A low-power management method and system for electricity meters

Abstract

This invention belongs to the field of electricity metering technology and discloses a low-power management method and system for electricity meters. The method includes: collecting real-time load data; acquiring historical load data and calculating average load based on the real-time load; dynamically adjusting the data frequency based on the average load; dynamically adjusting the transmission strategy according to the historical load and data frequency; acquiring power data and compressing the power data; and transmitting the compressed power data based on the transmission strategy. This invention can effectively adapt to real-time load changes, optimize data transmission time, and reduce transmission load, thereby reducing the power consumption of the electricity meter, improving operating efficiency and service life, and meeting the needs of smart grids for sustainable development and high performance.

Description

Technical Field

[0001] This invention relates to the field of electricity metering technology, and more specifically, to a low-power management method and system for electricity meters. Background Technology

[0002] With the development of smart grids, the functions of electricity meters have expanded from traditional metering to areas such as intelligent monitoring, data acquisition, and communication transmission. However, electricity meters face challenges in practical applications, such as high power consumption, long-term operation requirements, and sustainable development pressures. In order to improve the reliability of electricity meters, extend their service life, and respond to the needs of green development, a low-power management method is urgently needed to achieve a balance between high performance and low energy consumption, thereby improving the operating efficiency and economy of smart grids.

[0003] Patent CN112098713B discloses a method for controlling the operation of an electricity meter, including: 1) starting the electricity meter; 2) determining whether it is powered on: after the electricity meter starts, the MCU of the management chip determines whether a power failure signal is received. If yes, proceed to step 3); if no, proceed to step 4); 3) the electricity meter enters a low-power state. If a wake-up signal is received, proceed to step 5); if no wake-up signal is received, return to step 2); 4) the MCU of the management chip runs an operating system version and returns to step 2 after running for a certain period of time; 5) the MCU of the management chip runs a bare-metal version and returns to step 2 after running for a certain period of time. This invention can run dual systems through the same MCU core, can run at low speed when power is lost, reduces the operating current under backup power supply, and meets the requirements of low-power operation.

[0004] However, while the above technologies can achieve low-power management of electricity meters, they mainly rely on power failure information monitoring and low-power state switching. That is, they rely too much on fixed logic processes for low-power management, lacking dynamic and flexible fine-grained management. They fail to adaptively optimize and adjust data acquisition, processing and transmission based on actual operating conditions, making it difficult to achieve accurate adaptation to complex scenarios and dynamic loads.

[0005] In view of this, the present invention proposes a low-power management method and system for electricity meters to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides the following technical solution: a low-power management method for an electricity meter, comprising:

[0007] Collect real-time load;

[0008] Obtain historical load data and combine it with real-time load data to calculate the average load.

[0009] The data frequency is dynamically adjusted based on the average load.

[0010] The transmission strategy is dynamically adjusted based on historical load and data frequency.

[0011] Acquire power data, compress the power data, and transmit the compressed power data based on the transmission strategy.

[0012] Furthermore, the real-time load is the load power collected in real time, and the load power is the rate at which the load equipment in the power system consumes electrical energy; the method for collecting the real-time load is: acquiring real-time voltage, real-time current, and power factor, and calculating the real-time load; the expression for the real-time load is: In the formula, For real-time load, For real-time voltage, For real-time current, Power factor;

[0013] The historical load refers to the load power collected at historical moments;

[0014] The method for calculating the average load includes:

[0015] Set the selected duration and obtain the current time, which corresponds to the real-time load. Subtract the selected duration from the current time to obtain the selected time. Based on the selected time and the current time, construct the calculation period, the range of which is... ,in To select a time, Given the current time; obtain the load power corresponding to the calculation period in the historical load and mark it as the calculation load; count the number of calculation loads and increment it by one, marking it as the load quantity; add each calculation load and the real-time load in sequence, and then divide by the load quantity to obtain the average load.

[0016] Furthermore, the step of setting the selected duration includes:

[0017] Step A1: Preset the selected duration range Segmentation ratio and number of iterations ;

[0018] Step A2: Based on the segmentation ratio From the selected time range Two new points are divided in the middle. and , , ;

[0019] Step A3: Based on the points respectively and And at the current moment, select the corresponding computing load from the historical load; based on The selected computational loads are used as the first set. The standard deviation corresponding to the first set is calculated and marked as the first standard deviation; then, based on... The selected computational loads are used as the second set. The standard deviation of the second set is calculated and marked as the second standard deviation.

[0020] Step A4: Using the first standard deviation as the first data point, The first data and the second standard deviation are used as the second data; the first data and the second data are respectively input into the trained matching analysis model to predict the corresponding matching degree; the matching degree corresponding to the first data is marked as the first matching degree, and the matching degree corresponding to the second data is marked as the second matching degree; the matching analysis model is a deep learning network model;

[0021] Step A5: If the first matching degree is greater than or equal to the second matching degree, the minimum value of the updated selection duration range is the minimum value of the selection duration range before the update, and the maximum value is the point corresponding to the second matching degree; if the first matching degree is less than the second matching degree, the minimum value of the updated selection duration range is the point corresponding to the first matching degree, and the maximum value is the maximum value of the selection duration range before the update.

