A battery car load identification method based on multi-feature fusion and a related device thereof

Abstract

The application discloses a battery car load identification method based on multi-feature fusion and a related device thereof. Indoor charging data of a battery car is divided into three stages of opening, continuous charging and ending to extract transient features and steady features of the battery car. Basic power consumption data of users in a transformer area is obtained, and transient features and steady features of each to-be-identified electrical equipment are extracted according to the basic power consumption data. Transient feature similarity and steady feature similarity are calculated according to the transient features and the steady features of each to-be-identified electrical equipment and the transient features and the steady features of the battery car, and fusion features of each to-be-identified electrical equipment and fusion features of the battery car are obtained based on the transient feature similarity and the steady feature similarity. Whether the to-be-identified electrical equipment is the battery car is identified according to the fusion features of the battery car and the fusion features of each to-be-identified electrical equipment. The multi-feature fusion identification can effectively reduce the misidentification of the battery car, improve the battery car identification precision and user safety.

Description

Technical Field

[0001] This application relates to the field of load identification technology, and in particular to a method and related apparatus for identifying the load of electric vehicles based on multi-feature fusion. Background Technology

[0002] With the development of smart electricity use, monitoring methods in various industries are gradually moving towards intelligence and digitalization. However, on the low-voltage power distribution side, illegal charging of electric bicycles indoors still occurs frequently, causing multiple fire incidents and posing significant safety hazards. Therefore, providing a method for identifying the load of electric bicycles is a technical problem that those skilled in the art need to solve. Summary of the Invention

[0003] This application provides a method and related apparatus for identifying the load of electric bicycles based on multi-feature fusion, which is used to improve user safety.

[0004] In view of this, the first aspect of this application provides a method for identifying the load of electric bicycles based on multi-feature fusion, including:

[0005] Acquire indoor charging data of electric vehicles, and divide the indoor charging data of electric vehicles into three stages: start-up, continuous charging and end-up, in order to extract transient and steady-state features of electric vehicles.

[0006] Acquire basic electricity consumption data of users within the distribution area, and extract transient and steady-state characteristics of each electrical device to be identified based on the basic electricity consumption data;

[0007] Based on the transient and steady-state characteristics of each of the electrical devices to be identified and the transient and steady-state characteristics of the electric vehicle, the transient feature similarity and steady-state feature similarity are calculated, and the fused features of each of the electrical devices to be identified and the electric vehicle are obtained based on the transient feature similarity and the steady-state feature similarity.

[0008] The fusion features of the electric vehicle and the fusion features of each of the electrical devices to be identified are used to determine whether the electrical device to be identified is an electric vehicle.

[0009] Optionally, acquiring indoor charging data for electric bicycles includes:

[0010] Voltage and current data of electric vehicles are extracted based on smart meters;

[0011] Based on the voltage and current data of the electric vehicle, the active power, reactive power, and odd harmonic current of the electric vehicle are extracted to obtain the indoor charging data of the electric vehicle.

[0012] Optionally, the transient characteristics of the electric vehicle include the change value of the start time, the change value of the turn-off time, the change value of the active power during start-up, the change value of the active power during turn-off, the change value of the reactive power during start-up, the change value of the reactive power during turn-off, the change value of the odd harmonic current during start-up, and the change value of the odd harmonic current during turn-off.

[0013] The formula for calculating the transient characteristics of the electric vehicle is:

[0014] ΔT on =t0-t1,ΔT off =t4-t3;

[0015] ΔP on =P(t1)-P(t0),ΔP off =P(t4)-P(t3);

[0016] ΔQ on =Q(t1)-Q(t0),ΔQ off =Q(t4)-Q(t3);

[0017] ΔI h,on =I h (t1)-I h (t0),ΔI h,off =I h (t4)-I h (t3);

[0018] In the formula, ΔT on ΔT off These represent the changes in on time and off time, respectively, ΔP on ΔP off These represent the changes in active power when the circuit is open and the changes in active power when the circuit is closed, respectively; ΔQ on ΔQ off These represent the changes in reactive power when the circuit is open and the changes in reactive power when the circuit is closed, respectively; ΔI h,on ΔI h,off These represent the changes in odd-order harmonic current during startup and shutdown, respectively. t0 is the time before the device starts up, t1 is the time when the device ends startup, t3 is the time before the device ends charging, and t4 is the time after the device ends charging. P(t0) represents the active power before the device starts up, P(t1) represents the active power after the device starts up, P(t3) represents the active power before the device ends charging, and P(t4) represents the active power after the device ends charging. Q(t0) represents the reactive power before the device starts up, Q(t1) represents the reactive power after the device starts up, Q(t3) represents the reactive power before the device ends charging, and Q(t4) represents the reactive power after the device ends charging. h (t0) represents the magnitude of the odd-order harmonic current before the device is turned on, I h(t1) represents the magnitude of the odd harmonic current after the device is turned on, I h (t3) represents the magnitude of the odd harmonic current before the device finishes charging, I h (t4) represents the magnitude of the odd harmonic current after the device finishes charging.

