A power consumption data calibration method and system

Abstract

The application discloses a kind of power utilization data calibration method and system, method includes: obtaining power utilization equipment time series power utilization data;Based on sliding window gradient detection and relative change rate continuous criterion identifies transient impact data;Generation normal subsequence and transient subsequence set, realize batch processing by sequence length clustering;Transient subsequence is associated with adjacent normal subsequence, according to the first correction based on dynamic threshold positioning cutoff point, recover range saturation data by reverse difference compensation;Second correction utilizes the calibration model driven by operating parameter to govern harmonic distortion.It is more accurate to separate equipment state switching transient and noise interference, realize the same type transient batch calibration, effectively power utilization data calibration accuracy and efficiency.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of power consumption data processing, and particularly relates to a power consumption data calibration method and system. BACKGROUND

[0002] In the field of smart grid, transient impact current generated by state switching of power consumption equipment (such as motor start-stop and load mutation) can cause millisecond-level peak values of power consumption data, which exceed the upper limit of the range of electric energy meters. Traditional electric energy metering devices are limited by sampling rate and range, and cannot completely collect real power consumption in the transient process, resulting in data truncation distortion.

[0003] In the prior art, the correction for the transient interval is mostly processed by a single model end to end, without decoupling systematic truncation error caused by range saturation and nonlinear error caused by harmonic distortion, which leads to the lack of physical consistency of the corrected data, and further affects the accuracy of power consumption data calibration. SUMMARY

[0004] The present application provides a power consumption data calibration method and system, which is used to solve the technical problem of the lack of physical consistency of the corrected data caused by the decoupling of systematic truncation error caused by range saturation and nonlinear error caused by harmonic distortion.

[0005] In a first aspect, the present application provides a power consumption data calibration method, comprising:

[0006] obtaining power consumption data of power consumption equipment at different time nodes;

[0007] determining whether there is transient impact power consumption data in each power consumption data according to a preset transient transition detection strategy, wherein the transient impact power consumption data is power consumption data generated by state switching of the power consumption equipment;

[0008] if there is at least one transient impact power consumption data, generating at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence, and performing sequence division on the at least one transient impact power consumption data subsequence to obtain at least one transient impact power consumption data subsequence set;

[0009] associating one transient impact power consumption data subsequence and at least one normal power consumption data subsequence adjacent to the one transient impact power consumption data subsequence to obtain at least one association relationship;

[0010] selecting a target transient impact power consumption data subsequence in a certain transient impact power consumption data subsequence set, and using a preset first data correction strategy to correct the target transient impact power consumption data subsequence according to a target association relationship containing the target transient impact power consumption data subsequence, to obtain a certain corrected transient impact power consumption data subsequence and a certain correction amount.

[0011] According to the certain correction amount, other transient impact power consumption data subsequences in the certain transient impact power consumption data subsequence set are directly corrected to obtain other corrected transient impact power consumption data subsequences;

[0012] The corrected transient impact power consumption data subsequences and the at least one normal power consumption data subsequence are spliced based on the chronological order of the time nodes to obtain a to-be-calibrated power consumption data sequence, and the to-be-calibrated power consumption data sequence is corrected again according to a preset second data correction strategy, so that the calibration power consumption data of the power consumption equipment at different time nodes is finally obtained.

[0013] In a second aspect, the present application provides a power consumption data calibration system, comprising:

[0014] An acquisition module configured to acquire power consumption data of a power consumption equipment at different time nodes;

[0015] A judgment module configured to judge whether transient impact power consumption data exists in each power consumption data according to a preset transient transition detection strategy, wherein the transient impact power consumption data is power consumption data generated by the power consumption equipment when the state is switched;

[0016] A division module configured to generate at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence if there is at least one transient impact power consumption data, and to perform sequence division on the at least one transient impact power consumption data subsequence to obtain at least one transient impact power consumption data subsequence set;

[0017] An association module configured to associate one transient impact power consumption data subsequence and at least one normal power consumption data subsequence adjacent to the one transient impact power consumption data subsequence to obtain at least one association relationship;

[0018] A first correction module configured to select a target transient impact power consumption data subsequence from a certain transient impact power consumption data subsequence set, and to correct the target transient impact power consumption data subsequence according to a target association relationship containing the target transient impact power consumption data subsequence by using a preset first data correction strategy to obtain a certain corrected transient impact power consumption data subsequence and a certain correction amount;

[0019] A second correction module configured to correct other transient impact power consumption data subsequences in the certain transient impact power consumption data subsequence set according to the certain correction amount to obtain other corrected transient impact power consumption data subsequences;

[0020] The third correction module is configured to splice each correction transient impact power consumption data sub-sequence and the at least one normal power consumption data sub-sequence based on the chronological order of the time nodes to obtain a to-be-calibrated power consumption data sequence, and to correct the to-be-calibrated power consumption data sequence again according to a preset second data correction strategy, so as to finally obtain the calibrated power consumption data of the power consumption equipment at different time nodes.

[0021] In a third aspect, an electronic device is provided, which includes at least one processor and a memory connected with the at least one processor in communication, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the power consumption data calibration method of any one of the embodiments of the present application.

[0022] In a fourth aspect, the present application further provides a computer readable storage medium having a computer program stored thereon, and the program instructions are executed by a processor to enable the processor to perform the steps of the power consumption data calibration method of any one of the embodiments of the present application.