[0022] Step A6: Set a preset width threshold, subtract the minimum value from the maximum value in the updated selected duration range to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, then adjust the updated selected duration range according to the segmentation ratio. Two new points were redefined. and Then return to step A4; if the interval width is less than the width threshold, then the average of the maximum and minimum values ​​in the updated selection duration range will be used as the selection duration.

[0023] Furthermore, the data frequency includes sampling frequency, processing frequency, and transmission frequency;

[0024] Methods for dynamically adjusting data frequency include:

[0025] Define the search space radius threshold and iteration threshold Search space This includes n frequency sets, where n is an integer greater than 1. The method for constructing the n frequency sets is as follows: Obtain the frequency range, which includes the sampling frequency range, the processing frequency range, and the transmission frequency range. Randomly select a value from each range within the frequency range to construct a frequency set, resulting in a total of n distinct frequency sets. Assign sequentially increasing numerical labels to the n frequency sets, marking them as set labels. The range of set labels is... ;

[0026] Set the initial aperture center. and initial aperture radius And set the number of iterations. ;in, , Determine the superiority function and the iterative process; the iterative process is as follows: generate m candidate solutions within the aperture range, and reduce the aperture radius. , Calculate the merit of each candidate solution, and move the aperture center to the candidate solution with the highest merit; repeat the iterative process, and for each repetition, set the iteration count. Until or When the time is reached, stop the repeated loop process, obtain the candidate solution corresponding to the aperture center, and mark it as the best solution; dynamically adjust the data frequency according to the frequency set corresponding to the set label of the best solution.

[0027] Furthermore, the expression for the superiority function is: In the formula, For excellence, The fitness level is determined by the following method: based on the set labels corresponding to the candidate solutions, the data frequencies in the corresponding frequency sets are obtained and marked as candidate frequencies; the candidate frequencies and average load are used as evaluation data; the evaluation data are input into the trained fitness analysis model to predict the corresponding fitness level; the fitness analysis model is a deep learning network model.

[0028] The expression for the candidate solution is: In the formula, For the i-th candidate solution, The aperture center corresponds to the t-th repetition cycle. Let f be the aperture radius corresponding to the t-th repetition cycle. For random coefficients, , ;

[0029] The expression for reducing the aperture radius is: In the formula, This is the reduced aperture radius. The shrinkage coefficient, .

[0030] Furthermore, the dynamic adjustment transmission strategy includes determining the transmission period and evaluating the transmission frequency;

[0031] The method for determining the transmission time period includes:

[0032] Based on historical load, predict the load power corresponding to future times and mark them as future loads. Future times include all times from the current time to the end of the day, and the load power in the future loads corresponds one-to-one with the times in the future times; preset low load periods. and medium load section According to the low load segment and medium load section Determine the power of each load in the future load, if Then the corresponding future time will be marked as a low-load time. For the load power in the future load; if If so, the corresponding future time will be marked as a medium load time; if If so, the corresponding future time will not be marked;

[0033] Sort future times from earliest to latest to generate a time sorting table; mark consecutive low-load times in the time sorting table as low-load periods, and consecutive medium-load times in the time sorting table as medium-load periods; mark the latest low-load time in each low-load period as the latest time, and compare each latest time; mark the low-load period corresponding to the latest latest time as the transmission period; if there is no low-load period in the time sorting table, mark the latest medium-load time in each medium-load time as the latest time, and mark the medium-load period corresponding to the latest latest time as the transmission period.

[0034] Furthermore, the method for predicting the load power at future times includes:

[0035] Historical and real-time loads are used as prediction data. This prediction data is then input into a trained load prediction model to predict the load power at future times. The load prediction model is an RNN neural network model, and its training process includes:

[0036] Pre-continuous collection Each load power is assigned, and a load training set is constructed. , To predict the amount of load power in the data; based on the load training set, train a load prediction model to predict the load power at future times;

[0037] The system presets the sliding step size L and the sliding window length E; it transforms the load power in the load training set into multiple training samples using the sliding window method, with each training sample containing E load powers; it uses each training sample as input to the load prediction model, predicts the load power after the sliding step size L as output, and uses the actual L load powers corresponding to each training sample as prediction targets. It evaluates the model accuracy using the mean absolute percentage error (MAPE) of the prediction results. When the calculated MAPE is less than the preset MAPE, the load prediction model corresponding to the load power is considered trained successfully; and it generates a load prediction model that predicts the load power at future times based on the load power.

