A VI waveform feature extraction method for user electricity consumption behavior perception analysis

Abstract

The present invention discloses a VI waveform feature extraction method for user electricity consumption behavior perception analysis, which relates to the field of power grid operation and maintenance. At present, the electricity consumption behavior feature extraction technology at home and abroad has poor versatility from the perspective of load characteristics, extracts harmonics as feature data, and cannot further distinguish variable speed drive loads. The present invention collects current-voltage trajectory curves, takes into account the closed area of ​​the trajectory curve, the average curve distortion, the number of curve self-intersection points, and the five features of the current-voltage trajectory curve, the slope of the middle section of the average curve near zero point, and the clockwise or counterclockwise average curvature of the VI trajectory, and performs mathematical feature extraction on them to obtain a feature data set for user behavior perception analysis. The feature extraction speed is fast and the implementation is simple. Multiple features jointly participate in user behavior perception with high accuracy. The proposed method can provide data support for user load identification, and has good economic benefits and practical value.

Description

Technical Field

[0001] The present invention relates to the field of power grid operation and maintenance, and in particular to a VI waveform feature extraction method for user power consumption behavior perception analysis. Background Art

[0002] As society's awareness of ecological civilization continues to grow, people's concern about energy consumption is also increasing. Consumers' demands for power supply quality are constantly increasing. Traditional extensive management can no longer meet household users' demands for intelligent and streamlined power supply. Smart electricity consumption has become an inevitable trend. Smart electricity consumption is a key component of building a smart grid and the core of an interactive service system. Its key technologies are mainly reflected in the advanced metering infrastructure (AMI) standards, systems, and terminal technologies. User electricity behavior perception and analysis technology, as one of the most important components of AMI, is the first step in realizing the smart grid.

[0003] When it comes to sensing user electricity usage behavior, quickly and efficiently extracting characteristic data is a crucial step. Currently, various methods exist for extracting electricity usage characteristics, both domestically and internationally, based on load characteristics, including instantaneous power, harmonics, and noise. Extracting instantaneous power as characteristic data offers robustness but is unsuitable for permanently operating or continuously changing loads. Extracting voltage noise as characteristic data suffers from poor interference immunity and versatility. Extracting harmonics as characteristic data fails to differentiate variable-speed drive loads. Summary of the Invention

[0004] The technical problem to be solved and the technical task to be addressed by this invention are to improve and enhance existing technical solutions and provide a method for extracting VI waveform features for user electricity usage behavior perception analysis, thereby achieving the goal of quickly and comprehensively extracting feature data. To this end, this invention adopts the following technical solution.

[0005] A VI waveform feature extraction method for user electricity consumption behavior perception analysis includes the following steps:

[0006] 1) Total electricity load data collection;

[0007] The voltage and current data of a single electrical appliance are collected through an oscilloscope, and the VI trajectory curve is drawn;

[0008] 2) Data preprocessing;

[0009] Data preprocessing includes current and voltage standardization; current and voltage standardization is achieved by dividing the current and voltage signals by their root mean square respectively; the standardized current and voltage data are plotted with voltage as the horizontal axis and current as the vertical axis to form a current and voltage VI trajectory curve;

[0010] 3) VI feature extraction;

[0011] After preprocessing the collected data, the preprocessed voltage and current signals are used to make a VI trajectory area and extract characteristic values ​​from the trajectory area;

[0012] 4) Acquire feature data sets and store them to form a feature library to provide data support for user load identification;

[0013] The feature data set includes: closed area, distortion of the average curve, number of self-intersection points, slope of the middle section of the average curve, and circulation direction;

[0014] When load identification is required, supervised or unsupervised classification algorithms are used to obtain VI waveform feature data and analyze the user's electricity consumption behavior perception.

[0015] This technical solution collects current-voltage trajectory curves, taking into account five features: the enclosed area of ​​the trajectory curve, the average curve distortion, the number of self-intersection points, and mathematically extracting features from the current-voltage trajectory curve, the slope near the zero point of the average curve's mid-segment, and the average clockwise or counterclockwise curvature of the V-V trajectory. This yields a feature dataset for user behavior perception analysis. Feature extraction is fast and simple to implement, and multiple features contribute to user behavior perception with high accuracy. This provides data support for user load identification, demonstrating both economic efficiency and practical value.