[0023] The power consumption data calibration method and system of the present application can achieve batch efficient calibration of similar transient events as much as possible through the transient sub-sequence set division and the transmission strategy of the correction amount. The two-time correction cooperative mechanism is adopted, the first correction compensates for the truncation distortion caused by the range saturation and reconstructs, and the second correction governs the nonlinear errors such as harmonic distortion based on the dynamic model driven by the operating parameters, and the double-layer cascade greatly improves the data fidelity compared with the single correction scheme. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0025] Figure 1 A flow chart of a power consumption data calibration method provided by an embodiment of the present application;

[0026] Figure 2 A structural block diagram of a power consumption data calibration system provided by an embodiment of the present application;

[0027] Figure 3 A structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0029] Referring to Figure 1 , a flowchart of a power consumption data calibration method is shown.

[0030] As Figure 1 shown, the power consumption data calibration method specifically includes the following steps:

[0031] In step S101, power consumption data of the power consumption equipment at different time nodes is obtained.

[0032] In this step, the power consumption equipment can be different electrical equipment, i.e., power consumption data of the electrical equipment at different time nodes is obtained, wherein the power consumption data specifically can include power consumption. For example, the power consumption of the electrical equipment from 8 o'clock to 12 o'clock is obtained, specifically, the stage from 8 o'clock to 12 o'clock can experience the start, state mode switching or shutdown of the electrical equipment.

[0033] In step S102, it is judged whether there is transient impact power consumption data in each power consumption data according to a preset transient transition detection strategy, wherein the transient impact power consumption data is power consumption data generated by the power consumption equipment when the state is switched.

[0034] In this step, based on the order of time nodes, the change gradient between adjacent two power consumption data is obtained in sequence to obtain a change gradient sequence; according to the preset sliding window, the change gradient sequence is slid, and each time the sliding window is slid, it is judged whether the change gradient in the sliding window is greater than a preset threshold; if each change gradient is not greater than the preset threshold, it is determined that there is no transient impact power consumption data in each power consumption data; if a change gradient is greater than the preset threshold, a relative change rate between each target change gradient and the change gradient is obtained, wherein the target change gradient is a change gradient located after the change gradient in the change gradient sequence; it is judged whether the number of relative change rates greater than a preset change rate threshold continuously is greater than a preset threshold; if the number is greater than the preset threshold, the power consumption data corresponding to the change gradient and the power consumption data corresponding to the relative change rate greater than the preset change rate threshold continuously are defined as the transient impact power consumption data, otherwise not.

[0035] In one specific application scenario, there is a power transition at the start-stop moment of an electrical device (rated power 120 kW), which causes the power meter to be unable to accurately record the transient power consumption due to the range limit (0-150 kW). Therefore, a transient transition detection strategy is needed to identify the transient impact interval in the power consumption data, providing a basis for subsequent calibration.

[0036] The power consumption sequence (unit: kWh) generated by the electrical device with 1 Hz sampling accumulation is obtained, and the timestamp is accurate to seconds:

[0037] Standby state: T0-5s: 10.000 → T0-4s: 10.000 → T0-3s: 10.000 → T0-2s: 10.000 → T0-1s: 10.000 → T0 s: 10.000;

[0038] After the T0s start command is issued: T0+1s: 10.350 → T0+2s: 10.800 → T0+3s: 11.400 → T0+4s: 12.600 → T0+5s: 150.000 → T0+6s: 150.000 → T0+7s: 150.000 → T0+8s: 150.000 → T0+9s: 152.100 → T0+10s: 50.100 → T0+11s: 52.100.

[0039] Based on the sequence of time nodes, the change gradient (absolute value of the difference between the adjacent two power consumption data) between the adjacent two power consumption data is obtained in turn, i.e. 10-10, 10-10, 10-10, 10-10, 10-10, 10.35-10, 10.8-10.35, 11.4-10.8, 12.6-11.4, 150-12.6, 150-150, 150-150, 150-150, 152.1-150, 50.1-152.1, 52.1-50.1. The change gradient sequence is change gradient A (value 0), change gradient B (value 0), change gradient C (value 0), change gradient D (value 0), change gradient E (value 0), change gradient F (value 0.35), change gradient G (value 0.45), change gradient H (value 0.6), change gradient J (value 1.2), change gradient K (value 137.4), change gradient P (value 0), change gradient Q (value 0), change gradient R (value 0), change gradient S (value 2.1), change gradient T (value 102), change gradient U (value 2).

[0040] Where the gradient K is greater than the preset threshold (50), the first relative rate of change between gradient P and gradient K (|0-137.4| / 137.4=1), the second relative rate of change between gradient Q and gradient K (|0-137.4| / 137.4=1), the third relative rate of change between gradient R and gradient K (|0-137.4| / 137.4=1), the fourth relative rate of change between gradient S and gradient K (|2.1-137.4| / 137.4=0.48), the fifth relative rate of change between gradient T and gradient K (|102-137.4| / 137.4=0.26), and the sixth relative rate of change between gradient U and gradient K (|2-137.4| / 137.4=0.99) are calculated respectively.

[0041] Specifically, the first relative rate of change, the second relative rate of change, the third relative rate of change, and the fourth relative rate of change are continuously greater than the preset rate of change threshold (0.4), and the number of them is 4, which is greater than the preset number threshold (2). Among them, the first relative rate of change, the second relative rate of change, the third relative rate of change, and the fourth relative rate of change correspond to the change gradient P, the change gradient Q, the change gradient R, and the change gradient S, respectively. Therefore, the electricity consumption data corresponding to the change gradient K, the change gradient P, the change gradient Q, the change gradient R, and the change gradient S (T0+5s:150.000, T0+6s:150.000, T0+7s:150.000, T0+8s:150.000, T0+9s:152.100) are defined as transient impact electricity consumption data.

