CN114236301B - A non-intrusive industrial load event detection method and system - Google Patents

Abstract

The invention relates to a non-intrusive industrial load event detection method and system, and relates to the technical field of intelligent power consumption management. Intercept N adjacent first data segments from the electrical data set, and calculate the average power difference between any two adjacent first data segments; according to the average power difference corresponding to multiple adjacent two first data segments value to determine whether a load state change event occurs; build a second power consumption data set according to the detected load state change event; intercept M adjacent second data segments from the second power consumption data set, and calculate any adjacent The difference in the number of intersections between two second data segments; whether a load pattern change event occurs is determined according to the difference in the number of intersections between a plurality of adjacent second data segments, thereby realizing industrial load event type detection.

Description

A non-intrusive industrial load event detection method and system

technical field

The invention relates to the technical field of intelligent power consumption management, in particular to a non-invasive industrial load event detection method and system.

Background technique

The carbon peaking and carbon neutrality goals put forward new requirements for low-carbon energy conservation for industrial development, prompting industrial users to continuously improve the level of production energy efficiency. Industrial production consumes a lot of electricity, and the power consumption data of its load can largely reflect the actual situation of the factory's production operations. Therefore, in industrial production, it is necessary to monitor the load power consumption information, and then obtain the user's production situation and restore the load action, which can help industrial users improve production energy efficiency and achieve low-carbon transformation.

The traditional intrusive load detection method directly collects the independent data of the operating load, and then performs detection based on the collected data. Although the obtained load information is true and reliable, it requires a large number of equipment and the cost is high, and its installation and maintenance are very important to the operation of the load. It has a great influence and is not suitable for industrial load event detection.

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a non-intrusive industrial load event detection method and system, so as to effectively identify the types of load events in industrial production.

To achieve the above objects, the present invention provides a non-intrusive industrial load event detection method, the method comprising:

The electricity consumption data is collected synchronously from the power metering place of the industrial user through the non-intrusive collection device, forming the first electricity consumption data set;

Intercept N adjacent first data segments from the first power consumption data set, and calculate the average power difference between any two adjacent first data segments; wherein, N is a positive integer greater than 2;

Determine whether a load state change event occurs according to the average power difference corresponding to a plurality of adjacent first data segments;

According to the detected load state change event, construct a second power consumption data set;

Intercept M adjacent second data segments from the second power consumption data set, and calculate the difference in the number of intersections between any two adjacent second data segments; wherein, M is a positive integer greater than 2;

Whether a load pattern change event occurs is determined according to the difference in the number of intersections between a plurality of adjacent second data segments.

Optionally, the method further includes:

Determine the load action corresponding to the load state change event and/or the load mode change event.

Optionally, the intercepting N adjacent first data segments from the first power consumption data set, and calculating the average power difference between any two adjacent first data segments, specifically includes:

Take the collection point corresponding to the voltage value from negative to positive as the initial point of the voltage cycle;

Taking the initial point as a critical point, intercepting N adjacent first data segments from the first electricity consumption data set; each of the first data segments includes electricity consumption data collected in N ₁ voltage cycles; wherein, N ₁ is a positive integer greater than 1;

Calculate the load power consumption of each voltage period in each of the first data sections according to the power consumption data collected in each voltage period in each of the first data sections;

Calculate the average electric power value of the load in each of the first data sections according to the load electric power of each voltage period in each of the first data sections;

The average power difference between the two adjacent first data segments is calculated according to the load average electric power values of the two adjacent first data segments.