[0019] Optionally, the steady-state characteristics of the electric vehicle include the root mean square of the active power waveform, the peak value of the active power waveform, the root mean square of the reactive power waveform, the peak value of the reactive power waveform, the root mean square of the odd harmonic current waveform, and the peak value of the odd harmonic current.

[0020] The formula for calculating the steady-state characteristics of the electric vehicle is as follows:

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027] In the formula, P rms P is the root mean square of the active power waveform. CF Q is the peak power coefficient. rms The root mean square of the reactive power waveform, Q CF I is the reactive power crest factor. h,rms For the root mean square of the odd harmonic current waveform, I h,CF P is the peak value of the odd harmonic current. off This is a set of active power data during the slow decline phase, where P(t2) is the active power at the moment trickle charging begins, P(t3) is the active power before the device finishes charging, and Q... off This is a dataset of reactive power during the slow decline phase, where Q(t2) represents the reactive power at the start of trickle charging, Q(t3) represents the reactive power at the end of trickle charging, and I... h,off For the odd harmonic current data set during the slow descent phase, I h (t2) represents the magnitude of the odd-order harmonic current at the moment trickle charging begins, I h (t3) represents the odd harmonic current at the end of trickle charging.

[0028] Optionally, the step of calculating transient feature similarity and steady-state feature similarity based on the transient and steady-state features of each of the electrical devices to be identified and the electric vehicle, and obtaining the fused features of each of the electrical devices to be identified and the electric vehicle based on the transient feature similarity and the steady-state feature similarity, includes:

[0029] The similarity between the transient features of each electrical device to be identified and the electric vehicle is calculated based on the transient features of each electrical device to be identified and the transient features of the electric vehicle. The similarity between the steady-state features of each electrical device to be identified and the electric vehicle is also calculated based on the steady-state features of each electrical device to be identified and the steady-state features of the electric vehicle.

[0030] The transient feature weights and steady-state feature weights are calculated based on the transient feature similarity and steady-state feature similarity between each of the electrical devices to be identified and the electric vehicle.

[0031] Based on the transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities, feature fusion is performed on the transient and steady-state features of each electrical device to be identified, as well as the transient and steady-state features of the electric vehicle, to obtain the fused features of each electrical device to be identified and the fused features of the electric vehicle.

[0032] Optionally, the process of obtaining the fusion features is as follows:

[0033] The transient and steady-state characteristics of the target device are standardized, wherein the target device is the electrical equipment to be identified or the electric vehicle.

[0034] The transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities are substituted into a preset feature fusion formula to perform feature fusion on the transient and steady-state features of the target device, resulting in the fused features of the target device. The preset feature fusion formula is as follows:

[0035]

[0036]

[0037] In the formula, U represents the fusion feature of the target device, and X represents the fusion feature of the target device. i,nrom Sim is the i-th feature after standardization of the target device. i f is the similarity weight coefficient for the i-th feature. j,i e is the j-th element of the i-th feature of the electric vehicle. j,i Let X be the j-th element of the i-th feature of the electrical equipment to be identified, m be the dimension of the i-th feature, and n be the number of features of the target equipment; when X i,nromWhen it is a transient feature, β i For transient feature weights; when X i,nrom When it is a steady-state characteristic, β i These are the steady-state feature weights.

[0038] A second aspect of this application provides a battery vehicle load identification device based on multi-feature fusion, comprising:

[0039] The first feature extraction unit is used to acquire indoor charging data of electric vehicles and divide the indoor charging data of electric vehicles into three stages: start-up, continuous charging and end-up, in order to extract transient and steady-state features of the electric vehicles.

[0040] The second feature extraction unit is used to obtain basic electricity consumption data of users in the transformer area, and extract transient and steady-state features of each electrical device to be identified based on the basic electricity consumption data.

[0041] The feature fusion unit is used to calculate transient feature similarity and steady-state feature similarity based on the transient and steady-state features of each of the electrical devices to be identified and the transient and steady-state features of the electric vehicle, and to obtain the fused features of each of the electrical devices to be identified and the fused features of the electric vehicle based on the transient feature similarity and the steady-state feature similarity.

[0042] The identification unit is used to identify whether the electrical device to be identified is an electric vehicle based on the fusion features of the electric vehicle and the fusion features of each electrical device to be identified.

