Synchronous acquisition method for internal temperature of user electric meter in low-voltage transformer area

Abstract

The invention provides a synchronous acquisition method for the internal temperature of a user ammeter in a low-voltage transformer area, and belongs to the technical field of electric meter reading. Comprising the following steps: identifying an ammeter communication protocol and address when an ammeter module is initialized, and detecting whether the ammeter supports a temperature detection function; the terminal establishes and maintains an ammeter information database according to a detection result of the ammeter module, and records an ammeter protocol type, a communication address and a temperature measurement function state; the terminal broadcasts a freezing instruction and specifies the ammeter to freeze electric energy data at a target moment; the ammeter module analyzes the freezing instruction to obtain a target moment; for the ammeter supporting temperature measurement, an ammeter module reads the internal temperature of the ammeter at a target moment and returns the internal temperature when a subsequent terminal requests; for an electricity meter which does not support temperature measurement, an electricity meter module collects electric energy in a preset time period before a target moment and time sequence data of the temperature of the electricity meter module, and returns the time sequence data when a subsequent terminal requests; and the terminal predicts the internal temperature of the electric meter at the target moment according to the corresponding time sequence data before the target moment of the electric meter.

Description

Technical Field

[0001] This invention belongs to the field of power meter reading technology, specifically relating to a method for synchronously acquiring the internal temperature of a user's electricity meter in a low-voltage distribution area. Background Technology

[0002] In low-voltage power distribution network management and status analysis, temperature monitoring of user meters is a crucial step. Currently, some smart meters have built-in temperature sensors that can directly read their internal temperature values ​​via communication protocols. However, a large number of older smart meters still exist in the power grid. Although these meters' communication protocols comply with standard specifications, some models lack built-in temperature detection circuitry, preventing them from detecting the internal temperature. It should be noted that while the 2007 version of DL / T 645 "Communication Protocol for Multifunctional Energy Meters" added an "internal temperature" data identifier (distinct from the 1997 version), temperature detection is not mandatory, leading to some meters not having this function enabled during production.

[0003] Within a low-voltage distribution area, old and new electricity meters are often mixed together, and some operating meters do not have built-in temperature measurement functions. Therefore, relying solely on direct readings cannot obtain the temperature of all meters in the low-voltage distribution area, creating a monitoring blind spot. How to obtain the internal temperature of all meters in the low-voltage distribution area at the same time is a technical problem that needs to be solved. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a method for synchronously acquiring the internal temperature of electricity meters in low-voltage distribution areas. This method can efficiently collect basic electrical energy data of the electricity meter without modifying the current meter hardware or firmware, and can also collect extended data such as temperature data and harmonic data.

[0005] To solve one or more or all of the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for synchronously acquiring the internal temperature of a user's electricity meter in a low-voltage distribution area includes: during meter module initialization, identifying the meter's communication protocol and address, and detecting whether the meter supports temperature detection; the terminal establishing and maintaining a meter information database based on the meter module's detection results, recording the meter's protocol type, communication address, and temperature measurement function status; the terminal broadcasting a freeze command matching the protocol type in the database, specifying that the meter freezes its energy data at a target time; the meter module parsing the freeze command to obtain the target time; for meters that support temperature measurement, the meter module reading the meter's internal temperature at the target time and transmitting it back upon subsequent terminal requests; for meters that do not support temperature measurement, the meter module collecting time-series data of energy and its own temperature for a preset period before the target time and transmitting it back upon subsequent terminal requests; and the terminal predicting the meter's internal temperature at the target time based on the time-series data corresponding to the meter before the target time.

[0006] Furthermore, the method for detecting whether the electricity meter supports the temperature detection function includes: attempting to read the data identifier corresponding to the communication protocol of the electricity meter. If the reading fails after multiple attempts or invalid data is read, it is determined that the electricity meter does not support the temperature measurement function.

[0007] Furthermore, the method for the terminal to broadcast a freeze command matching the protocol type according to the protocol type in the database includes: the terminal determining the protocol type supported by the electricity meter according to the electricity meter information database, constructing corresponding freeze commands according to different protocols, and broadcasting freeze commands of different protocols respectively.

[0008] Furthermore, the method for the meter module to collect time-series data of electrical energy and its own temperature for a preset period before the target time includes: setting the preset duration as x and the target time as T, then collecting and recording the electrical energy data of the meter and the temperature data of the meter module itself at a preset frequency starting from time Tx, until the target time T.