Optionally, the determining whether a load state change event occurs according to the average power difference corresponding to a plurality of adjacent first data segments specifically includes:

Step S31: Calculate the variance of the load power consumption corresponding to each of the first data sections according to the load power consumption of each voltage period in each of the first data sections and the average load power consumption value of the load in each of the first data sections;

；如果△P _X 大于

，则说明事件为负荷状态变动事件，并执行“步骤S33”；如果△P _X 小于或等于

，则令X=X+1，判断X是否小于或等于N，如果X小于或等于N，则执行“步骤S32”，如果X大于N，则结束；其中，

均为大于0小于1的有理数，P' _X 表示第X个第一数据段对应的负荷用电功率方差，P _X-1表示第X-1个第一数据段内负荷平均用电功率值；Step S32: Determine whether the average power difference ΔP _X is greater than

; if △ P _X is greater than

, then the event is a load state change event, and "step S33" is executed; if ΔP _X is less than or equal to

, then let X=X+1, judge whether X is less than or equal to N, if X is less than or equal to N, execute "step S32", if X is greater than N, end; where,

All are rational numbers greater than 0 and less than 1, P' _X represents the variance of the load power consumption corresponding to the Xth first data segment, and P _X-1 represents the load average power consumption value in the X-1th first data segment;

Step S33: Determine whether the first event ending condition is met. If the first event ending condition is met, the load state change event has ended. If the first event ending condition is not met, "step S33" is continued.

Optionally, the first event end condition is specifically:

Among them, X ₁ and X ₂ represent the serial numbers of two non-adjacent first data segments, X ₁ is less than X ₂ , P' _n represents the load power variance corresponding to the n-th first data segment, and P _n-1 represents the first data segment. The average power consumption value of the load in the n-1 first data segment, K ₁ represents a set integer, ΔP _n represents the average power difference between the n-th first data segment and the n-1-th first data segment value.

Optionally, the determining whether a load pattern change event occurs according to the difference in the number of intersections between multiple adjacent second data segments specifically includes:

；如果交点数量差值△S _Y大于

，则说明事件为负荷模式变动事件，并执行“步骤S62”；如果交点数量差值△S _Y小于或等于

，则令Y=Y+1，判断Y是否小于或等于M，如果Y小于或等于M，则返回“步骤S61”，如果Y大于M，则结束；其中，

为大于0小于1的有理数，S _Y-1表示第Y-1个第二数据段对应的交点个数；Step S61: Determine whether the difference in the number of intersections ΔS _Y is greater than

;If the difference in the number of intersections △ S _Y is greater than

, then the event is a load pattern change event, and "step S62" is executed; if the difference in the number of intersections △ S _Y is less than or equal to

, then let Y=Y+1, judge whether Y is less than or equal to M, if Y is less than or equal to M, return to "step S61", if Y is greater than M, end; where,

is a rational number greater than 0 and less than 1, and S _Y-1 represents the number of intersections corresponding to the second data segment Y-1;

Step S62: Determine whether the second event ending condition is met. If the second event ending condition is met, the load mode change event has ended; if the second event ending condition is not met, continue to execute "Step S62".

Optionally, the end condition of the second event is specifically:

Among them, Y ₁ and Y ₂ represent the serial numbers of two non-adjacent second data segments, Y ₁ is less than Y ₂ , Sn _-1 represents the number of intersections corresponding to the n-1th second data segment , and K ₂ represents the set is a fixed integer, and ΔS _n represents the difference in the number of intersections between the n-th second data segment and the n-1-th second data segment.

The present invention also provides a non-intrusive industrial load event detection system, the system comprising:

The first power consumption data set building module is used for synchronously collecting power consumption data from the power metering place of the industrial user through the non-intrusive collection device to form the first power consumption data set;

an average power difference calculation module, configured to intercept N adjacent first data segments from the first power consumption data set, and calculate the average power difference between any two adjacent first data segments; wherein , N is a positive integer greater than 2;

a load state change event determination module, configured to determine whether a load state change event occurs according to the average power difference corresponding to a plurality of adjacent first data segments;

The second power consumption data set building module is used to construct a second power consumption data set according to the detected load state change event;

a calculation module for the difference in the number of intersection points, configured to intercept M adjacent second data segments from the second power consumption data set, and calculate the difference in the number of intersection points between any two adjacent second data segments; wherein , M is a positive integer greater than 2;

The load pattern change event determination module is configured to determine whether a load pattern change event occurs according to the difference in the number of intersections between a plurality of adjacent second data segments.