[0043] Optionally, the feature fusion unit is specifically used for:

[0044] The similarity between the transient features of each electrical device to be identified and the electric vehicle is calculated based on the transient features of each electrical device to be identified and the transient features of the electric vehicle. The similarity between the steady-state features of each electrical device to be identified and the electric vehicle is also calculated based on the steady-state features of each electrical device to be identified and the steady-state features of the electric vehicle.

[0045] The transient feature weights and steady-state feature weights are calculated based on the transient feature similarity and steady-state feature similarity between each of the electrical devices to be identified and the electric vehicle.

[0046] Based on the transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities, feature fusion is performed on the transient and steady-state features of each electrical device to be identified, as well as the transient and steady-state features of the electric vehicle, to obtain the fused features of each electrical device to be identified and the fused features of the electric vehicle.

[0047] A third aspect of this application provides a battery vehicle load identification device based on multi-feature fusion, the device including a processor and a memory;

[0048] The memory is used to store program code and transmit the program code to the processor;

[0049] The processor is used to execute any one of the electric vehicle load identification methods based on multi-feature fusion as described in the first aspect, according to the instructions in the program code.

[0050] The fourth aspect of this application provides a computer-readable storage medium, characterized in that the computer-readable storage medium is used to store program code, which, when executed by a processor, implements the electric vehicle load identification method based on multi-feature fusion as described in the first aspect.

[0051] As can be seen from the above technical solutions, this application has the following advantages:

[0052] This application provides a method for identifying electric vehicle load based on multi-feature fusion, comprising: acquiring indoor charging data of electric vehicles, dividing the indoor charging data into three stages—start, continuous charging, and end—to extract transient and steady-state features of the electric vehicles; acquiring basic electricity consumption data of users within the transformer substation area, and extracting transient and steady-state features of each electrical device to be identified based on the basic electricity consumption data; calculating transient feature similarity and steady-state feature similarity based on the transient and steady-state features of each electrical device to be identified and the transient and steady-state features of the electric vehicles, and obtaining fused features of each electrical device to be identified and fused features of the electric vehicles based on the transient feature similarity and steady-state feature similarity; and identifying whether the electrical device to be identified is an electric vehicle based on the fused features of the electric vehicles and the fused features of each electrical device to be identified.

[0053] In this application, the waveform of the indoor charging data of the electric vehicle is divided into three stages: start-up, continuous charging and end-up, and corresponding transient and steady-state features are extracted. The transient and steady-state features are then fused with the multi-layer similarity principle to obtain the fused features of the electric vehicle and the electrical device to be identified. Based on the fused features, the charging status of the electric vehicle can be effectively identified, which can play a preventive role and improve user safety. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A flowchart illustrating a method for identifying the load of an electric vehicle based on multi-feature fusion, provided in an embodiment of this application;

[0056] Figure 2 A schematic diagram of active power, reactive power, and odd harmonics for charging an electric vehicle provided in an embodiment of this application;

[0057] Figure 3 This is a schematic diagram of the entire charging cycle of an electric vehicle provided in an embodiment of this application;

[0058] Figure 4 A waveform comparison diagram of the electric vehicle and the lighting lamp provided in the embodiments of this application;

[0059] Figure 5 This is a schematic diagram of a battery vehicle load identification device based on multi-feature fusion, provided in an embodiment of this application. Detailed Implementation

[0060] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0061] For easier understanding, please refer to Figure 1 This application provides a method for identifying the load of electric bicycles based on multi-feature fusion, including:

[0062] Step 101: Obtain indoor charging data for electric vehicles. Divide the indoor charging data for electric vehicles into three stages: start-up, continuous charging, and end-up, in order to extract the transient and steady-state characteristics of the electric vehicles.

[0063] Voltage and current data of electric vehicles are extracted using smart meters. Based on this data, the active power, reactive power, and odd harmonic current of the electric vehicle are extracted to obtain indoor charging data. The obtained active power p, reactive power q, and odd harmonic current i... h The waveform of the sequence is as follows Figure 2 As shown. Considering that the electric vehicle charging device is mainly a power electronic rectifier, it will generate a large number of odd harmonics. The window size is set to 1 second. The active power p, reactive power q, and odd harmonic current i within the window are calculated. h As the feature subspace for identification, its calculation formula is as follows:

[0064]

[0065]

[0066]

[0067] In the formula, Let k be the phase angle of the voltage harmonic. Let u be the phase angle of the k-th current harmonic. (0) i (0) These are the fundamental RMS values ​​of voltage and current, respectively, u (k) i (k) These represent the effective values ​​of the k-th harmonic voltage and the k-th harmonic current, respectively, N. k This represents the total number of data items within the window.