[0042] In one specific embodiment, after determining whether there is transient impact power consumption data in each power consumption data according to the preset transient transition detection strategy, if there is no transient impact power consumption data, the power consumption data is sorted according to the order of time nodes to directly obtain the second power consumption data sequence to be calibrated, and the second power consumption data sequence to be calibrated is corrected according to the second data correction strategy to finally obtain the calibration power consumption data of the power equipment at different time nodes.

[0043] Step S103: If there is at least one transient impact power consumption data, then at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence are generated, and the at least one transient impact power consumption data subsequence is divided into sequences to obtain at least one transient impact power consumption data subsequence set.

[0044] In this step, the sequence length of at least one transient impact power consumption data subsequence is obtained, where the sequence length is the number of transient impact power consumption data in the transient impact power consumption data subsequence; the transient impact power consumption data subsequences with the same sequence length are clustered to obtain at least one set of transient impact power consumption data subsequences.

[0045] For example, a power consumption sequence (unit: kWh) generated by sampling at 1Hz, with timestamps accurate to the second:

[0046] Standby state: T0-5s:10.000→T0-4s:10.000→T0-3s:10.000→T0-2s:10.000→T0-1s:10.000→T0 s:10.000;

[0047] After the T0s start command is issued: T0+1s: 10.350 → T0+2s: 10.800 → T0+3s: 11.400 → T0+4s: 12.600 → T0+5s: 150.000 → T0+6s: 150.000 → T0+7s: 150.000 → T0+8s: 150.000 → T0+9s: 152.100 → T0+10s: 50.100 → T0+11s: 52.100.

[0048] Specifically, the transient impact power data subsequence is: (T0+5s:150.000, T0+6s:150.000, T0+7s:150.000, T0+8s:150.000, T0+9s:152.100).

[0049] A normal electricity consumption data subsequence is: (T0-5s:10.000, T0-4s:10.000, T0-3s:10.000, T0-2s:10.000, T0-1s:10.000, T0s:10.000, T0+1s:10.350, T0+2s:10.800, T0+3s:11.400, T0+4s:12.600);

[0050] Another normal electricity consumption data subsequence is: (T0+10s:50.100, T0+11s:52.100).

[0051] Step S104: Associate a transient impact power consumption data subsequence with at least one normal power consumption data subsequence adjacent to the transient impact power consumption data subsequence to obtain at least one association relationship.

[0052] In this step, when the electrical equipment changes from standby state to start state, the start state will generate a transient impact power consumption data subsequence. There is an adjacent normal power consumption data subsequence before the acquisition time of the transient impact power consumption data subsequence, and there is another adjacent normal power consumption data subsequence after the acquisition time of the transient impact power consumption data subsequence.

[0053] For example, the transient impact power consumption data subsequence (T0+5s:150.000, T0+6s:150.000, T0+7s:150.000, T0+8s:150.000, T0+9s:152.100) and a normal power consumption data subsequence (T0-5s:10.000, T0-4s:10.000, T0-3s:10.000) are respectively compared to these subsequences. The following data sequences are correlated: T0-2s:10.000, T0-1s:10.000, T0s:10.000, T0+1s:10.350, T0+2s:10.800, T0+3s:11.400, T0+4s:12.600, and another normal electricity consumption data subsequence (T0+10s:50.100, T0+11s:52.100).

[0054] Step S105: Select a target transient impact power consumption data subsequence from a certain set of transient impact power consumption data subsequences, and according to the target correlation relationship containing the target transient impact power consumption data subsequence, use a preset first data correction strategy to correct the target transient impact power consumption data subsequence to obtain a corrected transient impact power consumption data subsequence and a correction amount.

[0055] In this step, one transient impact power consumption data subsequence is arbitrarily selected from a certain set of transient impact power consumption data subsequences as the target transient impact power consumption data subsequence. The sequence lengths of all transient impact power consumption data subsequences in a certain set of transient impact power consumption data subsequences are the same.

[0056] It should be noted that, based on the target correlation relationship containing the target transient impact power consumption data subsequence, a preset first data correction strategy is used to correct the target transient impact power consumption data subsequence to obtain a corrected transient impact power consumption data subsequence, and a specific correction amount is performed in the following steps:

[0057] Step S1051: Obtain the target normal power consumption data subsequence associated with the target transient impact power consumption data subsequence according to the target correlation relationship, and determine whether the number of target normal power consumption data subsequences is greater than a preset number threshold.

[0058] In this embodiment, by determining whether the number of target normal power consumption data subsequences is greater than a preset threshold, the number of target normal power consumption data subsequences associated with the target transient impact power consumption data subsequence can be determined. The preset threshold is set to 1. Here, the target normal power consumption data subsequence is the normal power consumption data subsequence associated with the target normal power consumption data subsequence.

