, D vim,β Indicates the first importance of the reference;

[0166] Calculating the reference first importance according to a preset second importance calculation formula to obtain the reference relative importance;

[0167] The second importance calculation formula is as follows:

[0168]

[0169] in, Indicates the relative importance of the reference; D vim.n Indicates the first importance of the nth reference; represents the sum of the p reference first importances.

[0170] Optionally, in performing model training according to the second power data to obtain an anomaly detection model, the training module 430 specifically includes:

[0171] Obtaining a preset first model; the first model includes a filtering network and a classification network;

[0172] Obtain x model parameter combinations corresponding to the first model; each model parameter combination includes a filtering network parameter and a classification network parameter; x is an integer greater than 1;

[0173] Configuring the first model according to the filtering network parameters and the classification network parameters of each model parameter combination in the x model parameter combinations to obtain x second models;

[0174] Performing feature extraction on the second power data through a filter network of each of the x second models to obtain x target feature data;

[0175] Performing feature classification on corresponding target feature data in the x target feature data using a classification network of each second model in the x second models to obtain x classification results;

[0176] Evaluate the x classification results according to a preset loss function to obtain x loss function values; each loss function value corresponds to a second model;

[0177] Determine a second model corresponding to the minimum loss function value among the x loss function values ​​as the anomaly detection model.

[0178] Optionally, the target channel area includes b dual-mode communication modules, where b is an integer greater than 1; the first test result includes b reference first test results; each reference first test result corresponds to a dual-mode communication module; and in performing the channel transmission test on the power line channel and the wireless channel to obtain the first test result, the testing module 460 specifically includes:

[0179] Transmitting a preset first test message to a target node through the power line channel and the wireless channel respectively through the first dual-mode communication module, so that the target node receives the first test message through the power line channel and generates a first response message and transmits it to the first dual-mode communication module, and receives the first test message through the wireless channel and generates a second response message and transmits it to the first dual-mode communication module; the first dual-mode communication module is any one of the b dual-mode communication modules;

[0180] Obtaining a sending time of the first test message, a first receiving time of the first response message, and a second receiving time of the second response message;

[0181] Determine a difference between the first receiving time and the sending time as a first transmission duration;

[0182] Determine a difference between the second receiving time and the sending time as a second transmission duration;

[0183] Comparing the first test message with the first response message and the second response message respectively to obtain a first bit error rate and a second bit error rate;

[0184] If the first transmission duration is less than a preset first transmission duration threshold, the second transmission duration is less than a preset second transmission duration threshold, and the first bit error rate and the second bit error rate are both less than a preset bit error rate threshold, then determining that the first state of the first dual-mode communication module is a normal state; otherwise, determining that the first state is an abnormal state;

[0185] The first state is determined to be a reference first test result corresponding to the first dual-mode communication module.

[0186] Optionally, the second test result includes a plurality of reference second test results; each reference second test result corresponds to a dual-mode communication module; in terms of performing a channel switching test on the power line channel and the wireless channel to obtain the second test result, the testing module 460 specifically further includes:

[0187] Determining that the first dual-mode communication module is in a first operating mode; the first operating mode is that the power line channel is turned on and the wireless channel is turned off;

[0188] transmitting a preset second test message through the power line channel by the first dual-mode communication module, and recording a third reception time of a third response message;

[0189] Performing channel switching on the first dual-mode communication module so that the first dual-mode communication module is in a second working mode; the second working mode is that the power line channel is closed and the wireless channel is opened;

[0190] Transmitting the second test message through the wireless channel by the first dual-mode communication module, and recording a fourth reception time of a fourth response message;

[0191] Obtaining a reference transmission duration of the wireless channel;

[0192] Determine a difference between the fourth receiving time and the third receiving time as a first transmission duration;

[0193] Determine a difference between the first transmission duration and the reference transmission duration as a first switching delay;

[0194] If the first switching delay is less than a preset switching delay threshold, determining that the second state of the first dual-mode communication module is a normal state; otherwise, determining that the second state is an abnormal state;

[0195] Determine that the second state is a reference second test result corresponding to the first dual-mode communication module.