[0016] As a preferred technical means: in step 2), the current and voltage normalization calculation formula is:

[0017]

[0018] v norm(n) =v n / V rms (1≤n≤N) (2)

[0019]

[0020] i norm(n) =i n / I rms (1≤n≤N) (4)

[0021] Where: N is the total number of points in the collected trajectory curve, V rms is the RMS voltage in the VI trace, V norm(n)The voltage v = v n The standard voltage at rms is the RMS current in the VI trajectory curve, i norm(n) For current i=i n The standard current when

[0022] As a preferred technical means: in step 3), VI waveform feature extraction includes:

[0023] 301) Closed Area:

[0024]

[0025] Where: A norm is the area covered by the VI trajectory curve, V L is the minimum voltage in the curve, V R is the maximum voltage in the curve, v n is any voltage value between the minimum and maximum voltage, i nu is the voltage v in the trajectory curve n The larger of the two corresponding current values, i nd is the voltage v in the trajectory curve n The smaller of the two corresponding current values;

[0026] 302) Distortion of the average curve:

[0027]

[0028] Where: m n v n The average current of the trajectory curve at ;

[0029] Connect the first and last coordinates of the VI trajectory curve to obtain a straight line, whose equation is shown in formula (7):

[0030]

[0031] Where: I L V L The current value of the trajectory curve, I R V R The current value of the trajectory curve at ;

[0032] DMC=max(|m n -i n ′|) (8)

[0033] Where: DMC is the distortion of the average curve;

[0034] 303) Number of self-intersection points:

[0035] Get the number of self-intersection points N in the VI trajectory curve ip ;

[0036] 304) Slope of the middle section of the average curve:

[0037] The slope of the middle section of the average curve near zero point can characterize the power-electronic characteristics of the electrical equipment, which is used to represent the tangent of the angle between the tangent line of the average curve near zero point and the V axis. The formula is shown in (9):

[0038]

[0039] Where: θ is the angle between the tangent line of the average curve of the VI trajectory curve near zero point and the V axis, The average curve is close to zero Current value is the point to the right of zero The current value;

[0040] 305) Circulation direction:

[0041] "Circulation direction" refers to the average counterclockwise or clockwise curvature of the VI trace. Clockwise curvature refers to the total capacitive load behavior, while counterclockwise curvature refers to the total inductive load behavior. The curvature is calculated using the following model:

[0042]

[0043]

[0044] Where: k n is the curvature of any point of the VI trajectory curve, i n 、i n+1 、i n+2 v=v n 、v=v n+1 、v=v n+2 The corresponding current value.

[0045] As a preferred technical means: in step 4), the VI waveform feature data set is obtained as follows: X={A norm ,DMC,N ip , tanθ, K}.

[0046] As a preferred technical means: when performing load identification, the graphic features in the load VI curve trajectory are used as identification samples, and identification labels are defined for each load. An identifier that can identify different load devices is trained through the feature library, and the load device is identified through the identifier.

[0047] Beneficial effects: For the same or similar power loads, the current and voltage trajectories of this technical solution are similar. Due to the different electrical power consumption, the VI trajectories will form different shapes. By extracting features from the VI trajectories and classifying them, user behavior perception can be achieved. This technical solution draws a VI trajectory curve based on the collected voltage and current data. Through the VI trajectory curve, combined with the feature extraction model proposed by the present invention, a feature data set is finally obtained to provide data support for user power consumption behavior perception. The feature extraction speed is fast and the implementation is simple. Multiple features participate in user behavior perception together with high accuracy. It can provide data support for user load identification and has good economic benefits and practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of the present invention.

[0049] Figure 2 It is a flow chart of user load identification of the present invention. DETAILED DESCRIPTION

[0050] The technical solution of the present invention is further described in detail below with reference to the accompanying drawings.

[0051] like Figure 1 As shown, the present invention includes the following steps:

[0052] 1) Total electricity load data collection;

[0053] The voltage and current data of a single electrical appliance are collected through an oscilloscope, and the VI trajectory curve is drawn;

[0054] 2) Data preprocessing;

[0055] Data preprocessing includes current and voltage standardization; current and voltage standardization is achieved by dividing the current and voltage signals by their root mean square respectively; the standardized current and voltage data are plotted with voltage as the horizontal axis and current as the vertical axis to form a current and voltage VI trajectory curve;

[0056] The current and voltage normalization calculation formula is:

[0057]

[0058] v norm(n) =v n / V rms (1≤n≤N) (2)

[0059]

[0060] i norm(n) =i n / I rms(1≤n≤N) (4)

[0061] Where: N is the total number of points in the collected trajectory curve, V rms is the RMS voltage in the VI trace, V norm(n) The voltage v = v n The standard voltage at rms is the RMS current in the VI trajectory curve, i norm(n) For current i=i n The standard current when

[0062] 3) VI feature extraction;