[0059] In one specific embodiment, after determining whether the number of target normal power consumption data subsequences is greater than a preset number threshold, if it is not greater than the preset number threshold, the sequence length of the target transient impact power consumption data subsequence is obtained. A first target normal power consumption data subsequence segment with the same sequence length and adjacent to the target transient impact power consumption data subsequence is extracted from the first target normal power consumption data subsequence. The first target normal power consumption data subsequence is a normal power consumption data subsequence adjacent to the target transient impact power consumption data subsequence. The second average value of each normal power consumption data in the first target normal power consumption data subsequence segment is calculated, and the second average value is defined as the second dynamic normal power consumption threshold. The target transient impact power consumption data subsequence is searched sequentially in the positive direction to see if there is another target transient impact power consumption data that is not greater than the second dynamic normal power consumption threshold.

[0060] The second dynamic normal power consumption threshold, calculated based on the segment, reflects the current load level of electrical equipment in real time, replacing manually set fixed thresholds and solving the problem of rigid transient judgment standards under light / heavy load conditions. This allows for better identification of the target transient impact power consumption data at the time of transient impact cutoff.

[0061] Step S1052: If the number is greater than a preset threshold, the sequence length of the target transient impact power consumption data subsequence is obtained. The first target normal power consumption data subsequence and the second target normal power consumption data subsequence are respectively extracted from the first target normal power consumption data subsequence and the second target normal power consumption data subsequence, with the same sequence length and adjacent to the target transient impact power consumption data subsequence. The first target normal power consumption data subsequence and the second target normal power consumption data subsequence are both normal power consumption data subsequences adjacent to the target transient impact power consumption data subsequence.

[0062] Step S1053: Calculate the first average value of each normal power consumption data in the first target normal power consumption data sub-sequence segment and the second target normal power consumption data sub-sequence segment, and define the first average value as the first dynamic normal power consumption threshold.

[0063] Step S1054: Search sequentially in the target transient impact power consumption data subsequence along the preset positive direction to see if there is a target transient impact power consumption data that is not greater than the first dynamic normal power consumption threshold. Here, the positive direction is the direction from the first time node to the last.

[0064] For example, a power consumption sequence (unit: kWh) generated by sampling at 1Hz, with timestamps accurate to the second:

[0065] Standby state: T0-5s:10.000→T0-4s:10.000→T0-3s:10.000→T0-2s:10.000→T0-1s:10.000→T0 s:10.000;

[0066] After the T0s start command is issued: T0+1s: 10.350 → T0+2s: 10.800 → T0+3s: 11.400 → T0+4s: 12.600 → T0+5s: 150.000 → T0+6s: 150.000 → T0+7s: 150.000 → T0+8s: 150.000 → T0+9s: 152.100 → T0+10s: 50.100 → T0+11s: 52.100. The positive direction refers to the direction of T0-5s → T0-4s → T0-3s → T0-2s → T0-1s → T0s → … → T0+11s.

[0067] Step S1055: When there is a target transient impact power consumption data, the search is stopped, the target transient impact power consumption data is defined as the cutoff target transient impact power consumption data, and the cutoff target transient impact power consumption data and the first power consumption data difference between the target transient impact power consumption data that is in the opposite direction to the cutoff target transient impact power consumption data are obtained.

[0068] In this embodiment, by determining the cutoff target transient impact power consumption data and the first power consumption data difference between the target transient impact power consumption data that is inversely adjacent to the cutoff target transient impact power consumption data, potential normal power consumption data that may exist in the target transient impact power consumption data subsequence can be eliminated without performing data calibration based on the first power consumption data difference on the potential normal power consumption data.

[0069] For example, suppose the obtained transient impact power consumption data subsequence is: (T0+15s:150.000, T0+16s:150.000, T0+17s:150.000, T0+18s:150.000, T0+19s:152.100, T0+20s:145.500, T0+21s:150.00), where T0+20s:145.500 and T0+21s:150.00 may be the actual power consumption of the electrical equipment, which does not exceed the maximum range of the electricity meter. By calculating the first average value of each normal power consumption data in the first target normal power consumption data subsequence segment and the second target normal power consumption data subsequence segment, and defining the first average value as the first dynamic normal power consumption threshold (146), the system searches sequentially along a preset positive direction for a target transient impact power consumption data (T0+20s:145.500) that is not greater than the first dynamic normal power consumption threshold. This avoids performing the first calibration on the normal power consumption data (T0+21s:150.000). This significantly improves the accuracy of power consumption data calibration.

[0070] Step S1056: Add the first power consumption data difference to each of the reverse target transient impact power consumption data in the target transient impact power consumption data subsequence to obtain a certain corrected transient impact power consumption data subsequence corresponding to the target transient impact power consumption data subsequence. Each of the reverse target transient impact power consumption data is the target transient impact power consumption data with a time node before the time node corresponding to the cutoff target transient impact power consumption data.

[0071] For example, the transient impact power consumption data subsequence is: (T0+15s:150.000, T0+16s:150.000, T0+17s:150.000, T0+18s:150.000, T0+19s:152.100, T0+20s:145.500, T0+21s:150.00). The cutoff transient impact power consumption data is: T0+20s:145.500; the reverse transient impact power consumption data is: T0+15s:150.000, T0+16s:150.000, T0+17s:150.000, T0+18s:150.000.

[0072] In summary, the first data correction strategy dynamically generates a first dynamic normal power consumption threshold by associating adjacent normal power consumption data subsequences, replacing the traditional fixed threshold or linear interpolation method. This effectively addresses the benchmark drift caused by equipment load fluctuations. Furthermore, it ensures the spatiotemporal correlation between the correction benchmark and the transient event by truncating associated normal subsequence segments based on sequence length matching and calculating their average value. It locates the first cutoff target transient impact power consumption data not exceeding the dynamic threshold along the positive direction, accurately identifying the state transition completion point. Using the difference between the first power consumption data of the cutoff point and the adjacent point in the opposite direction as a compensation amount, it performs directional compensation correction on the reverse time series data, eliminating truncation distortion caused by range saturation while fully preserving the nonlinear dynamic characteristics of the transient process. These features form a closed-loop correction chain, achieving physical consistency reconstruction of transient impact data while avoiding subjective setting of correction parameters, providing high-fidelity basic data for accurate energy consumption metering.