[0068] like Figure 3 As shown, the charging waveform of an electric vehicle can be divided into three stages: start-up, continuous charging, and end-up. The start-up stage is a transient event, typically lasting less than 0.2 seconds. The continuous charging time is related to the current battery level, with little change in overall power and harmonics. Finally, in the end-up stage, active power and harmonics continuously decrease, with a duration generally greater than 20 minutes. Dividing the indoor charging data of electric vehicles into these three stages (start-up, continuous charging, and end-up) allows for the extraction of transient and steady-state characteristics of the electric vehicle.

[0069] The transient characteristics of electric vehicles include changes in start-up time, change in turn-off time, change in active power during start-up, change in active power during turn-off, change in reactive power during start-up, change in reactive power during turn-off, change in odd harmonic current during start-up, and change in odd harmonic current during turn-off.

[0070] The formula for calculating the transient characteristics of an electric bicycle is:

[0071] ΔT on =t0-t1,ΔT off =t4-t3;

[0072] ΔP on =P(t1)-P(t0),ΔP off =P(t4)-P(t3);

[0073] ΔQ on =Q(t1)-Q(t0),ΔQ off =Q(t4)-Q(t3);

[0074] ΔI h,on =I h (t1)-I h (t0),ΔI h,off =I h (t4)-I h (t3);

[0075] In the formula, ΔT on ΔT offThese represent the changes in on time and off time, respectively, ΔP on ΔP off These represent the changes in active power when the circuit is open and the changes in active power when the circuit is closed, respectively; ΔQ on ΔQ off These represent the changes in reactive power when the circuit is open and the changes in reactive power when the circuit is closed, respectively; ΔI h,on ΔI h,off These represent the changes in odd-order harmonic current during startup and shutdown, respectively. t0 is the time before the device starts up, t1 is the time when the device ends startup, t3 is the time before the device ends charging, and t4 is the time after the device ends charging. P(t0) represents the active power before the device starts up, P(t1) represents the active power after the device starts up, P(t3) represents the active power before the device ends charging, and P(t4) represents the active power after the device ends charging. Q(t0) represents the reactive power before the device starts up, Q(t1) represents the reactive power after the device starts up, Q(t3) represents the reactive power before the device ends charging, and Q(t4) represents the reactive power after the device ends charging. h (t0) represents the magnitude of the odd-order harmonic current before the device is turned on, I h (t1) represents the magnitude of the odd harmonic current after the device is turned on, I h (t3) represents the magnitude of the odd harmonic current before the device finishes charging, I h (t4) represents the magnitude of the odd harmonic current after the device finishes charging.

[0076] The steady-state characteristics of electric vehicles include the root mean square of active power waveform, the peak value of active power waveform, the root mean square of reactive power waveform, the peak value of reactive power waveform, the root mean square of odd harmonic current waveform, and the peak value of odd harmonic current waveform.

[0077] Steady-state characteristics are mainly waveform morphology characteristics, such as Figure 2 As shown, active power, reactive power, and odd harmonics all exhibit slow changes during operation. The formula for calculating the steady-state characteristics of the electric vehicle is:

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084] In the formula, P rmsP is the root mean square of the active power waveform. CF Q is the peak power coefficient. rms The root mean square of the reactive power waveform, Q CF I is the reactive power crest factor. h,rms For the root mean square of the odd harmonic current waveform, I h,CF P is the peak value of the odd harmonic current. off This is a set of active power data during the slow decline phase, where P(t2) is the active power at the moment trickle charging begins, P(t3) is the active power before the device finishes charging, and Q... off This is a dataset of reactive power during the slow decline phase, where Q(t2) represents the reactive power at the start of trickle charging, Q(t3) represents the reactive power at the end of trickle charging, and I... h,off For the odd harmonic current data set during the slow descent phase, I h (t2) represents the magnitude of the odd-order harmonic current at the moment trickle charging begins, I h (t3) represents the odd harmonic current at the end of trickle charging.

[0085] Step 102: Obtain basic electricity consumption data of users within the distribution area, and extract transient and steady-state characteristics of each electrical device to be identified based on the basic electricity consumption data.

[0086] Basic electricity consumption data of users within the distribution area was collected, and transient and steady-state features of each electrical device to be identified were extracted based on the basic electricity consumption data. The specific feature extraction process is similar to that of electric bicycles, and will not be described in detail here. A distribution area user dataset Ω was constructed based on the extracted features of each electrical device to be identified. The dataset format is shown in Table 1. Figure 4 Taking electric vehicles and lighting equipment as examples, the extracted transient and steady-state characteristics of electric vehicles and lighting equipment are shown in Table 1.