[0063] After preprocessing the collected data, the preprocessed voltage and current signals are used to make a VI trajectory area and extract characteristic values ​​from the trajectory area;

[0064] VI waveform feature extraction includes:

[0065] 301) Closed Area:

[0066]

[0067] Where: A norm is the area covered by the VI trajectory curve, V L is the minimum voltage in the curve, V R is the maximum voltage in the curve, v n is any voltage value between the minimum and maximum voltage, i nu is the voltage v in the trajectory curve n The larger of the two corresponding current values, i nd is the voltage v in the trajectory curve n The smaller of the two corresponding current values;

[0068] 302) Distortion of the average curve:

[0069]

[0070] Where: m n v n The average current of the trajectory curve at ;

[0071] Connect the first and last coordinates of the VI trajectory curve to obtain a straight line, whose equation is shown in formula (7):

[0072]

[0073] Where: I L V L The current value of the trajectory curve, I R VR The current value of the trajectory curve at ;

[0074] DMC=max(|m n -i n ′|) (8)

[0075] Where: DMC is the distortion of the average curve;

[0076] 303) Number of self-intersection points:

[0077] Get the number of self-intersection points N in the VI trajectory curve ip ;

[0078] 304) Slope of the middle section of the average curve:

[0079] The slope of the middle section of the average curve near zero point can characterize the power-electronic characteristics of the electrical equipment, which is used to represent the tangent of the angle between the tangent line of the average curve near zero point and the V axis. The formula is shown in (9):

[0080]

[0081] Where: θ is the angle between the tangent line of the average curve of the VI trajectory curve near zero point and the V axis, The average curve is close to zero Current value is the point to the right of zero The current value;

[0082] 305) Circulation direction:

[0083] "Circulation direction" refers to the average counterclockwise or clockwise curvature of the VI trace. Clockwise curvature refers to the total capacitive load behavior, while counterclockwise curvature refers to the total inductive load behavior. The curvature is calculated using the following model:

[0084]

[0085]

[0086] Where: k n is the curvature of any point of the VI trajectory curve, i n 、i n+1 、i n+2 v=v n 、v=v n+1 、v=v n+2 The corresponding current value.

[0087] 4) Acquire feature data sets and store them to form a feature library to provide data support for user load identification;

[0088] The characteristic data set includes: closed area, average curve distortion, number of self-intersection points, slope of the middle section of the average curve, and circulation direction; the VI waveform characteristic data set is obtained as follows: X = {A norm ,DMC,N ip , tanθ, K}.

[0089] When load identification is required, supervised or unsupervised classification algorithms are used to obtain VI waveform feature data and analyze the user's electricity consumption behavior perception.

[0090] When performing load identification, the graphic features in the load VI curve trajectory are used as identification samples, and identification labels are defined for each load. An identifier that can identify different load devices is trained through the feature library, and the load device is identified through the identifier.

[0091] like Figure 2 As shown, user load identification includes the following steps:

[0092] 401) Acquire data, the data including training data, label data, weak identifier and iteration number; the training data is the graphical feature data in the load VI curve trajectory; the training data and label data are input into the training data set to form a training data set, wherein the training data set is:

[0093] T={(x1,y1),(x2,y2),…,(x n ,y n )} (12)

[0094] Where: T is the training data set; x i is the load characteristic; y i The label corresponding to the load characteristic indicates the working status of the equipment. When it is -1, it means it is in the stopped state, and when it is 1, it means it is working.

[0095] 402) Initialize the weak identifier weight distribution;

[0096]

[0097] Where: D1 is the weight distribution set of the initial training sample, w i,n is the nth training weight of the i-th weak recognizer.

[0098] 403) Data training to obtain a weak identifier;

[0099] 404) Determine whether the recognition error rate is less than or equal to the set threshold; if so, proceed to the next step; if not, return to step 403); wherein the recognition error rate calculation formula is:

[0100]

[0101] Where: e m is the recognition error rate, G m (x) is the mth weak identifier.

[0102] 405) Calculate the weight of the weak identifier in the strong identifier;

[0103]

[0104] Where: α m is the weight of the mth weak identifier in the strong identifier.

[0105] 406) Update weak identifier weights;

[0106]

[0107]

[0108] Where: z m is the normalization factor (to make the probability distribution of the samples sum to 1).

[0109] 407) obtaining a strong identifier for identifying the working status of the corresponding load device according to the weight of the weak identifier;

[0110]

[0111] Where: F is a strong identifier, x is the newly added sample data.

[0112] 408) When identification is required, the acquired new identification sample data is input into the strong identifier, and the strong identifier identifies the identification sample data and outputs the load type of the corresponding device.