[0073] Step S106: Based on the certain correction amount, directly correct other transient impact power data subsequences in the set of transient impact power data subsequences to obtain other corrected transient impact power data subsequences.

[0074] In this step, a certain correction amount is the difference in the first electricity consumption data.

[0075] Specifically, other transient impact power consumption data subsequences are aligned with the target transient impact power consumption data subsequence, and cut-off transient impact power consumption data in other transient impact power consumption data subsequences are determined with the same sorting position as the cut-off target transient impact power consumption data;

[0076] A certain correction amount is added to each cutoff transient impact power consumption data in the other transient impact power consumption data subsequence to obtain other corrected transient impact power consumption data subsequences corresponding to the other transient impact power consumption data subsequences. Among them, each reverse other transient impact power consumption data is other transient impact power consumption data whose time node is before the time node corresponding to the cutoff transient impact power consumption data.

[0077] In this embodiment, the transient impact power consumption data subsequences are clustered by sequence length. Since the sequence lengths are the same, the switching reasons for the various transient impact power consumption data subsequences generated by the power equipment are also very likely to be the same. Therefore, by obtaining a certain correction amount, the reverse transient impact power consumption data in other transient impact power consumption data subsequences are quickly corrected. This maximizes the efficiency of power consumption data correction. For example, if the power equipment is started at time T0, and then shut down and started again at time T20, the sequence lengths of the transient impact power consumption data sequences generated by the transient impact at time T0 and time T20 are also very likely to be the same.

[0078] Step S107: The various modified transient impact power consumption data subsequences are spliced ​​together with the at least one normal power consumption data subsequence based on the chronological order of time nodes to obtain the power consumption data sequence to be calibrated. The power consumption data sequence to be calibrated is then corrected again according to the preset second data correction strategy to finally obtain the calibration power consumption data of the power equipment at different time nodes.

[0079] In this step, it is assumed that there is a transient impact power consumption data subsequence (T0+5s:150.000, T0+6s:150.000, T0+7s:150.000, T0+8s:150.000, T0+9s:152.100).

[0080] A normal electricity consumption data subsequence is: (T0-5s:10.000, T0-4s:10.000, T0-3s:10.000, T0-2s:10.000, T0-1s:10.000, T0s:10.000, T0+1s:10.350, T0+2s:10.800, T0+3s:11.400, T0+4s:12.600);

[0081] Another normal electricity consumption data subsequence is: (T0+10s:50.100, T0+11s:52.100).

[0082] After the transient impact power consumption data subsequence is corrected by the first data correction strategy, the corrected transient impact power consumption data subsequence is obtained. Then, it is spliced ​​according to the order of T0-5s, T0-4s, T0-3s, T0-2s, T0-1s, T0s, T0+1s, T0+2s, T0+3s, T0+4s, T0+5s, T0+6s, T0+7s, T0+8s, T0+9s, T0+10s, T0+11s to obtain the power consumption data sequence to be calibrated.

[0083] It should be noted that, according to the preset second data correction strategy, the power consumption data sequence to be calibrated is corrected again, and the final calibration power consumption data of the equipment at different time points specifically includes:

[0084] The current operating parameters of the electrical equipment are input into the preset power consumption data calibration model. The power consumption data calibration model outputs the correction coefficients corresponding to the current operating parameters. Based on the correction coefficients, each power consumption data to be calibrated in the power consumption data sequence is corrected to obtain the calibration power consumption data of the electrical equipment at different time points.

[0085] Furthermore, operating parameters may include, but are not limited to, ambient temperature, equipment load rate, and historical energy consumption patterns. The ratio of the actual electricity consumption (measured with high precision) under different operating parameters to the electricity consumption to be calibrated (raw data from the electricity meter) is used as a correction coefficient. Then, the operating parameters and the correction coefficient are input to train the DSN network (correction coefficient), thus obtaining the electricity consumption data calibration model.

[0086] In this embodiment, the first data correction strategy solves the data existence problem caused by range saturation through dynamic threshold positioning and difference compensation; the second data correction strategy addresses the waveform fidelity problem caused by harmonics / phase shift based on the complete waveform and operating parameters; the two form a technical closed loop of "reconstruction first, then optimization", which can surpass end-to-end single correction schemes in terms of accuracy, robustness and efficiency as much as possible, and is especially suitable for highly dynamic transient application scenarios.

[0087] In summary, the method presented in this application, through the partitioning of transient subsequence sets and the strategy of transferring correction amounts, can achieve efficient batch calibration of similar transient events as much as possible. It employs a two-stage correction coordination mechanism—the first correction compensates for and reconstructs truncation distortion caused by range saturation, while the second correction relies on a dynamic model driven by operating parameters to manage nonlinear errors such as harmonic distortion. This two-stage cascade significantly improves data fidelity compared to a single-stage correction scheme. In highly dynamic power consumption scenarios, this technical solution simultaneously achieves a leap in metering accuracy and system resource optimization, providing reliable technical support for refined energy consumption management.