[0087] Table 1

[0088]

[0089] Step 103: Calculate the transient feature similarity and steady-state feature similarity based on the transient and steady-state features of each electrical device to be identified and the electric vehicle, and obtain the fusion features of each electrical device to be identified and the electric vehicle based on the transient feature similarity and steady-state feature similarity.

[0090] Considering that other household appliances and electric bicycles have certain similarities in waveform, such as Figure 4As shown, the transient features of its lights and electric bicycles are almost identical, with only the steady-state features possessing a certain degree of distinguishability. Therefore, the multi-dimensional features are fused based on the principle of similarity. Taking the electric bicycle and lights as examples, let's assume the feature set of feature 1 of the electric bicycle is F1 = {f 1,1 …f n,1 The feature set of feature 1 corresponding to the lighting lamp is E1 = {e} 1,1 …e n,1 The format corresponding to E1 and F1 is {first feature value of the first event, ..., first feature value of the nth event}. The similarity Sim1 between feature 1 of the electric vehicle and feature 1 corresponding to the lighting is calculated as follows:

[0091]

[0092] Therefore, the similarity Sim between the electric vehicle and the i-th feature of the electrical device to be identified is... i The calculation formula can be expressed as:

[0093]

[0094] In the formula, f j,i e is the j-th element of the i-th feature of the electric vehicle. j,i Let m be the j-th element of the i-th feature of the electrical equipment to be identified, and m be the dimension of the i-th feature.

[0095] The transient feature similarity between each electrical device to be identified and the electric vehicle is calculated based on their transient features and the transient features of the electric vehicle. The steady-state feature similarity between each electrical device to be identified and the electric vehicle is calculated based on their steady-state features and the steady-state features of the electric vehicle. The calculation process for transient and steady-state feature similarity can be found in the Sim algorithm described above. i Calculation formula.

[0096] The transient feature weights and steady-state feature weights are calculated based on the transient feature similarity and steady-state feature similarity of each electrical device to be identified and the electric vehicle. In addition to considering the similarity between features within the same category, the weight W between the fused transient and steady-state features is also considered; otherwise, the classifier may bias towards either transient or steady-state features. For example, ... Figure 4 As shown, ΔP on In this feature analysis, the LED is 110W and the electric scooter is 250W. Although they are similar in trajectory images, there is still a certain degree of differentiation in their transient feature values. After calculating the transient feature similarity and steady-state feature similarity between each electrical device and the electric scooter using the similarity formula mentioned in the previous steps, the transient feature similarity between each electrical device and the electric scooter is averaged to obtain the average transient feature similarity Sim between each electrical device and the electric scooter. tThe steady-state feature similarity between each electrical device to be identified and the electric vehicle is averaged to obtain the average steady-state feature similarity Sim between each electrical device to be identified and the electric vehicle. s Then, based on the average transient feature similarity Sim between each electrical device to be identified and the electric vehicle, t Sim, average steady-state feature similarity s Calculate the transient feature weights W for each electrical device and electric vehicle to be identified. t Steady-state characteristic weights W s The calculation process is as follows:

[0097] W s =Sim s / (Sim t +Sim s );

[0098] W t =Sim t / (Sim t +Sim s );

[0099] Based on transient feature weights, steady-state feature weights, transient feature similarity, and steady-state feature similarity, feature fusion is performed on the transient and steady-state features of each electrical device to be identified, as well as the transient and steady-state features of the electric vehicle, to obtain the fused features of each electrical device to be identified and the fused features of the electric vehicle. Specifically, the transient and steady-state features of the target device (electrical device to be identified or electric vehicle) are standardized, i.e.:

[0100]

[0101] In the formula, X i Let X be the i-th feature of the target device, where max() is the function to find the maximum value and min() is the function to find the minimum value. i,nrom The i-th feature of the target device after standardization.

[0102] Substituting the transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities into a preset feature fusion formula, we perform feature fusion on the transient and steady-state features of the target device to obtain the fused features of the target device. The preset feature fusion formula is as follows:

[0103]

[0104] In the formula, U represents the fusion feature of the target device, and X represents the fusion feature of the target device. i,nrom Sim is the i-th feature after standardization of the target device. i Let X be the similarity weight coefficient for the i-th feature, and n be the number of features of the target device; when X... i,nrom When it is a transient feature, βi The transient feature weight W t When X i,nrom When it is a steady-state characteristic, β i The steady-state feature weight W s .

[0105] Step 104: Identify whether the electrical device to be identified is an electric vehicle based on the fusion characteristics of the electric vehicle and the fusion characteristics of each electrical device to be identified.