[0196] It can be seen that by comprehensively considering power data and communication status, abnormal fault conditions in the dual-mode communication process can be identified quickly and accurately, thereby improving the stability and reliability of the power system.

[0197] It should be noted that the specific implementation of each operation can adopt the corresponding description of the method embodiment shown above. The dual-mode communication meter reading anomaly analysis device 400 can be used to execute the above method embodiment of the present application, which will not be described in detail.

[0198] An embodiment of the present application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any method described in the above method embodiments, and the above computer includes an electronic device.

[0199] The present application also provides a computer program product comprising a non-transitory computer-readable storage medium storing a computer program, wherein the computer program is operable to cause a computer to perform some or all of the steps of any of the methods described in the above method embodiments. The computer program product may be a software installation package, and the computer may comprise an electronic device.

[0200] It should be noted that, for the above-mentioned various embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations. Those skilled in the art should know that this application is not limited by the order of the actions described, because some steps in the embodiments of the present application can be performed in other orders or simultaneously. In addition, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions, steps, modules or units involved are not necessarily required by the embodiments of the present application.

[0201] In the above embodiments, the embodiments of the present application have different focuses on the description of each embodiment. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0202] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

[0203] The steps of the method or algorithm described in the embodiments of the present application can be implemented in hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, CD-ROMs, or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and storage medium can be located in an ASIC. In addition, the ASIC can be located in a terminal device or a management device. Of course, the processor and storage medium can also be present in a terminal device or a management device as discrete components.

[0204] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, they can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0205] The modules / units included in the devices and products described in the above embodiments may be software modules / units, hardware modules / units, or partly software modules / units and partly hardware modules / units. For example, for the devices and products applied to or integrated in the chip, the modules / units included therein may all be implemented in the form of hardware such as circuits, or at least part of the modules / units may be implemented in the form of software programs, which run on the processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for the devices and products applied to or integrated in the chip module, the modules / units included therein may all be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as chip, circuit module, etc.) or different components of the chip module, or at least part of the modules / units may be It is implemented in the form of a software program, which runs on the processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in the terminal equipment, the various modules / units contained therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or different components in the terminal equipment, or, at least some modules / units can be implemented in the form of a software program, which runs on the processor integrated inside the terminal equipment, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.

[0206] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the embodiments of the present application. It should be understood that the above description is only a specific implementation method of the embodiments of the present application and is not intended to limit the scope of protection of the embodiments of the present application. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the embodiments of the present application should be included in the scope of protection of the embodiments of the present application.

Claims

1. A dual-mode communication meter reading abnormality analysis method, characterized in that: An abnormality analysis module applied to a dual-mode communication system, the dual-mode communication system comprising the abnormality analysis module, an electric meter measurement module, and a channel area, where a is an integer greater than 1, the method comprising: Acquiring first power data of a target channel area during a preset time period and a target communication state of the dual-mode channel; the target channel area is any one of the a channel areas; the dual-mode channel includes a power line channel and a wireless channel; and the communication state includes a smooth state and a blocked state; performing data preprocessing and feature extraction on the first power data to obtain second power data; Performing model training based on the second power data to obtain an anomaly detection model; Acquire meter reading information of the electric meter metering module in the preset time period corresponding to the target channel area; Analyzing the meter reading information according to the anomaly detection model to obtain a first anomaly analysis result; If the communication state of the power line channel and / or the wireless channel in the target communication state is the blocked state, determining the target communication state as a second abnormality analysis result; determining a first target abnormality analysis result according to the first abnormality analysis result and the second abnormality analysis result; If the communication states of the power line channel and the wireless channel in the target communication state are both the unblocked state, performing a channel transmission test on the power line channel and the wireless channel to obtain a first test result; and performing a channel switching test on the power line channel and the wireless channel to obtain a second test result; A third abnormality analysis result is determined according to the first test result and the second test result, and a second target abnormality analysis result is determined according to the first abnormality analysis result and the third abnormality analysis result.