[0088] Please see Figure 2 The diagram shows a structural block diagram of an electrical data calibration system according to this application.

[0089] like Figure 2 As shown, the power consumption data calibration system 200 includes an acquisition module 210, a judgment module 220, a division module 230, an association module 240, a first correction module 250, a second correction module 260, and a third correction module 270.

[0090] The system includes: an acquisition module 210 configured to acquire power consumption data of electrical equipment at different time points; a judgment module 220 configured to determine whether transient impact power consumption data exists in each power consumption data according to a preset transient transition detection strategy, wherein the transient impact power consumption data is the power consumption data generated by the electrical equipment during state switching; a partitioning module 230 configured to generate at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence if at least one transient impact power consumption data exists, and to partition the at least one transient impact power consumption data subsequence to obtain at least one set of transient impact power consumption data subsequences; an association module 240 configured to associate a transient impact power consumption data subsequence with at least one normal power consumption data subsequence adjacent to the transient impact power consumption data subsequence to obtain at least one association relationship; and a first correction module 250 configured to correct a transient impact power consumption data subsequence. A target transient impact power consumption data subsequence is selected from the subsequence set. Based on the target correlation relationship containing the target transient impact power consumption data subsequence, a preset first data correction strategy is used to correct the target transient impact power consumption data subsequence to obtain a corrected transient impact power consumption data subsequence and a correction amount. A second correction module 260 is configured to directly correct other transient impact power consumption data subsequences in the set of the target transient impact power consumption data subsequences based on the correction amount to obtain other corrected transient impact power consumption data subsequences. A third correction module 270 is configured to concatenate each corrected transient impact power consumption data subsequence with the at least one normal power consumption data subsequence based on the chronological order of time nodes to obtain a power consumption data sequence to be calibrated. The power consumption data sequence to be calibrated is then corrected again according to the preset second data correction strategy to finally obtain the calibration power consumption data of the power equipment at different time nodes.

[0091] It should be understood that Figure 2 The modules and references described in the document Figure 1 The steps described in the text correspond to those in the method described above. Therefore, the operations, features, and corresponding technical effects described above also apply to the method described in the text. Figure 2 The various modules in the document will not be described in detail here.

[0092] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor performs the power consumption data calibration method in any of the above method embodiments.

[0093] In one embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, which are configured as follows:

[0094] Obtain power consumption data of electrical equipment at different time points;

[0095] The system determines whether transient impact power consumption data exists in each power consumption data according to a preset transient transition detection strategy. The transient impact power consumption data refers to the power consumption data generated by the power-consuming equipment when the state changes.

[0096] If there is at least one transient impact power consumption data, then at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence are generated, and the at least one transient impact power consumption data subsequence is divided into sequences to obtain at least one transient impact power consumption data subsequence set.

[0097] Associating a transient impact power consumption data subsequence with at least one normal power consumption data subsequence adjacent to the transient impact power consumption data subsequence, at least one association relationship is obtained;

[0098] A target transient impact power consumption data subsequence is selected from a certain set of transient impact power consumption data subsequences. Based on the target correlation relationship containing the target transient impact power consumption data subsequence, a preset first data correction strategy is used to correct the target transient impact power consumption data subsequence to obtain a corrected transient impact power consumption data subsequence and a correction amount.

[0099] Based on the aforementioned correction amount, other transient impact power consumption data subsequences in the aforementioned transient impact power consumption data subsequence set are directly corrected to obtain other corrected transient impact power consumption data subsequences;

[0100] Each modified transient impact power consumption data subsequence is concatenated with the at least one normal power consumption data subsequence based on the chronological order of time nodes to obtain the power consumption data sequence to be calibrated. The power consumption data sequence to be calibrated is then corrected again according to a preset second data correction strategy to finally obtain the calibration power consumption data of the power equipment at different time nodes.

[0101] Computer-readable storage media may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the power consumption data calibration system, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory, and may also include memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely configured relative to a processor, which can be connected to the power consumption data calibration system via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0102] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the power data calibration method described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the power data calibration system. The output device 340 may include a display screen or other display device.

[0103] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0104] In one implementation, the above-described electronic device is used in an electrical data calibration system for a client, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0105] Obtain power consumption data of electrical equipment at different time points;

[0106] The system determines whether transient impact power consumption data exists in each power consumption data according to a preset transient transition detection strategy. The transient impact power consumption data refers to the power consumption data generated by the power-consuming equipment when the state changes.

[0107] If there is at least one transient impact power consumption data, then at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence are generated, and the at least one transient impact power consumption data subsequence is divided into sequences to obtain at least one transient impact power consumption data subsequence set.

[0108] Associating a transient impact power consumption data subsequence with at least one normal power consumption data subsequence adjacent to the transient impact power consumption data subsequence, at least one association relationship is obtained;

[0109] A target transient impact power consumption data subsequence is selected from a certain set of transient impact power consumption data subsequences. Based on the target correlation relationship containing the target transient impact power consumption data subsequence, a preset first data correction strategy is used to correct the target transient impact power consumption data subsequence to obtain a corrected transient impact power consumption data subsequence and a correction amount.

[0110] Based on the aforementioned correction amount, other transient impact power consumption data subsequences in the aforementioned transient impact power consumption data subsequence set are directly corrected to obtain other corrected transient impact power consumption data subsequences;

[0111] Each modified transient impact power consumption data subsequence is concatenated with the at least one normal power consumption data subsequence based on the chronological order of time nodes to obtain the power consumption data sequence to be calibrated. The power consumption data sequence to be calibrated is then corrected again according to a preset second data correction strategy to finally obtain the calibration power consumption data of the power equipment at different time nodes.