[0106] The identification of whether an electrical device to be identified is an electric vehicle can be based on the fusion features of the electric vehicle and the fusion features of each electrical device to be identified. The fusion features of the electric vehicle and the fusion features of each electrical device to be identified can be input into multiple classifiers for identification. If all classifiers identify a certain electrical device to be identified as an electric vehicle, the identification result of the electrical device to be identified as an electric vehicle will be output.

[0107] In this embodiment, the waveform of the indoor charging data of the electric vehicle is divided into three stages: start-up, continuous charging, and end, to extract corresponding transient and steady-state features. The transient and steady-state features are then fused using a multi-layer similarity principle to obtain a fused feature between the electric vehicle and the electrical device to be identified. Based on this fused feature, the charging status of the electric vehicle can be effectively identified, preventing potential problems and improving user safety. This embodiment applies non-intrusive load identification technology to effectively solve the problem of monitoring indoor charging of electric vehicles. By decomposing the load based on data collected at the user's home, the charging status of the electric vehicle can be effectively identified, preventing potential problems. Considering that in real-world scenarios, electric vehicle chargers are power electronic devices, and given the current proliferation of power electronic devices with various similar waveforms, this application uses multi-feature fusion for identification, which can effectively reduce false identification of electric vehicles and improve the accuracy of electric vehicle identification.

[0108] The above is an embodiment of a battery vehicle load identification method based on multi-feature fusion provided by this application. The following is an embodiment of a battery vehicle load identification device based on multi-feature fusion provided by this application.

[0109] Please refer to Figure 5 This application provides a battery vehicle load identification device based on multi-feature fusion, comprising:

[0110] The first feature extraction unit is used to acquire indoor charging data of electric vehicles and divide the indoor charging data of electric vehicles into three stages: start-up, continuous charging and end-up, in order to extract transient and steady-state features of electric vehicles.

[0111] The second feature extraction unit is used to obtain the basic electricity consumption data of users in the transformer area, and extract the transient and steady-state features of each electrical device to be identified based on the basic electricity consumption data.

[0112] The feature fusion unit is used to calculate transient feature similarity and steady-state feature similarity based on the transient and steady-state features of each electrical device to be identified and the transient and steady-state features of the electric vehicle, and to obtain the fused features of each electrical device to be identified and the fused features of the electric vehicle based on the transient feature similarity and steady-state feature similarity.

[0113] The identification unit is used to identify whether the electrical device to be identified is an electric vehicle based on the fusion characteristics of the electric vehicle and the fusion characteristics of each electrical device to be identified.

[0114] As a further improvement, the feature fusion unit is specifically used for:

[0115] The similarity between the transient features of each electrical device to be identified and the electric vehicle is calculated based on the transient features of each electrical device to be identified and the transient features of the electric vehicle. The similarity between the steady-state features of each electrical device to be identified and the electric vehicle is calculated based on the steady-state features of each electrical device to be identified and the steady-state features of the electric vehicle.

[0116] The transient feature weights and steady-state feature weights are calculated based on the transient feature similarity and steady-state feature similarity between each electrical device to be identified and the electric vehicle.

[0117] Based on transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities, feature fusion is performed on the transient and steady-state features of each electrical device to be identified, as well as the transient and steady-state features of the electric vehicle, to obtain the fused features of each electrical device to be identified and the fused features of the electric vehicle.

[0118] In this embodiment, the waveform of the indoor charging data of the electric vehicle is divided into three stages: start-up, continuous charging, and end, to extract corresponding transient and steady-state features. The transient and steady-state features are then fused using a multi-layer similarity principle to obtain a fused feature between the electric vehicle and the electrical device to be identified. Based on this fused feature, the charging status of the electric vehicle can be effectively identified, preventing potential problems and improving user safety. This embodiment applies non-intrusive load identification technology to effectively solve the problem of monitoring indoor charging of electric vehicles. By decomposing the load based on data collected at the user's home, the charging status of the electric vehicle can be effectively identified, preventing potential problems. Considering that in real-world scenarios, electric vehicle chargers are power electronic devices, and given the current proliferation of power electronic devices with various similar waveforms, this application uses multi-feature fusion for identification, which can effectively reduce false identification of electric vehicles and improve the accuracy of electric vehicle identification.

[0119] This application embodiment also provides an electric vehicle load identification device based on multi-feature fusion, the device including a processor and a memory;

[0120] The memory is used to store program code and transfer the program code to the processor;

[0121] The processor is used to execute the electric vehicle load identification method based on multi-feature fusion in the aforementioned method embodiments according to the instructions in the program code.

[0122] This application also provides a computer-readable storage medium for storing program code, which, when executed by a processor, implements the electric vehicle load identification method based on multi-feature fusion in the aforementioned method embodiments.