Optionally, the average power difference calculation module specifically includes:

The initial point determination unit is used to take the collection point corresponding to the voltage value from negative to positive as the initial point of the voltage cycle;

A first intercepting unit, configured to take an initial point as a critical point, intercepting N adjacent first data segments from the first power consumption data set; each of the _first data segments includes N1 voltage period collections The electricity consumption data of ; wherein, N ₁ is a positive integer greater than 1;

a first load electric power calculation unit, configured to calculate the load electric power of each voltage period in each of the first data sections according to the electric power consumption data collected in each voltage period in each of the first data sections;

a first load average electric power value calculation unit, configured to calculate the average load electric power value of the load in each of the first data segments according to the load electric power of each voltage period in each of the first data segments;

The average power difference calculation unit is configured to calculate the average power difference between the two adjacent first data segments according to the load average electric power values of the two adjacent first data segments.

Optionally, the load state change event determination module specifically includes:

A load electric power variance calculation unit, configured to calculate the load corresponding to each of the first data segments according to the load electric power of each voltage period in each of the first data segments and the average load electric power value of the load in each of the first data segments power consumption variance;

；如果△P _X 大于

，则说明事件为负荷状态变动事件，并执行“第二判断单元”；如果△P _X 小于或等于

，则令X=X+1，判断X是否小于或等于N，如果X小于或等于N，则执行“第一判断单元”，如果X大于N，则结束；其中，

均为大于0小于1的有理数，P' _X 表示第X个第一数据段对应的负荷用电功率方差，P _X-1 表示第X-1个第一数据段内负荷平均用电功率值；The first judgment unit is used to judge whether the average power difference Δ P _X is greater than

; if △ P _X is greater than

, then the event is a load state change event, and the "second judgment unit" is executed; if △ P _X is less than or equal to

, then let X=X+1, judge whether X is less than or equal to N, if X is less than or equal to N, execute the "first judgment unit", if X is greater than N, end; where,

All are rational numbers greater than 0 and less than 1, P' _X represents the variance of the load power consumption corresponding to the Xth first data segment, and P _X-1 represents the load average power consumption value in the X-1th first data segment;

The second judging unit is used to judge whether the first event ending condition is met. If the first event ending condition is met, it means that the load state change event has ended. ".

According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

The invention synchronously collects power consumption data from the power metering place of industrial users through a non-intrusive collection device, and then effectively detects the type of industrial load events based on the collected power consumption data, which not only solves the problems of complex data collection and high cost in traditional collection , and can also detect industrial load events.

Description of drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings required in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some of the present invention. In the embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative labor.

Fig. 1 is the flow chart of the non-intrusive industrial load event detection method of the present invention;

2 is a structural diagram of a non-intrusive industrial load event detection system according to the present invention;

3 is a schematic diagram of the non-intrusive industrial load event detection architecture of the present invention;

Fig. ₄ is the current and voltage waveforms that the present invention detects the state change event at time t1 and its extraction;

FIG. 5 shows the mode change event detected at time t ₂ and the power waveform extracted by the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

The purpose of the present invention is to provide a non-intrusive industrial load event detection method and system, so as to effectively identify the types of load events in industrial production.

In order to make the above objects, features and advantages of the present invention more clearly understood, the present invention will be described in further detail below with reference to the accompanying drawings and specific embodiments.

Example 1

In the following embodiments, in order to distinguish the difference between the data segments in step S2 and step S5, the data segment intercepted in step S2 is referred to as the first data segment, and the data segment intercepted in step S5 is referred to as the second data segment. In addition, the load event in the present invention includes a load state change event and a load mode change event; the slow and large change of the power signal caused by the load switching action is recorded as a load state change event; The event that the constant power signal changes regularly is recorded as the load pattern change event.

As shown in FIG. 1 , the present invention discloses a non-invasive industrial load event detection method, which includes:

Step S1 : synchronously collect power consumption data from the power metering site of the industrial user through a non-intrusive collection device to form a first power consumption data set.

Step S2: Intercept N adjacent first data segments from the first power consumption data set, and calculate the average power difference between any two adjacent first data segments; wherein, N is greater than 2. positive integer.

Step S3: Determine whether a load state change event occurs according to the average power difference corresponding to a plurality of adjacent two first data segments.