[0112] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for calibrating electrical data, characterized in that, include: Obtain power consumption data of electrical equipment at different time points; The system determines whether transient impact power consumption data exists in each power consumption data according to a preset transient transition detection strategy. The transient impact power consumption data refers to the power consumption data generated by the power-consuming equipment when the state changes. If there is at least one transient impact power consumption data, then at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence are generated, and the at least one transient impact power consumption data subsequence is divided into sequences to obtain at least one transient impact power consumption data subsequence set. Associating a transient impact power consumption data subsequence with at least one normal power consumption data subsequence adjacent to the transient impact power consumption data subsequence, at least one association relationship is obtained; A target transient impact power consumption data subsequence is selected from a certain set of transient impact power consumption data subsequences. Based on the target correlation relationship containing the target transient impact power consumption data subsequence, a preset first data correction strategy is used to correct the target transient impact power consumption data subsequence to obtain a corrected transient impact power consumption data subsequence and a correction amount. Based on the aforementioned correction amount, other transient impact power consumption data subsequences in the aforementioned transient impact power consumption data subsequence set are directly corrected to obtain other corrected transient impact power consumption data subsequences; Each modified transient impact power consumption data subsequence is concatenated with the at least one normal power consumption data subsequence based on the chronological order of time nodes to obtain the power consumption data sequence to be calibrated. The power consumption data sequence to be calibrated is then corrected again according to a preset second data correction strategy to finally obtain the calibration power consumption data of the power equipment at different time nodes.

2. The method for calibrating electricity consumption data according to claim 1, characterized in that, The step of determining whether transient impact power consumption data exists in each power consumption data according to the preset transient transition detection strategy includes: Based on the chronological order of time nodes, the gradient of change between two adjacent electricity consumption data points is obtained sequentially to obtain a gradient of change sequence; The sliding window slides over the changing gradient sequence according to a preset sliding window. Each time the sliding window slides, it is determined whether the changing gradient in the sliding window is greater than a preset threshold. If none of the change gradients are greater than the preset threshold, then it is determined that there is no transient impact power consumption data in the various power consumption data. If a certain gradient change is greater than a preset threshold, the relative rate of change between each target gradient change and the certain gradient change is obtained, wherein the target gradient change is the gradient change that is ordered after the certain gradient change in the gradient change sequence. Determine whether the number of consecutive relative rates of change greater than a preset rate of change threshold exceeds the preset threshold. If the quantity is greater than a preset threshold, the electricity consumption data corresponding to a certain change gradient and the electricity consumption data corresponding to a relative change rate that is continuously greater than a preset change rate threshold are both defined as transient impact electricity consumption data; otherwise, they are not defined as transient impact electricity consumption data.

3. The method for calibrating electricity consumption data according to claim 1, characterized in that, The step involves correcting the target transient impact power consumption data subsequence using a preset first data correction strategy based on the target correlation relationship containing the target transient impact power consumption data subsequence, to obtain a corrected transient impact power consumption data subsequence, and a correction amount including: Based on the target association relationship, obtain the target normal power consumption data subsequence associated with the target transient impact power consumption data subsequence, and determine whether the number of the target normal power consumption data subsequence is greater than a preset number threshold; If the number exceeds a preset threshold, the sequence length of the target transient impact power consumption data subsequence is obtained. From the first target normal power consumption data subsequence and the second target normal power consumption data subsequence, segments of the first target normal power consumption data subsequence and segments of the second target normal power consumption data subsequence with the same sequence length and adjacent to the target transient impact power consumption data subsequence are extracted respectively. The first target normal power consumption data subsequence and the second target normal power consumption data subsequence are both normal power consumption data subsequences adjacent to the target transient impact power consumption data subsequence. Calculate the first average value of each normal power consumption data in the first target normal power consumption data sub-segment and the second target normal power consumption data sub-segment, and define the first average value as the first dynamic normal power consumption threshold; The system searches sequentially along a preset positive direction in the target transient impact power consumption data subsequence to find whether there exists a target transient impact power consumption data that is not greater than the first dynamic normal power consumption threshold, wherein the positive direction is the direction from the first to the last time node; When a certain target transient impact power consumption data exists, the search stops, the certain target transient impact power consumption data is defined as the cutoff target transient impact power consumption data, and the cutoff target transient impact power consumption data and the first power consumption data difference between the target transient impact power consumption data that is in the opposite direction to the cutoff target transient impact power consumption data are obtained. The difference in the first power consumption data is added to each of the reverse target transient impact power consumption data in the target transient impact power consumption data subsequence to obtain a certain corrected transient impact power consumption data subsequence corresponding to the target transient impact power consumption data subsequence. The reverse target transient impact power consumption data are target transient impact power consumption data with time nodes before the time node corresponding to the cutoff target transient impact power consumption data.