[0038] Furthermore, the method for evaluating the transmission frequency includes:

[0039] Mark the latest time corresponding to the transmission period as the end time, and mark all times from the current time to the end time as the analysis period; calculate the average load corresponding to each time in the analysis period based on the load power and historical load; calculate the transmission frequency corresponding to each time in the analysis period based on the average load; obtain the historical end time, which is the time corresponding to the last time the energy meter ended data transmission; construct an estimated period based on the end time and the historical end time, with the range of the estimated period being... ,in For the end of history, The end time is determined by summing the transmission frequencies corresponding to each time point in the estimated time period to obtain the total transmission frequency. The number of time points within the transmission time period is counted, and the total transmission frequency is divided by the number of time points to obtain the transmission frequency corresponding to each time point within the transmission time period.

[0040] Furthermore, the electricity data refers to all data collected by the electricity meter within the estimated time period;

[0041] The steps for compressing the power data include:

[0042] Step B1: Count the number of each value in the power data and use it as the frequency of occurrence; treat each value as a node and use the frequency of occurrence of each value as the node frequency of the corresponding node; sort all nodes according to the node frequency from small to large to generate a node sorting table.

[0043] Step B2: According to the ascending order of the node sorting table, obtain the two nodes at the very beginning and mark them as child nodes, then delete the child nodes from the node sorting table;

[0044] Step B3: Merge the two child nodes into one parent node and add it to the node sorting table. The node frequency of the parent node is the sum of the node frequencies of the two child nodes.

[0045] Step B4: Repeat steps B2 to B3 until only one node remains in the node sorting table. The loop ends, and the remaining node in the node sorting is taken as the root node. The root node is taken as the starting point and extended downwards. The two child nodes corresponding to each node are obtained in turn until the child nodes corresponding to all nodes are obtained, forming a node tree, and then proceeding to step B5.

[0046] Step B5: Starting from the root node of the node tree, follow the path of the node tree, and when a node is encountered on the left, record its code. When a node is located on the right, record the encoding. ,coding With encoding It is a binary value; until the path of the node tree ends, the recorded codes are combined according to the order of the codes to form the compressed code of the value corresponding to the last node.

[0047] Step B6: Repeat step B5 until the compressed code corresponding to each value in the power data is obtained, and the node tree is used as compressed data; wherein, the order of each compressed code in the compressed data is consistent with the order of each value in the power data.

[0048] A low-power management system for an electricity meter, comprising implementing the aforementioned low-power management method for an electricity meter, including:

[0049] The load acquisition module is used to acquire real-time load data.

[0050] The load calculation module is used to obtain historical load and combine it with real-time load to calculate the average load;

[0051] The frequency adjustment module dynamically adjusts the data frequency based on the average load.

[0052] The strategy adjustment module dynamically adjusts the transmission strategy based on historical load and data frequency.

[0053] The data compression module is used to acquire power data, compress the power data, and transmit the compressed power data based on the transmission strategy.

[0054] The technical effects and advantages of the low-power management method and system for electricity meters of the present invention are as follows:

[0055] By collecting and analyzing real-time and historical load data, the average load is accurately calculated. Based on the average load, a heuristic algorithm is used to dynamically adjust the data frequency. Combining load forecasting and data compression technologies, the transmission strategy is dynamically adjusted to transmit compressed data, enabling refined management and control of the electricity meter. This effectively adapts to real-time load changes, optimizes data transmission time, and reduces transmission load, thereby reducing the power consumption of the electricity meter, improving operating efficiency and service life, meeting the needs of smart grids for sustainable development and high performance, and providing an innovative solution for balancing low power consumption and high performance in electricity meters. Attached Figure Description

[0056] Figure 1 This is a schematic diagram of a low-power management system for an energy meter according to Embodiment 1 of the present invention;

[0057] Figure 2 This is a schematic diagram of the node tree in Embodiment 1 of the present invention;

[0058] Figure 3 This is a flowchart of a low-power management method for an energy meter according to Embodiment 2 of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] Example 1

[0061] Please see Figure 1 As shown in the figure, the low-power management system for an electricity meter described in this embodiment includes: and ; each module is connected via wired and / or wireless means to realize data transmission between modules;

[0062] The load acquisition module is used to collect real-time load data.

[0063] Real-time load refers to the load power collected in real time, which is the rate at which the load equipment in the power system consumes electrical energy. The method for collecting real-time load data involves obtaining real-time voltage, real-time current, and power factor, and then calculating the real-time load. The expression for real-time load is: In the formula, For real-time load, For real-time voltage, For real-time current, The power factor is used; real-time voltage, real-time current, and power factor are obtained through an energy meter.

[0064] The load calculation module is used to obtain historical load and combine it with real-time load to calculate the average load.

[0065] Historical load refers to the load power collected at historical moments; methods for calculating average load include:

[0066] Set the selected duration and obtain the current time, which corresponds to the real-time load. Subtract the selected duration from the current time to obtain the selected time. Based on the selected time and the current time, construct the calculation period, the range of which is... ,in To select a time, Given the current time; obtain the load power corresponding to the calculation period in the historical load and mark it as the calculation load; count the number of calculation loads and increment it by one, marking it as the load quantity; add each calculation load and the real-time load in sequence, and then divide by the load quantity to obtain the average load.