2. The method according to claim 1, wherein The performing data preprocessing and feature extraction on the first power data to obtain second power data includes: performing data cleaning and data standardization on the first power data to obtain reference first power data; Performing feature extraction on the reference first power data to obtain power feature data; the power feature data includes p feature sub-data; the feature sub-data includes any one of the following: electrical quantity feature, time domain feature, frequency domain feature; p is an integer greater than 1; Randomly divide the p feature sub-data into C random subsets, and generate C decision trees based on the C random subsets; C is a positive integer less than p; Calculate the relative importance of each feature sub-data in the p feature sub-data according to the C decision trees to obtain p relative importances; Sort the p relative importances in descending order to obtain a first sorting result; The characteristic sub-data corresponding to the first M relative importances in the first sorting result are obtained to obtain the second power data; M is a positive integer less than p.

3. The method according to claim 2, wherein The step of calculating the relative importance of each of the p feature sub-data according to the C decision trees to obtain the p relative importances includes: The Gini index value of the reference feature sub-data at the node m of the decision tree i is calculated according to a preset Gini index calculation formula to obtain a reference Gini index value; the reference feature sub-data is any one feature sub-data among the p feature sub-data; the decision tree i is the i-th decision tree among the C decision trees; the decision tree i includes S nodes, and the node m is the m-th node among the S nodes; S is an integer greater than 1; The Gini index calculation formula is as follows: Among them, G ini,i,m represents the reference Gini index value; p m,k represents the ratio between the number of messages of category k in node m of the decision tree i and the number of messages in node m, where each message corresponds to a reference feature sub-data and a message category; Calculate the reference Gini index value according to a preset change calculation formula to obtain a reference change of the reference feature sub-data at the node m of the decision tree i; The change calculation formula is as follows: D vim,β,i,m =G ini,i,m -G imi,i,e -G imi,i,r Among them, D vim,β,i,m represents the reference variation; β represents the reference feature sub-data; G imi,i,e G represents the Gini index value of the first new node after the node m of the decision tree i branches; imi,i,r represents the Gini index value of the second new node after the node m of the decision tree i branches; Importance calculation is performed on the reference variation to obtain a reference relative importance; the reference relative importance is the relative importance corresponding to the reference feature sub-data in the p relative importances.

4. The method according to claim 3, wherein The calculating the importance of the reference variation to obtain the reference relative importance includes: Calculating the reference change amount according to a preset first importance calculation formula to obtain a reference first importance; The first importance calculation formula is as follows: Among them, D vim,β Indicates the first importance of the reference; Calculating the reference first importance according to a preset second importance calculation formula to obtain the reference relative importance; The second importance calculation formula is as follows: in, Indicates the relative importance of the reference; D vim.n Indicates the first importance of the nth reference; represents the sum of the p reference first importances.

5. The method according to any one of claims 1 to 4, characterized in that The performing model training according to the second power data to obtain an anomaly detection model includes: Obtaining a preset first model; the first model includes a filtering network and a classification network; Obtain x model parameter combinations corresponding to the first model; each model parameter combination includes a filtering network parameter and a classification network parameter; x is an integer greater than 1; Configuring the first model according to the filtering network parameters and the classification network parameters of each model parameter combination in the x model parameter combinations to obtain x second models; Performing feature extraction on the second power data through a filter network of each of the x second models to obtain x target feature data; Performing feature classification on corresponding target feature data in the x target feature data using a classification network of each second model in the x second models to obtain x classification results; Evaluate the x classification results according to a preset loss function to obtain x loss function values; each loss function value corresponds to a second model; Determine a second model corresponding to the minimum loss function value among the x loss function values ​​as the anomaly detection model.