Step S4: Construct a second power consumption data set according to the detected load state change event.

Step S5: Intercept M adjacent second data segments from the second power consumption data set, and calculate the difference in the number of intersections between any two adjacent second data segments; where M is greater than 2. positive integer.

Step S6: Determine whether a load pattern change event occurs according to the difference in the number of intersections between a plurality of adjacent second data segments.

Each step is discussed in detail below:

Step S1 : synchronously collect power consumption data from the power metering site of the industrial user through a non-intrusive collection device to form a first power consumption data set; the power consumption data includes: voltage and current. In this embodiment, the power consumption data is collected by means of high frequency collection.

Step S2: Intercept N adjacent first data segments from the first power consumption data set, and calculate the average power difference between any two adjacent first data segments, specifically including:

Step S21: take the collection point corresponding to the voltage value from negative to positive as the initial point of the voltage cycle, and the specific judgment formula is:

(1)

(1)

Among them, U _A,z represents the voltage collected at the zth sampling point of phase A.

Step S22: Taking the initial point as the critical point, intercepting N adjacent first data segments from the first electricity consumption data set; each of the first data segments includes electricity consumption data collected in N ₁ voltage cycles ; In this embodiment, N ₁ takes 500.

Step S23: Calculate the load power consumption of each voltage period in each of the first data sections according to the power consumption data collected in each voltage period in each of the first data sections. The specific calculation formula is:

(2)

(2)

Among them, P _X,i represents the load power consumption of the i-th voltage cycle in the X-th first data segment, T is the total number of points collected in each voltage cycle, and U _A,z represents the z-th sampling point of the A phase. Voltage, I _A,z represents the current collected at the zth sampling point of phase A, X∈[1,N], i∈[1, N ₁ ].

Step S24: Calculate the average electric power value of the load in each of the first data sections according to the load electric power of each voltage period in each of the first data sections, and the specific calculation formula is:

(3)

(3)

Among them, P _X represents the average power consumption value of the load in the X-th first data segment, N ₁ represents the total number of voltage cycles included in each first data segment, and P _X,i represents the X-th first data segment. The load electric power of the i-th voltage cycle.

Step S25: Calculate the average power difference between two adjacent first data segments according to the load average power consumption values of the two adjacent first data segments, and the specific calculation formula is:

(4)

(4)

Among them, P _X represents the average power consumption value of the load in the X-th first data segment, P _X-1 represents the load average power consumption value in the X-1-th first data segment, and △ P _X represents the X-th first data segment. The average power difference between the segment and the X-1 th first data segment.

Step S3: Determine whether a load state change event occurs according to the average power difference corresponding to a plurality of adjacent two first data segments, which specifically includes:

Step S31: Calculate the variance of the load power consumption corresponding to each of the first data sections according to the load power consumption of each voltage period in each of the first data sections and the average load power consumption value in each of the first data sections. The specific formula is: for:

(5)

(5)

Among them, P'x represents the load power variance corresponding to the X-th first data segment, P _X represents the load average power consumption value in the X-th first data segment, N ₁ represents the total number of voltage cycles, P _{X, i} represents the load electric power of the i-th voltage cycle in the X-th first data segment.

；如果△P _X 大于

，则说明事件为负荷状态变动事件，并执行“步骤S33”；如果△P _X 小于或等于

，则令X=X+1，判断X是否小于或等于N，如果X小于或等于N，则执行“步骤S32”，如果X大于N，则结束；其中，

均为大于0小于1的有理数，P' _X 表示第X个第一数据段对应的负荷用电功率方差，P _X-1 表示第X-1个第一数据段内负荷平均用电功率值；本实施例中，

为0.02，

为0.3。Step S32: Determine whether the average power difference ΔP _X is greater than

; if △ P _X is greater than

, then the event is a load state change event, and "step S33" is executed; if ΔP _X is less than or equal to

, then let X=X+1, judge whether X is less than or equal to N, if X is less than or equal to N, execute "step S32", if X is greater than N, end; where,

All are rational numbers greater than 0 and less than 1, P' _X represents the variance of the load power consumption corresponding to the Xth first data segment, and P _X-1 represents the load average power consumption value in the X-1th first data segment; this implementation In the example,

is 0.02,

is 0.3.