[0067] The steps to set the selected duration include:

[0068] Step A1: Preset the selected duration range Segmentation ratio and number of iterations This embodiment is preferred. The selected duration range, segmentation ratio, and number of iterations are all set by those skilled in the art according to the actual situation;

[0069] Step A2: Based on the segmentation ratio From the selected time range Two new points are divided in the middle. and , , ;

[0070] Step A3: Based on the points respectively and And at the current moment, select the corresponding computing load from the historical load; based on The selected computational loads are used as the first set. The standard deviation corresponding to the first set is calculated and marked as the first standard deviation; then, based on... The selected computational loads are used as the second set. The standard deviation of the second set is calculated and marked as the second standard deviation.

[0071] Step A4: Using the first standard deviation as the first data point, The first data and the second standard deviation are used as the second data; the first data and the second data are respectively input into the trained matching analysis model to predict the corresponding matching degree; the matching degree corresponding to the first data is marked as the first matching degree, and the matching degree corresponding to the second data is marked as the second matching degree;

[0072] Step A5: If the first matching degree is greater than or equal to the second matching degree, the minimum value of the updated selection duration range is the minimum value of the selection duration range before the update, and the maximum value is the point corresponding to the second matching degree; if the first matching degree is less than the second matching degree, the minimum value of the updated selection duration range is the point corresponding to the first matching degree, and the maximum value is the maximum value of the selection duration range before the update; for example, the selection duration range... Passing Point and After the update, if the first matching degree is greater than or equal to the second matching degree, the update selection time range is [range to be filled in]. If the first matching degree is less than the second matching degree, then the update selection time range is [range to be filled in]. ;

[0073] Step A6: Preset a width threshold. The width threshold is preset by those skilled in the art based on the actual situation. Subtract the minimum value from the maximum value in the updated selection time range to obtain the interval width. Compare the interval width with the width threshold. If the interval width is greater than or equal to the width threshold, then adjust the updated selection time range according to the segmentation ratio. Two new points were redefined. and Then return to step A4; if the interval width is less than the width threshold, then the average of the maximum and minimum values ​​in the updated selection duration range will be used as the selection duration.

[0074] In step A4 above, the training process of the matching analysis model includes:

[0075] The first and second data are marked as research data. Z sets of research data are collected in advance. A corresponding matching degree is set for each of the Z sets of research data, where Z is an integer greater than 1. The matching degree corresponding to the research data is determined by a person skilled in the art during the historical operation of the electricity meter. The Z sets of research data are collected, and each set of research data is analyzed in turn based on practical experience to determine the matching degree corresponding to each set of research data. The corresponding matching degree is set for each of the Z sets of research data in turn.

[0076] The research data and their corresponding matching degrees are converted into a set of feature vectors. Each set of feature vectors is used as input to the matching analysis model, which outputs a set of predicted matching degrees for each set of research data and uses the actual matching degree for each set of research data as the prediction target. The actual matching degree is the pre-set matching degree corresponding to the research data. The training objective is to minimize the sum of prediction errors for all research data. The formula for calculating the prediction error is as follows: ,in The prediction error is represented by K, where K is the group number of the feature vector corresponding to the research data. The predicted matching degree corresponding to the Kth group of research data. Let K be the actual matching degree corresponding to the Kth group of research data; train the matching analysis model until the sum of prediction errors converges and then stop training.

[0077] The matching analysis model described above is specifically a deep neural network model; it includes an input layer, hidden layers, and an output layer; each hidden layer contains multiple neurons, and each neuron is connected to the neurons in the next layer. The connections contain weights that determine the importance and influence of data transmission in the neural network; each neuron between the hidden layer and the output layer applies an activation function, which introduces non-linearity, allowing the network to learn more complex patterns and features.

[0078] The frequency adjustment module dynamically adjusts the data frequency based on the average load.

[0079] Data frequency includes sampling frequency, processing frequency, and transmission frequency; sampling frequency is the number of times the energy meter collects data per unit time, processing frequency is the number of times the energy meter processes data per unit time, and transmission frequency is the number of times the energy meter transmits data per unit time; the unit of unit time is consistent with the unit of the current moment.

[0080] Methods for dynamically adjusting data frequency include:

[0081] Define the search space radius threshold and iteration threshold radius threshold and iteration threshold Pre-configured by those skilled in the art based on actual conditions; search space This includes n frequency sets, where n is an integer greater than 1. The method for constructing the n frequency sets is as follows: Based on the technical parameters of the electricity meter, obtain the frequency range, which includes the sampling frequency range, processing frequency range, and transmission frequency range. Randomly select a value from each range within the frequency range to construct a frequency set, resulting in a total of n distinct frequency sets. Assign sequentially increasing numerical labels to the n frequency sets, marking them as set labels. The range of the set labels is... ;

[0082] Set the initial aperture center. and initial aperture radius And set the number of iterations. ;in, , Determine the superiority function and the iterative process; the iterative process is as follows: generate m candidate solutions within the aperture range, and reduce the aperture radius. , Calculate the merit of each candidate solution, and move the aperture center to the candidate solution with the highest merit; repeat the iterative process, and for each repetition, set the iteration count. Until or When the time is reached, stop the repeated loop process, obtain the candidate solution corresponding to the aperture center, and mark it as the best solution; dynamically adjust the data frequency according to the frequency set corresponding to the set label of the best solution.