[0123] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0124] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0125] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0126] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0127] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0128] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0129] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of this application through a computer device (which may be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0130] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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. Such 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 this application.

Claims

1. A method for identifying the load of electric bicycles based on multi-feature fusion, characterized in that, include: Acquire indoor charging data of electric vehicles, and divide the indoor charging data of electric vehicles into three stages: start-up, continuous charging and end-up, in order to extract transient and steady-state features of electric vehicles. Acquire basic electricity consumption data of users within the distribution area, and extract transient and steady-state characteristics of each electrical device to be identified based on the basic electricity consumption data; Based on the transient and steady-state characteristics of each of the electrical devices to be identified and the transient and steady-state characteristics of the electric vehicle, transient feature similarity and steady-state feature similarity are calculated. Then, based on the transient feature similarity and steady-state feature similarity, fused features of each of the electrical devices to be identified and fused features of the electric vehicle are obtained, including: The similarity between the transient features of each electrical device to be identified and the electric vehicle is calculated based on the transient features of each electrical device to be identified and the transient features of the electric vehicle. The similarity between the steady-state features of each electrical device to be identified and the electric vehicle is also calculated based on the steady-state features of each electrical device to be identified and the steady-state features of the electric vehicle. The transient feature weights and steady-state feature weights are calculated based on the transient feature similarity and steady-state feature similarity between each of the electrical devices to be identified and the electric vehicle. Based on the transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities, feature fusion is performed on the transient and steady-state features of each electrical device to be identified, as well as the transient and steady-state features of the electric vehicle, to obtain the fused features of each electrical device to be identified and the fused features of the electric vehicle. The process of obtaining the fusion features is as follows: The transient and steady-state characteristics of the target device are standardized, wherein the target device is the electrical equipment to be identified or the electric vehicle. The transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities are substituted into a preset feature fusion formula to perform feature fusion on the transient and steady-state features of the target device, resulting in the fused features of the target device. The preset feature fusion formula is as follows: ； ； In the formula, U represents the fusion feature of the target device, and X represents the fusion feature of the target device. i,nrom Sim is the i-th feature after standardization of the target device. i f is the similarity weight coefficient for the i-th feature. j,i e is the j-th element of the i-th feature of the electric vehicle. j,i Let X be the j-th element of the i-th feature of the electrical equipment to be identified, m be the dimension of the i-th feature, and n be the number of features of the target equipment; when X i,nrom When it is a transient feature, For transient feature weights; when X i,nrom When it is a steady-state characteristic, Weights for steady-state features; The fusion features of the electric vehicle and the fusion features of each of the electrical devices to be identified are used to determine whether the electrical device to be identified is an electric vehicle.

2. The method for identifying the load of electric bicycles based on multi-feature fusion according to claim 1, characterized in that, The acquisition of indoor charging data for electric bicycles includes: Voltage and current data of electric vehicles are extracted based on smart meters; Based on the voltage and current data of the electric vehicle, the active power, reactive power, and odd harmonic current of the electric vehicle are extracted to obtain the indoor charging data of the electric vehicle.

3. The method for identifying the load of electric bicycles based on multi-feature fusion according to claim 2, characterized in that, The transient characteristics of the electric vehicle include the change value of the start time, the change value of the turn-off time, the change value of the active power when start-up, the change value of the active power when turn-off, the change value of the reactive power when start-up, the change value of the reactive power when turn-off, the change value of the odd harmonic current when start-up, and the change value of the odd harmonic current when turn-off. The formula for calculating the transient characteristics of the electric vehicle is: ； ； ； ； In the formula, ΔT on ΔT off These represent the changes in on time and off time, respectively, ΔP on ΔP off These represent the changes in active power when the circuit is open and the changes in active power when the circuit is closed, respectively; ΔQ on ΔQ off These represent the changes in reactive power when the circuit is open and the changes in reactive power when the circuit is closed, respectively; ΔI h,on ΔI h,off These represent the changes in odd-order harmonic current during startup and shutdown, respectively. t0 is the time before the device starts up, t1 is the time when the device ends startup, t3 is the time before the device ends charging, and t4 is the time after the device ends charging. P(t0) represents the active power before the device starts up, P(t1) represents the active power after the device starts up, P(t3) represents the active power before the device ends charging, and P(t4) represents the active power after the device ends charging. Q(t0) represents the reactive power before the device starts up, Q(t1) represents the reactive power after the device starts up, Q(t3) represents the reactive power before the device ends charging, and Q(t4) represents the reactive power after the device ends charging. h (t0) represents the magnitude of the odd-order harmonic current before the device is turned on, I h (t1) represents the magnitude of the odd harmonic current after the device is turned on, I h (t3) represents the magnitude of the odd harmonic current before the device finishes charging, I h (t4) represents the magnitude of the odd harmonic current after the device finishes charging.