4. The method for calibrating electricity consumption data according to claim 3, characterized in that, The correction amount is the difference in the first electricity consumption data; The step of directly correcting other transient impact power consumption data subsequences in the set of transient impact power consumption data subsequences according to the certain correction amount to obtain other corrected transient impact power consumption data subsequences includes: Align the other transient impact power consumption data subsequences with the target transient impact power consumption data subsequences, and determine the other transient impact power consumption data in the other transient impact power consumption data subsequences that have the same sorting position as the cutoff target transient impact power consumption data; The correction amount is added to each cutoff transient impact power consumption data in the other transient impact power consumption data subsequence to obtain other corrected transient impact power consumption data subsequences corresponding to the other transient impact power consumption data subsequences, wherein each reverse other transient impact power consumption data is other transient impact power consumption data with time nodes before the time nodes corresponding to the cutoff other transient impact power consumption data.

5. The method for calibrating electricity consumption data according to claim 3, characterized in that, After sequentially searching along a preset positive direction in the target transient impact power consumption data subsequence for the existence of a target transient impact power consumption data that is not greater than the dynamic normal power consumption threshold, the method further includes: When there is no target transient impact power consumption data, the last target transient impact power consumption data in the target transient impact power consumption data subsequence along the positive direction is directly defined as the cutoff target transient impact power consumption data. Obtain the power consumption data of the cutoff target transient impact, and the second power consumption data difference between the target normal power consumption data adjacent to the cutoff target transient impact power consumption data in the first target normal power consumption data subsequence, wherein the time node corresponding to the first target normal power consumption data subsequence is after the time node corresponding to the target transient impact power consumption data subsequence; The difference in the second power consumption data is added to each of the target transient impact power consumption data in the target transient impact power consumption data subsequence to obtain a certain modified transient impact power consumption data subsequence corresponding to the target transient impact power consumption data subsequence.

6. The method for calibrating electricity consumption data according to claim 3, characterized in that, After determining whether the number of the target normal electricity consumption data subsequences is greater than a preset number threshold, the method further includes: If the number is not greater than a preset threshold, the sequence length of the target transient impact power consumption data subsequence is obtained, and a first target normal power consumption data subsequence segment with the same sequence length and adjacent to the target transient impact power consumption data subsequence is extracted from the first target normal power consumption data subsequence. The first target normal power consumption data subsequence is a normal power consumption data subsequence adjacent to the target transient impact power consumption data subsequence. Calculate the second average value of each normal power consumption data in the first target normal power consumption data sub-sequence segment, and define the second average value as the second dynamic normal power consumption threshold. Along the positive direction, sequentially search the target transient impact power consumption data subsequence to see if there is another target transient impact power consumption data that is not greater than the second dynamic normal power consumption threshold.

7. The method for calibrating electricity consumption data according to claim 1, characterized in that, After determining whether transient impact power consumption data exists in each power consumption data according to a preset transient transition detection strategy, the method further includes: If there is no transient impact power consumption data, the power consumption data are sorted according to the order of time nodes to directly obtain the second power consumption data sequence to be calibrated. The second power consumption data sequence to be calibrated is then corrected according to the second data correction strategy to finally obtain the calibration power consumption data of the power equipment at different time nodes.

8. The method for calibrating electricity consumption data according to claim 1, characterized in that, The step of partitioning the at least one transient impact power data subsequence to obtain at least one set of transient impact power data subsequences includes: Obtain the sequence length of the at least one transient impact power consumption data subsequence, wherein the sequence length is the number of transient impact power consumption data in the transient impact power consumption data subsequence; Clustering the transient impact power data subsequences of the same sequence length yields at least one set of transient impact power data subsequences.

9. The method for calibrating electricity consumption data according to claim 1, characterized in that, The step of further correcting the power consumption data sequence to be calibrated according to a preset second data correction strategy, and finally obtaining the calibration power consumption data of the electrical equipment at different time points, includes: The current operating parameters of the electrical equipment are input into a preset power consumption data calibration model, and the power consumption data calibration model outputs a correction coefficient corresponding to the current operating parameters. The power consumption data to be calibrated in the power consumption data sequence to be calibrated is corrected according to the correction coefficient to obtain the calibration power consumption data of the power equipment at different time points.

10. An electricity data calibration system, characterized in that, include: The acquisition module is configured to acquire power consumption data of electrical equipment at different time points; The judgment module is configured to determine whether there is transient impact power consumption data in each power consumption data according to a preset transient transition detection strategy, wherein the transient impact power consumption data is the power consumption data generated by the power consumption equipment when the state is switched; The partitioning module is configured to generate at least one normal power consumption data subsequence and at least one transient impact power consumption data subsequence if there is at least one transient impact power consumption data. The module then partitions the at least one transient impact power consumption data subsequence to obtain at least one set of transient impact power consumption data subsequences. The association module is configured to associate a transient impact power consumption data subsequence with at least one normal power consumption data subsequence adjacent to the transient impact power consumption data subsequence to obtain at least one association relationship; The first correction module is configured to select a target transient impact power consumption data subsequence from a certain set of transient impact power consumption data subsequences, and correct the target transient impact power consumption data subsequence according to the target association relationship containing the target transient impact power consumption data subsequence, using a preset first data correction strategy to obtain a corrected transient impact power consumption data subsequence and a certain correction amount. The second correction module is configured to directly correct other transient impact power consumption data subsequences in the set of transient impact power consumption data subsequences according to a certain correction amount, so as to obtain other corrected transient impact power consumption data subsequences; The third correction module is configured to concatenate each corrected transient impact power consumption data subsequence with the at least one normal power consumption data subsequence based on the chronological order of time nodes to obtain a power consumption data sequence to be calibrated, and then correct the power consumption data sequence to be calibrated again according to a preset second data correction strategy, so as to finally obtain the calibration power consumption data of the power equipment at different time nodes.

Images