[0083] The expression for the superiority function is: In the formula, For excellence, The degree of fit is determined by the following method: based on the set labels corresponding to the candidate solutions, the data frequencies in the corresponding frequency set are obtained and marked as candidate frequencies; the candidate frequencies and average load are used as evaluation data; the evaluation data are input into the trained fit analysis model to predict the corresponding degree of fit; the training process of the fit analysis model is the same as that of the matching analysis model, and both are deep learning network models.

[0084] The expression for the candidate solution is: In the formula, For the i-th candidate solution, The aperture center corresponds to the t-th repetition cycle. Let f be the aperture radius corresponding to the t-th repetition cycle. For random coefficients, , .

[0085] The expression for reducing the aperture radius is: In the formula, This is the reduced aperture radius. The shrinkage coefficient, .

[0086] It should be noted that, in order to reduce the power consumption of the electricity meter, data transmission should be performed when the load power is low. That is, the electricity meter will buffer the collected and processed data to delay data transmission. Therefore, the purpose of dynamically adjusting the transmission frequency is to calculate the transmission frequency corresponding to each moment, which helps to analyze the actual transmission frequency of the electricity meter during the data transmission process.

[0087] The strategy adjustment module dynamically adjusts the transmission strategy based on historical load and data frequency.

[0088] Dynamically adjusting transmission strategies includes determining transmission periods and evaluating transmission frequencies;

[0089] Methods for determining transmission time periods include:

[0090] Based on historical load, predict the load power corresponding to future times and mark them as future loads. Future times include all times from the current time to the end of the day, and the load power in the future loads corresponds one-to-one with the times in the future times; preset low load periods. and medium load section Low load segment and medium load section Pre-configured by those skilled in the art based on actual conditions; according to low load segment and medium load section Determine the power of each load in the future load, if Then the corresponding future time will be marked as a low-load time. For the load power in the future load; if If so, the corresponding future time will be marked as a medium load time; if If so, the corresponding future time will not be marked;

[0091] Sort future times from earliest to latest to generate a time sorting table; mark consecutive low-load times in the time sorting table as low-load periods, and consecutive medium-load times in the time sorting table as medium-load periods; mark the latest low-load time in each low-load period as the latest time, and compare each latest time; mark the low-load period corresponding to the latest latest time as the transmission period; if there is no low-load period in the time sorting table, mark the latest medium-load time in each medium-load time as the latest time, and mark the medium-load period corresponding to the latest latest time as the transmission period.

[0092] Methods for predicting load power at future moments include:

[0093] Historical and real-time loads are used as prediction data. This prediction data is then input into a trained load prediction model to predict the load power at future times. The load prediction model is an RNN neural network model, and its training process includes:

[0094] Pre-continuous collection Each load power is assigned, and a load training set is constructed. , To predict the amount of load power in the data; based on the load training set, train a load prediction model to predict the load power at future times;

[0095] Based on the practical experience of those skilled in the art, a sliding step size L and a sliding window length E are preset; the load power in the load training set is transformed into multiple training samples using the sliding window method, with each training sample including E load powers; each training sample is used as the input to the load prediction model, the predicted load power after the sliding step size L is used as the output, and the actual L load powers corresponding to each training sample are used as the prediction target. The model accuracy is evaluated using the mean absolute percentage error (MAPE). When the calculated MAPE is less than the preset MAPE, the load prediction model corresponding to the load power is considered trained successfully; the formula for calculating the mean absolute percentage error (MAPE) is as follows: In the formula, For the first The predicted target corresponds to the predicted load power. For the first The predicted load power The number of predicted load powers; generate a load prediction model that predicts the load power at future times based on the predicted load power.

[0096] For example, load training set A contains 10 load powers. , For the first Each load power, Define a sliding window with a length of 3 and a sliding step size L of 1. Construct 7 training samples using the sliding window. Each training sample contains 3 consecutive load powers, and the next load power after the 3 consecutive load powers is used as the prediction target. For example: training sample... Training samples The corresponding prediction target is Training samples Training samples The corresponding prediction target is Similarly, this is used to train load prediction models.