6. The method according to claim 1, wherein The target channel area includes b dual-mode communication modules, where b is an integer greater than 1; the first test result includes b reference first test results; Each reference first test result corresponds to a dual-mode communication module; The performing a channel transmission test on the power line channel and the wireless channel to obtain a first test result includes: Transmitting a preset first test message to a target node through the power line channel and the wireless channel respectively through the first dual-mode communication module, so that the target node receives the first test message through the power line channel and generates a first response message and transmits it to the first dual-mode communication module, and receives the first test message through the wireless channel and generates a second response message and transmits it to the first dual-mode communication module; the first dual-mode communication module is any one of the b dual-mode communication modules; Obtaining a sending time of the first test message, a first receiving time of the first response message, and a second receiving time of the second response message; Determine a difference between the first receiving time and the sending time as a first transmission duration; Determine a difference between the second receiving time and the sending time as a second transmission duration; Comparing the first test message with the first response message and the second response message respectively to obtain a first bit error rate and a second bit error rate; If the first transmission duration is less than a preset first transmission duration threshold, the second transmission duration is less than a preset second transmission duration threshold, and the first bit error rate and the second bit error rate are both less than a preset bit error rate threshold, then determining that the first state of the first dual-mode communication module is a normal state; otherwise, determining that the first state is an abnormal state; The first state is determined to be a reference first test result corresponding to the first dual-mode communication module.

7. The method according to claim 6, wherein The second test result includes a plurality of reference second test results; Each reference second test result corresponds to a dual-mode communication module; The performing a channel switching test on the power line channel and the wireless channel to obtain a second test result includes: Determining that the first dual-mode communication module is in a first operating mode; the first operating mode is that the power line channel is turned on and the wireless channel is turned off; transmitting a preset second test message through the power line channel by the first dual-mode communication module, and recording a third reception time of a third response message; Performing channel switching on the first dual-mode communication module so that the first dual-mode communication module is in a second working mode; the second working mode is that the power line channel is closed and the wireless channel is opened; Transmitting the second test message through the wireless channel by the first dual-mode communication module, and recording a fourth reception time of a fourth response message; Obtaining a reference transmission duration of the wireless channel; Determine a difference between the fourth receiving time and the third receiving time as a first transmission duration; Determine a difference between the first transmission duration and the reference transmission duration as a first switching delay; If the first switching delay is less than a preset switching delay threshold, determining that the second state of the first dual-mode communication module is a normal state; otherwise, determining that the second state is an abnormal state; Determine that the second state is a reference second test result corresponding to the first dual-mode communication module.

8. A dual-mode communication meter reading abnormality analysis device, characterized in that: An abnormality analysis module applied to a dual-mode communication system, the dual-mode communication system comprising the abnormality analysis module, an electric meter measurement module, and a channel area, where a is an integer greater than 1, the device comprising an acquisition module, a processing module, a training module, an analysis module, a determination module, and a testing module, wherein: The acquisition module is configured to acquire first power data of a target channel area during a preset time period and a target communication state of the dual-mode channel; the target channel area is any one of the a channel areas; the dual-mode channel includes a power line channel and a wireless channel; and the communication state includes a smooth state and a blocked state. The processing module is configured to perform data preprocessing and feature extraction on the first power data to obtain second power data; The training module is configured to perform model training based on the second power data to obtain an anomaly detection model; The acquisition module is further configured to acquire meter reading information of the electric meter measurement module in the preset time period corresponding to the target channel area; The analysis module is configured to analyze the meter reading information according to the anomaly detection model to obtain a first anomaly analysis result; The determining module is configured to determine that the target communication state is a second abnormality analysis result if the communication state of the power line channel and / or the wireless channel in the target communication state is the blocked state; and determine a first target abnormality analysis result based on the first abnormality analysis result and the second abnormality analysis result; The testing module is configured to, if the communication states of the power line channel and the wireless channel in the target communication state are both in the unblocked state, perform a channel transmission test on the power line channel and the wireless channel to obtain a first test result; and perform a channel switching test on the power line channel and the wireless channel to obtain a second test result; The determination module is further configured to determine a third abnormality analysis result based on the first test result and the second test result, and to determine a second target abnormality analysis result based on the first abnormality analysis result and the third abnormality analysis result.

9. An electronic device, characterized in that: include: a processor, a memory, a communication interface, and one or more programs; The one or more programs are stored in the memory and configured to be executed by the processor, wherein the programs include instructions for executing the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 7.

Images