[0113] In this embodiment, an identifier F(x) capable of identifying different load devices can be trained through the existing feature library. After inputting the newly added sample data x, the identifier can output the corresponding device working status.

[0114] This technical solution collects current-voltage trajectory curves, taking into account five features: the enclosed area of ​​the trajectory curve, the average curve distortion, the number of self-intersection points, and mathematically extracting features from the current-voltage trajectory curve, the slope near the zero point of the average curve's mid-segment, and the average clockwise or counterclockwise curvature of the V-V trajectory. This yields a feature dataset for user behavior perception analysis. Feature extraction is fast and simple to implement, and multiple features contribute to user behavior perception with high accuracy. This provides data support for user load identification, demonstrating both economic efficiency and practical value.

[0115] above Figure 1The VI waveform feature extraction method for user electricity consumption behavior perception analysis shown is a specific embodiment of the present invention, which has reflected the essential characteristics and progress of the present invention. According to actual usage needs and under the guidance of the present invention, equivalent modifications in shape, structure, etc. can be made to it, which are all within the scope of protection of this solution.

Claims

1. A VI waveform feature extraction method for user electricity consumption behavior perception analysis, characterized by The following steps are involved: 1) Total electricity load data collection; The voltage and current data of a single electrical appliance are collected through an oscilloscope, and the VI trajectory curve is drawn; 2) Data preprocessing; Data preprocessing includes current and voltage standardization; current and voltage standardization is achieved by dividing the current and voltage signals by their root mean square respectively; the standardized current and voltage data are plotted with voltage as the horizontal axis and current as the vertical axis to form a current and voltage VI trajectory curve; 3) VI feature extraction; After preprocessing the collected data, the preprocessed voltage and current signals are used to make the VI trajectory area and extract the characteristic values ​​of the trajectory area; 4) Obtaining feature data sets and storing them to form a feature library to provide data support for user load identification; The feature data set includes: closed area, distortion of the average curve, number of self-intersection points, slope of the middle section of the average curve, and circulation direction; When load identification is required, supervised or unsupervised classification algorithms are used to obtain VI waveform feature data and analyze user power consumption behavior perception; In step 3), VI waveform feature extraction includes: 301) Closed Area: （5） Where: is the area covered by the VI trajectory curve, is the minimum voltage in the curve, is the maximum voltage in the curve, is any voltage value between the minimum and maximum voltage values. is the voltage in the trajectory curve The larger of the two corresponding current values, is the voltage in the trajectory curve The smaller of the two corresponding current values; 302) Distortion of the average curve: （6） Where: for The average current of the trajectory curve at ; Connect the first and last coordinates of the VI trajectory curve to obtain a straight line, whose equation is shown in formula (7): （7） Where: for The current value of the trajectory curve at for The current value of the trajectory curve at ; （8） Where: DMC is the distortion of the average curve; 303) Number of self-intersection points: Get the number of self-intersection points in the VI trajectory curve ; 304) Slope of the middle section of the average curve: The slope of the middle section of the average curve near zero point can characterize the power-electronic characteristics of the electrical equipment, which is used to represent the tangent of the angle between the tangent line of the average curve near zero point and the V axis. The formula is shown in (9): （9） Where: is the angle between the tangent line of the average curve near zero point of the VI trajectory curve and the V axis, is the average curve near zero point ( ) current value ( ), is a point close to the right of zero ( ) current value; 305) Circulation direction: "Circulation direction" refers to the average counterclockwise or clockwise curvature of the VI trace. Clockwise curvature refers to the total capacitive load behavior, while counterclockwise curvature refers to the total inductive load behavior. The curvature is calculated using the following model: （10） （11） Where: is the curvature of any point of the VI trajectory curve, 、 、 They are 、 、 The corresponding current value.

2. The method for extracting VI waveform features for user electricity consumption behavior perception analysis according to claim 1, characterized in that: In step 2), the current and voltage normalization calculation formula is: （1） （2） （3） （4） Where: is the total number of points in the collected trajectory curve, is the RMS voltage in the VI trace, is voltage The standard voltage at is the RMS current in the VI trajectory curve, Current The standard current when .

3. The method for extracting VI waveform features for user electricity consumption behavior perception analysis according to claim 2, characterized in that: In step 4), the VI waveform feature data set is obtained as: .

4. The method for extracting VI waveform features for user electricity consumption behavior perception analysis according to claim 1, characterized in that: When performing load identification, the graphic features in the load VI curve trajectory are used as identification samples, and identification labels are defined for each load. An identifier that can identify different load devices is trained through the feature library, and the load device is identified through the identifier.

Images