[0097] Methods for evaluating transmission frequencies include:

[0098] Mark the latest time corresponding to the transmission period as the end time, and mark all times from the current time to the end time as the analysis period; calculate the average load corresponding to each time in the analysis period based on the load power and historical load; calculate the transmission frequency corresponding to each time in the analysis period based on the average load; obtain the historical end time, which is the time corresponding to the last time the energy meter ended data transmission; construct an estimated period based on the end time and the historical end time, with the range of the estimated period being... ,in For the end of history, The end time is determined by summing the transmission frequencies corresponding to each time point in the estimated time period to obtain the total transmission frequency. The number of time points within the transmission time period is counted, and the total transmission frequency is divided by the number of time points to obtain the transmission frequency corresponding to each time point within the transmission time period.

[0099] The data compression module is used to acquire power data, compress the power data, and transmit the compressed power data based on the transmission strategy.

[0100] Electricity data refers to all data collected by the electricity meter during the estimated period. Electricity data includes electricity consumption data (such as active energy, reactive energy, etc.), power quality data (such as voltage, current, frequency, etc.), and power load data (such as active power, reactive power, peak load, etc.).

[0101] The steps for compressing power data include:

[0102] Step B1: Count the number of each value in the power data and use it as the frequency of occurrence; treat each value as a node and use the frequency of occurrence of each value as the node frequency of the corresponding node; sort all nodes according to the node frequency from small to large to generate a node sorting table.

[0103] Step B2: According to the ascending order of the node sorting table, obtain the two nodes at the very beginning and mark them as child nodes, then delete the child nodes from the node sorting table;

[0104] Step B3: Merge the two child nodes into one parent node and add it to the node sorting table. The node frequency of the parent node is the sum of the node frequencies of the two child nodes.

[0105] Step B4: Repeat steps B2 to B3 until only one node remains in the node sorting table. The loop ends, and the remaining node in the node sorting is taken as the root node. The root node is taken as the starting point and extended downwards. The two child nodes corresponding to each node are obtained in turn until the child nodes corresponding to all nodes are obtained, forming a node tree, and then proceeding to step B5.

[0106] Step B5: Starting from the root node of the node tree, follow the path of the node tree, and when a node is encountered on the left, record its code. When a node is located on the right, record the encoding. ,coding With encoding It is a binary value; until the path of the node tree ends, the recorded codes are combined according to the order of the codes to form the compressed code of the value corresponding to the last node.

[0107] Step B6: Repeat step B5 until the compressed code corresponding to each value in the power data is obtained, and the node tree is used as compressed data; wherein, the order of each compressed code in the compressed data is consistent with the order of each value in the power data.

[0108] For example, the power data includes 1, 2, 3, 3, where the frequency of 1 is 1, the frequency of 2 is 1, and the frequency of 3 is 2. Therefore, first, 1 and 2 are taken as child nodes, their corresponding parent nodes are obtained, and marked as the first parent node. Then, the first parent node and 3 are taken as child nodes, their corresponding parent nodes are obtained, and marked as the second parent node. Since the second parent node is the only remaining node in the node sorting table, the second parent node is the root node. The node tree corresponding to the power data is as follows: Figure 2 As shown; the compressed encoding of 1 is The compressed encoding of 2 is The compressed encoding of 3 is .

[0109] This embodiment accurately calculates the average load by collecting and analyzing real-time and historical load data. Based on the average load, a heuristic algorithm is used to dynamically adjust the data frequency. Combining load prediction and data compression technologies, the transmission strategy is dynamically adjusted to transmit compressed data, achieving refined management and control of the electricity meter. It can effectively adapt to real-time load changes, optimize data transmission time, and reduce transmission load, thereby reducing the power consumption of the electricity meter, improving operating efficiency and service life, meeting the needs of smart grids for sustainable development and high performance, and providing an innovative solution for balancing low power consumption and high performance in electricity meters.

[0110] Example 2

[0111] Please see Figure 3 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A low-power management method for an electricity meter is provided, the method including:

[0112] Collect real-time load;

[0113] Obtain historical load data and combine it with real-time load data to calculate the average load.

[0114] The data frequency is dynamically adjusted based on the average load.

[0115] The transmission strategy is dynamically adjusted based on historical load and data frequency.

[0116] Acquire power data, compress the power data, and transmit the compressed power data based on the transmission strategy.

[0117] Example 3

[0118] This application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code that, when executed by the one or more processors, can perform a low-power management method for an energy meter as described above.

[0119] The method or system according to the embodiments of this application can also be implemented using the architecture of the electronic device shown in this application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as ROM or hard disk, may store a low-power management method for an energy meter provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in this application is merely exemplary; when implementing different devices, one or more components in the electronic device shown in this application may be omitted according to actual needs.

[0120] Example 4

[0121] One embodiment of this application discloses a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by a processor, a low-power management method for an energy meter according to an embodiment of this application, as described with reference to the above figures, can be performed. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0122] Furthermore, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to perform instructions corresponding to the method steps provided in this application, such as a low-power management method for an electricity meter. When this computer program is executed by a central processing unit (CPU), it performs the functions defined in the method of this application.