[0009] Furthermore, the preset frequency is: sampling once every 1-3 seconds during the period from T-1 minute to T, sampling once every 10-30 seconds during the period from T-5 minutes to T-1 minute, and sampling once 30-60 seconds before T-5 minutes.

[0010] Furthermore, the method for the terminal to predict the internal temperature of the electricity meter at the target time based on the time series data corresponding to the target time includes: constructing a prediction model, inputting the time series data into the prediction model, and predicting the internal temperature of the electricity meter at the target time.

[0011] Furthermore, the method for constructing the prediction model includes: obtaining training samples from electricity meters that support temperature measurement, each sample containing electricity time-series data and meter module temperature data for a preset period before the target time, and the actual temperature value of the meter at the target time; extracting key features from the electricity time-series data and meter module temperature data, and combining the key features with the time-series data into a feature vector; using the sample dataset, with the feature vector as input and the actual temperature value of the meter as output, training the prediction model.

[0012] Furthermore, it also includes: after the target time, the terminal sends a command to the electricity meter to obtain the frozen electricity data.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention solves the problem of asynchronous acquisition of internal temperature data from electricity meters within a given area. It broadcasts a freeze command to the meter and meter module, broadcasting the target time for acquiring temperature data. For meters supporting temperature measurement, the meter module directly reads the internal temperature data at the target time. For meters without temperature measurement, the meter module collects power consumption time-series data and meter module temperature time-series data for a preset period before the target time. Based on the collected time-series data, it predicts internal meter problems, ensuring accurate temperature data alignment. When collecting time-series data, the meter module employs a stepped dynamic sampling strategy, increasing the sampling frequency as the target time approaches to better capture thermal inertia characteristics. Detailed Implementation

[0014] To better understand the present invention, the following embodiments further illustrate the content of the invention, but the scope of protection of the present invention is not limited to the following embodiments. Numerous specific details are set forth in the following description to provide a more thorough understanding of the invention. However, it will be apparent to those skilled in the art that the present invention can be practiced without one or more of these details.

[0015] Example 1: The purpose of this example is to provide a method for synchronously acquiring the internal temperature of user meters in low-voltage distribution areas. It includes: S1. During the initialization of the meter module, the meter communication protocol and address are identified, and it is checked whether the meter supports temperature detection function.

[0016] The meter module is installed inside the meter's module compartment, directly connected to the meter body via a serial port, and then connected to a terminal (such as a concentrator) via a communication link, enabling transparent data transmission and network communication between the terminal and the meter. Currently, the most commonly used meter modules are carrier modules that support carrier communication.

[0017] Currently available electricity meter modules all comply with national and industry standards and specifications, including power industry standards (such as DL / T 1487, 1488, 1489, 645, 698.45) and State Grid Corporation of China enterprise standards (Q / GDW 1376), and generally have temperature detection capabilities.

[0018] During power-on initialization, the meter module attempts to send an address request frame to the meter's MCU using a built-in protocol (such as DL / T645 or DL / T698.45). The communication protocol and address are determined based on the meter's response. Subsequently, the meter module attempts to read the internal temperature according to the determined protocol. For example, the DL / T645 protocol uses data identifier 02800007, and the DL / T698 protocol uses data identifier 20100200. If the reading fails or invalid data is read, it is determined that the meter does not support temperature measurement.

[0019] Finally, the meter module sends the meter's protocol type, communication address, and temperature measurement function support information to the terminal.

[0020] S2. The terminal establishes and maintains a meter information database based on the detection results of the meter module, and records the meter protocol type, communication address and temperature measurement function status.

[0021] The terminal receives meter information (including meter number, protocol type, communication address, and temperature measurement function status) sent by the meter module and adds the meter information to the meter information database.

[0022] S3. The terminal broadcasts a freeze command matching the protocol according to the protocol type in the database, specifying that the electricity meter freezes the electricity data at the target time.

[0023] The terminal checks the protocol types supported by its subordinate meters based on the meter information database, and constructs corresponding freeze commands according to different protocols. The freeze command specifies the target time for freezing data on the meter. Subsequently, the freeze commands for different protocols are broadcast separately to ensure that the meter can freeze data at the target time.

[0024] It should be noted that the terminal also needs to broadcast time synchronization instructions periodically or irregularly, or broadcast time synchronization instructions before broadcasting freeze instructions, to ensure that the time of the terminal, the electricity meter and the electricity meter module are synchronized.