4. The method for identifying the load of an electric vehicle based on multi-feature fusion according to claim 2, characterized in that, The steady-state characteristics of the electric vehicle include the root mean square of the active power waveform, the peak value of the active power waveform, the root mean square of the reactive power waveform, the peak value of the reactive power waveform, the root mean square of the odd harmonic current waveform, and the peak value of the odd harmonic current. The formula for calculating the steady-state characteristics of the electric vehicle is as follows: ； ， ； ； ， ； ； ， ； In the formula, P rms P is the root mean square of the active power waveform. CF Q is the peak power coefficient. rms The root mean square of the reactive power waveform, Q CF I is the reactive power crest factor. h,rms For the root mean square of the odd harmonic current waveform, I h,CF P is the peak value of the odd harmonic current. off This is a dataset of active power during the slow decline phase, where P(t2) represents the active power at the start of trickle charging, P(t3) represents the active power before the device finishes charging, τ is the time index, P(τ) represents the active power of the electric vehicle at time τ, and Q... off This is a dataset of reactive power during the slow decline phase. Q(t2) represents the reactive power at the start of trickle charging, Q(t3) represents the reactive power at the end of trickle charging, and Q(τ) represents the reactive power of the electric vehicle at time τ. h,off For the odd harmonic current data set during the slow descent phase, I h (t2) represents the magnitude of the odd-order harmonic current at the moment trickle charging begins, I h (t3) represents the odd harmonic current at the end of trickle charging, I h (τ) represents the magnitude of the odd harmonic current of the electric vehicle at time τ.

5. A battery-powered vehicle load identification device based on multi-feature fusion, characterized in that, include: The first feature extraction unit is used to acquire indoor charging data of electric vehicles and divide the indoor charging data of electric vehicles into three stages: start-up, continuous charging and end-up, in order to extract transient and steady-state features of the electric vehicles. The second feature extraction unit is used to obtain basic electricity consumption data of users in the transformer area, and extract transient and steady-state features of each electrical device to be identified based on the basic electricity consumption data. The feature fusion unit is used to calculate transient feature similarity and steady-state feature similarity based on the transient and steady-state features of each of the electrical devices to be identified and the transient and steady-state features of the electric vehicle, and to obtain the fused features of each of the electrical devices to be identified and the fused features of the electric vehicle based on the transient feature similarity and the steady-state feature similarity. The feature fusion unit is specifically used for: The similarity between the transient features of each electrical device to be identified and the electric vehicle is calculated based on the transient features of each electrical device to be identified and the transient features of the electric vehicle. The similarity between the steady-state features of each electrical device to be identified and the electric vehicle is also calculated based on the steady-state features of each electrical device to be identified and the steady-state features of the electric vehicle. The transient feature weights and steady-state feature weights are calculated based on the transient feature similarity and steady-state feature similarity between each of the electrical devices to be identified and the electric vehicle. Based on the transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities, feature fusion is performed on the transient and steady-state features of each electrical device to be identified, as well as the transient and steady-state features of the electric vehicle, to obtain the fused features of each electrical device to be identified and the fused features of the electric vehicle. The process of obtaining the fusion features is as follows: The transient and steady-state characteristics of the target device are standardized, wherein the target device is the electrical equipment to be identified or the electric vehicle. The transient feature weights, steady-state feature weights, transient feature similarities, and steady-state feature similarities are substituted into a preset feature fusion formula to perform feature fusion on the transient and steady-state features of the target device, resulting in the fused features of the target device. The preset feature fusion formula is as follows: ； ； In the formula, U represents the fusion feature of the target device, and X represents the fusion feature of the target device. i,nrom Sim is the i-th feature after standardization of the target device. i f is the similarity weight coefficient for the i-th feature. j,i e is the j-th element of the i-th feature of the electric vehicle. j,i Let X be the j-th element of the i-th feature of the electrical equipment to be identified, m be the dimension of the i-th feature, and n be the number of features of the target equipment; when X i,nrom When it is a transient feature, For transient feature weights; When X i,nrom When it is a steady-state characteristic, Weights for steady-state features; The identification unit is used to identify whether the electrical device to be identified is an electric vehicle based on the fusion features of the electric vehicle and the fusion features of each electrical device to be identified.

6. A battery vehicle load identification device based on multi-feature fusion, characterized in that, The device includes a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the electric vehicle load identification method based on multi-feature fusion as described in any one of the claims 1-4 according to the instructions in the program code.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program code, which, when executed by a processor, implements the electric vehicle load identification method based on multi-feature fusion as described in any one of claims 1-4.

Images