[0123] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0124] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A low-power management method for an electricity meter, characterized in that, include: Collect real-time load; Obtain historical load data and combine it with real-time load data to calculate the average load. The method for calculating the average load includes: Set the selected duration and obtain the current time; subtract the selected duration from the current time to obtain the selected time; construct the calculation period based on the selected time and the current time; obtain the load power corresponding to the calculation period in the historical load and mark it as the calculation load; count the number of calculation loads and increment it by one, marking it as the load quantity; add each calculation load and the real-time load in sequence, and then divide by the load quantity to obtain the average load; The steps for setting the selected duration include: Step A1: Preset the selected duration range Segmentation ratio and number of iterations ; Step A2: Based on the segmentation ratio From the selected time range Two new points are divided in the middle. and , , ; Step A3: Based on the points respectively and And at the current moment, select the corresponding computing load from the historical load; based on The selected computational loads are used as the first set. The standard deviation corresponding to the first set is calculated and marked as the first standard deviation; then, based on... The selected computational loads are used as the second set. The standard deviation of the second set is calculated and marked as the second standard deviation. Step A4: Using the first standard deviation as the first data point, The first data and the second standard deviation are used as the second data; the first data and the second data are respectively input into the trained matching analysis model to predict the corresponding matching degree; the matching degree corresponding to the first data is marked as the first matching degree, and the matching degree corresponding to the second data is marked as the second matching degree; the matching analysis model is a deep learning network model; Step A5: If the first matching degree is greater than or equal to the second matching degree, the minimum value of the updated selection duration range is the minimum value of the selection duration range before the update, and the maximum value is the point corresponding to the second matching degree; if the first matching degree is less than the second matching degree, the minimum value of the updated selection duration range is the point corresponding to the first matching degree, and the maximum value is the maximum value of the selection duration range before the update. Step A6: Set a preset width threshold, subtract the minimum value from the maximum value in the updated selected duration range to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, then adjust the updated selected duration range according to the segmentation ratio. Two new points were redefined. and Then return to step A4; if the interval width is less than the width threshold, then the average of the maximum and minimum values ​​in the updated selection duration range will be used as the selection duration. The data frequency is dynamically adjusted based on the average load. The transmission strategy is dynamically adjusted based on historical load and data frequency. Acquire power data, compress the power data, and transmit the compressed power data based on the transmission strategy.

2. The low-power management method for an electricity meter according to claim 1, characterized in that, The real-time load is the load power collected in real time, which is the rate at which the load equipment in the power system consumes electrical energy. The method for collecting the real-time load is to obtain the real-time voltage, real-time current, and power factor, and then calculate the real-time load. The expression for the real-time load is: In the formula, For real-time load, For real-time voltage, For real-time current, Power factor; The historical load refers to the load power collected at historical moments; The current time is the time corresponding to the real-time load being collected; the range of the calculation period is... ,in To select a time, This refers to the current moment.

3. The low-power management method for an electricity meter according to claim 2, characterized in that, The data frequency includes the sampling frequency, processing frequency, and transmission frequency; Methods for dynamically adjusting data frequency include: Define the search space radius threshold and iteration threshold Search space This includes n frequency sets, where n is an integer greater than 1. The method for constructing the n frequency sets is as follows: Obtain the frequency range, which includes the sampling frequency range, the processing frequency range, and the transmission frequency range. Randomly select a value from each range within the frequency range to construct a frequency set, resulting in a total of n distinct frequency sets. Assign sequentially increasing numerical labels to the n frequency sets, marking them as set labels. The range of set labels is... ; Set the initial aperture center. and initial aperture radius And set the number of iterations. ;in, , Determine the superiority function and the iterative process; the iterative process is as follows: generate m candidate solutions within the aperture range, and reduce the aperture radius. , Calculate the merit of each candidate solution, and move the aperture center to the candidate solution with the highest merit; repeat the iterative process, and for each repetition, set the iteration count. Until or When the time is reached, stop the repeated loop process, obtain the candidate solution corresponding to the aperture center, and mark it as the best solution; dynamically adjust the data frequency according to the frequency set corresponding to the set label of the best solution.

4. The low-power management method for an electricity meter according to claim 3, characterized in that, The expression for the superiority function is: In the formula, For excellence, The fitness level is determined by the following method: based on the set labels corresponding to the candidate solutions, the data frequencies in the corresponding frequency sets are obtained and marked as candidate frequencies; the candidate frequencies and average load are used as evaluation data; the evaluation data are input into the trained fitness analysis model to predict the corresponding fitness level; the fitness analysis model is a deep learning network model. The expression for the candidate solution is: In the formula, For the i-th candidate solution, The aperture center corresponds to the t-th repetition cycle. Let f be the aperture radius corresponding to the t-th repetition cycle. For random coefficients, , ; The expression for reducing the aperture radius is: In the formula, This is the reduced aperture radius. The shrinkage coefficient, .