[0025] In addition, when freezing data on an electricity meter, only the electrical energy data can be frozen. The temperature data inside the meter, being real-time data, cannot be frozen. The temperature inside the meter at the target time needs to be actively read at the target time.

[0026] S4. The meter module parses the freeze command to obtain the target time.

[0027] After receiving the freeze command sent by the terminal, the meter module not only forwards the freeze command to the meter, but also parses the freeze command to obtain the specified target time.

[0028] S5. For electricity meters that support temperature measurement, the meter module reads the internal temperature of the meter at the target time and transmits it back when requested by the terminal. For electricity meters that do not support temperature measurement, the meter module collects the time-series data of the power consumption and its own temperature for a preset period before the target time and transmits it back when requested by the terminal.

[0029] For electricity meters that support temperature measurement, the meter module is set to perform a timed task to directly read the meter temperature (internal temperature) at a target time. After the target time, the terminal will send a request command to the meter module, and the meter module will return the meter temperature data.

[0030] For electricity meters that do not support temperature measurement, a timed task is set for the meter module. For example, if the target time is T, the task is set to start collecting and recording the meter's energy data and the meter module's own temperature data at a preset frequency from time Tx until the target time T. After the target time, the terminal will send a request command to the meter module, and the meter module will return the recorded time-series data.

[0031] For example, the electrical energy data collected by the meter module includes current, voltage, active power, reactive power, apparent power, and power factor.

[0032] Preferably, x is 15 minutes. Furthermore, the sampling frequency of the meter module is dynamically adjusted before the target time, using a stepped sampling frequency. The sampling frequency is increased as the target time approaches. For example, sampling once every 1-3 seconds from T-1 minute to T, once every 10-30 seconds from T-5 minutes to T-1 minute, and once every 30-60 seconds before T-5 minutes.

[0033] S6. For meters that do not support temperature measurement, the terminal predicts the internal temperature of the meter at the target time based on the time sequence data corresponding to the target time before the target time.

[0034] After acquiring the time-series data of electricity and temperature collected by the meter module, the terminal inputs the time-series data into a pre-built prediction model to predict the internal temperature of the meter at the target time.

[0035] Methods for constructing predictive models include: Collect sample dataset: Obtain training samples from electricity meters that support temperature measurement function. Each sample contains the time series data of electricity and the temperature data of the meter module for a preset period before the target time, and the actual temperature value of the meter at the target time.

[0036] Feature engineering: Key features are extracted from electricity time-series data and meter module temperature data, such as data change trends, current imbalance (three-phase imbalance can lead to local overheating), and power change gradient (sudden load changes can cause instantaneous temperature rise). These key features are then combined with the time-series data to form a time-series feature vector.

[0037] Model training: Using the sample dataset, with the feature vector as input and the actual temperature value of the electricity meter as output, a prediction model is trained.

[0038] The prediction model can adopt a conventional time-series model structure. Taking a 1D convolutional neural network (1D-CNN) as an example, the network structure of the prediction model includes: The input layer is used to input feature vectors. Each dimension of the vector corresponds to a class of sample data, and all features need to be sampled synchronously at the same time step. The shape of the input tensor is [1, L, C], where L is the length of the feature vector and C is the number of feature channels, with a value of 4-6.

[0039] The first convolutional kernel is a 1D convolutional kernel with a kernel size of 3-5 and a number of 32 kernels. The stride is 1, and the padding method is "same" to keep the temporal dimension unchanged. The ReLU activation function is used. The pooling layer has a pooling window size of 2 and a stride of 2 to reduce the temporal resolution and extract significant local features.

[0040] The second convolutional kernel, apart from increasing the number of kernels (e.g., to 64), has the same structure as the first convolutional kernel.

[0041] The flattening layer rearranges the three-dimensional tensor after multiple convolutions and pooling into a one-dimensional vector, providing input for subsequent fully connected layers.

[0042] The first fully connected layer has 128 neurons and uses ReLU activation to further fuse global features.

[0043] The second fully connected layer has 64 neurons and also uses ReLU activation to further enhance the model's expressive power.

[0044] The output layer uses a single linear neuron to directly output a scalar, namely the predicted internal temperature of the meter.

[0045] Meanwhile, to achieve the regression task, mean squared error (MSE) or root mean squared error (RMSE) is used as the loss function during the training phase.