5. A low-power management method for an electricity meter according to claim 4, characterized in that, The dynamic adjustment transmission strategy includes determining the transmission period and evaluating the transmission frequency; The method for determining the transmission time period includes: Based on historical load, predict the load power corresponding to future times and mark them as future loads. Future times include all times from the current time to the end of the day, and the load power in the future loads corresponds one-to-one with the times in the future times; preset low load periods. and medium load section According to the low load segment and medium load section Determine the power of each load in the future load, if Then the corresponding future time will be marked as a low-load time. For the load power in the future load; if If so, the corresponding future time will be marked as a medium load time; if If so, the corresponding future time will not be marked; Sort future times from earliest to latest to generate a time sorting table; mark consecutive low-load times in the time sorting table as low-load periods, and consecutive medium-load times in the time sorting table as medium-load periods; mark the latest low-load time in each low-load period as the latest time, and compare each latest time; mark the low-load period corresponding to the latest latest time as the transmission period; if there is no low-load period in the time sorting table, mark the latest medium-load time in each medium-load time as the latest time, and mark the medium-load period corresponding to the latest latest time as the transmission period.

6. The low-power management method for an electricity meter according to claim 5, characterized in that, The method for predicting the load power at future times includes: Historical and real-time loads are used as prediction data. This prediction data is then input into a trained load prediction model to predict the load power at future times. The load prediction model is an RNN neural network model, and its training process includes: Pre-continuous collection Each load power is assigned, and a load training set is constructed. , To predict the amount of load power in the data; based on the load training set, train a load prediction model to predict the load power at future times; The system presets the sliding step size L and the sliding window length E; it transforms the load power in the load training set into multiple training samples using the sliding window method, with each training sample containing E load powers; it uses each training sample as input to the load prediction model, predicts the load power after the sliding step size L as output, and uses the actual L load powers corresponding to each training sample as prediction targets. It evaluates the model accuracy using the mean absolute percentage error (MAPE) of the prediction results. When the calculated MAPE is less than the preset MAPE, the load prediction model corresponding to the load power is considered trained successfully; and it generates a load prediction model that predicts the load power at future times based on the load power.

7. A low-power management method for an electricity meter according to claim 6, characterized in that, The method for evaluating the transmission frequency includes: Mark the latest time corresponding to the transmission period as the end time, and mark all times from the current time to the end time as the analysis period; calculate the average load corresponding to each time in the analysis period based on the load power and historical load; calculate the transmission frequency corresponding to each time in the analysis period based on the average load; obtain the historical end time, which is the time corresponding to the last time the energy meter ended data transmission; construct an estimated period based on the end time and the historical end time, with the range of the estimated period being... ,in For the end of history, The end time is determined by summing the transmission frequencies corresponding to each time point in the estimated time period to obtain the total transmission frequency. The number of time points within the transmission time period is counted, and the total transmission frequency is divided by the number of time points to obtain the transmission frequency corresponding to each time point within the transmission time period.

8. A low-power management method for an electricity meter according to claim 7, characterized in that, The electricity data refers to all data collected by the electricity meter within the estimated time period; The steps for compressing the power data include: Step B1: Count the number of each value in the power data and use it as the frequency of occurrence; treat each value as a node and use the frequency of occurrence of each value as the node frequency of the corresponding node; sort all nodes according to the node frequency from small to large to generate a node sorting table. Step B2: According to the ascending order of the node sorting table, obtain the two nodes at the very beginning and mark them as child nodes, then delete the child nodes from the node sorting table; Step B3: Merge the two child nodes into one parent node and add it to the node sorting table. The node frequency of the parent node is the sum of the node frequencies of the two child nodes. Step B4: Repeat steps B2 to B3 until only one node remains in the node sorting table. The loop ends, and the remaining node in the node sorting is taken as the root node. The root node is taken as the starting point and extended downwards. The two child nodes corresponding to each node are obtained in turn until the child nodes corresponding to all nodes are obtained, forming a node tree, and then proceeding to step B5. Step B5: Starting from the root node of the node tree, follow the path of the node tree, and when a node is encountered on the left, record its code. When a node is located on the right, record the encoding. ,coding With encoding It is a binary value; until the path of the node tree ends, the recorded codes are combined according to the order of the codes to form the compressed code of the value corresponding to the last node. Step B6: Repeat step B5 until the compressed code corresponding to each value in the power data is obtained, and the node tree is used as compressed data; wherein, the order of each compressed code in the compressed data is consistent with the order of each value in the power data.

9. A low-power management system for an electricity meter, implementing the low-power management method for an electricity meter according to any one of claims 1-8, characterized in that, include: The load acquisition module is used to acquire real-time load data. The load calculation module is used to obtain historical load and combine it with real-time load to calculate the average load; The frequency adjustment module dynamically adjusts the data frequency based on the average load. The strategy adjustment module dynamically adjusts the transmission strategy based on historical load and data frequency. The data compression module is used to acquire power data, compress the power data, and transmit the compressed power data based on the transmission strategy.

Images