[0046] It should be noted that during the execution of this method, the electricity meters are frozen with the electricity data at the target time. The sampling time of the frozen electricity data is consistent with the time when the internal temperature of the electricity meter is obtained by this method. Therefore, the terminal can also collect the frozen electricity data of each electricity meter as a reference for the analysis of the distribution network or electricity meter data. Therefore, optionally, this method also includes: S7, after the target time, the terminal sends a command to the electricity meter to obtain the electricity data frozen at the target time, and stores the obtained electricity data after associating it with the active internal temperature of the electricity meter.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solutions of the present invention, as long as they do not depart from the spirit and scope of the technical solutions of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A method for synchronously acquiring the internal temperature of a low-voltage area user electric meter, characterized in that, The application relates to a method for collecting and analyzing power consumption data of a power meter. The method comprises the following steps: a power meter module identifies a communication protocol and an address of the power meter during initialization, and detects whether the power meter supports a temperature detection function; a terminal establishes and maintains a power meter information database according to the detection result of the power meter module, and records a power meter protocol type, a communication address and a temperature detection function state; the terminal broadcasts a frozen instruction matching the protocol type in the database to specify that the power meter freezes power consumption data at a target time; the power meter module parses the frozen instruction to obtain the target time; for the power meter supporting temperature detection, the power meter module reads the internal temperature of the power meter at the target time, and returns the internal temperature to the terminal upon request; 2. The method for synchronously acquiring the internal temperature of the low-voltage area user electric meter according to claim 1, characterized in that, for the power meter not supporting temperature detection, the power meter module collects time sequence data of power consumption and the temperature of the power meter in a preset time period before the target time, and returns the time sequence data to the terminal upon request; and the terminal predicts the internal temperature of the power meter at the target time according to the time sequence data corresponding to the target time of the power meter.

3. The method for synchronously acquiring the internal temperature of the low-voltage area user electric meter according to claim 1, characterized in that, The method for detecting whether the power meter supports the temperature detection function comprises the following steps: reading data with a data identifier corresponding to the communication protocol of the power meter, and determining that the power meter does not support the temperature detection function if reading fails or invalid data is read.

4. The method for synchronously acquiring the internal temperature of the low-voltage area user electric meter according to claim 1, characterized in that, The method for broadcasting the frozen instruction matching the protocol type in the database by the terminal comprises the following steps: determining the protocol type supported by the power meter according to the power meter information database, constructing corresponding frozen instructions according to different protocols, and broadcasting the frozen instructions of different protocols respectively.

5. The method for synchronously acquiring the internal temperature of the low-voltage area user electric meter according to claim 4, characterized in that, The method for collecting the time sequence data of power consumption and the temperature of the power meter module in the preset time period before the target time comprises the following steps: setting a preset time length as x and a target time as T, collecting and recording the power consumption data of the power meter and the temperature data of the power meter module at a preset frequency from the time T-x to the target time T.

6. The method for synchronously acquiring the internal temperature of the low-voltage area user electric meter according to claim 1, characterized in that, The preset frequency is: sampling once every 1-3 seconds during T-1 minute-T, sampling once every 10-30 seconds during T-5 minutes-T-1 minute, and sampling once every 30-60 seconds before T-5 minutes.

7. The method for synchronously acquiring the internal temperature of the low-voltage area user electric meter according to claim 6, characterized in that, The method for predicting the internal temperature of the power meter at the target time according to the time sequence data corresponding to the target time of the power meter comprises the following steps: constructing a prediction model, inputting the time sequence data into the prediction model, and predicting the internal temperature of the power meter at the target time. The method for constructing the prediction model comprises the following steps: obtaining training samples from the power meter supporting the temperature detection function, wherein each sample comprises time sequence data of power consumption and temperature data of the power meter module in a preset time period before a target time, and an actual temperature value of the power meter at the target time; extracting key features from the time sequence data and the temperature data of the power meter module, combining the key features and the time sequence data into a feature vector; 8. The method for synchronously acquiring the internal temperature of a low-voltage area user electric meter according to claim 1, characterized in that, using a sample data set, taking the feature vector as input and taking the actual temperature value of the power meter as output, and training to obtain the prediction model. The application further comprises the following steps: after the target time, the terminal sends an instruction to the power meter to obtain the power consumption data frozen at the target time.