Step S33: Determine whether the first event ending condition is met. If the first event ending condition is met, the load state change event has ended. If the first event ending condition is not met, "step S33" is continued.

The specific conditions for the end of the first event are:

(6)

(6)

均为大于0小于1的有理数，P' _n表示第n个第一数据段对应的负荷用电功率方差，P_n-1表示第n-1个第一数据段内负荷平均用电功率值，K ₁表示设定的整数，△P _n表示第n个第一数据段与第n-1个第一数据段之间的平均功率差值。本实施例中，K ₁为20。Wherein, X ₁ and X ₂ represent the sequence numbers of two non-adjacent first data segments, X ₁ is less than X ₂ ,

are rational numbers greater than 0 and less than 1, P' _n represents the variance of load power consumption corresponding to the nth first data segment, P _n-1 represents the average load power consumption value of the n-1th first data segment, K ₁ represents a set integer, and ΔP _n represents the average power difference between the n-th first data segment and the n-1-th first data segment. In this embodiment, K ₁ is 20.

Step S4: Construct a second power consumption data set according to the detected load state change event. The second power consumption data set is current data and voltage data between two adjacent load state change events.

Step S5: Intercept M adjacent second data segments from the second power consumption data set, and calculate the difference in the number of intersections between any two adjacent second data segments, specifically including:

Step S51: Taking the initial point as a critical point, intercepting M adjacent second data segments from the second electricity consumption data set; each of the second data segments includes electricity consumption data collected in N ₂ voltage cycles ; In this embodiment, N ₂ takes 300.

Step S52: Calculate the load electric power of each voltage period in each second data section according to the power consumption data collected in each voltage period in each of the second data sections, and the specific calculation formula is:

(7)

(7)

Among them, P _Y,j represents the load power consumption of the jth voltage cycle in the Yth second data segment, T is the total number of points collected in each voltage cycle, U _A,z represents the zth sampling point of the A phase. Voltage, I _A,z represents the current collected at the zth sampling point of phase A, Y∈[1,M], j∈[1, N ₂ ].

Step S53: Calculate the average electric power value of the load in each of the second data sections according to the electric power of the load in each voltage period in each of the second data sections. The specific calculation formula is:

(8)

(8)

Among them, P _Y represents the average power consumption value of the load in the Y-th second data segment, N ₂ represents the total number of voltage cycles included in each second data segment, and P _{Y, j} represents the Y-th second data segment. The load electric power of the jth voltage cycle.

Step S54: Calculate the number of intersection points corresponding to each of the second data segments according to the load power consumption of each voltage period in each of the second data segments and the average load power consumption value of each of the second data segments. The specific formula is: :

(9)

(9)

表示设定参数，P _Y表示第Y个第二数据段内负荷平均用电功率值，N ₂表示每个第二数据段内包含电压周期的总个数，P _Y，j 表示第Y个第二数据段内第j个电压周期的负荷用电功率，S _Y表示第Y个第二数据段对应的交点个数，本实施例中交点个数为负荷用电功率曲线与功率平均值的交点的个数，

取5W。in,

Represents the setting parameters, P _Y represents the average power consumption value of the load in the Y-th second data segment, N ₂ represents the total number of voltage cycles included in each second data segment, P _{Y, j} represents the Y-th second data segment The load electric power of the jth voltage cycle in the data segment, S _Y represents the number of intersection points corresponding to the Yth second data segment, and the number of intersection points in this embodiment is the number of intersection points between the load electric power curve and the average power value ,

Take 5W.

Step S55: Calculate the difference in the number of intersections between two adjacent second data segments according to the number of intersections corresponding to two adjacent second data segments, and the specific formula is:

△ S _Y = S _Y - S _Y-1 (10)

Wherein, ΔS _Y represents the difference in the number of intersection points between the Y th second data segment and the Y-1 th second data segment, and S _Y represents the number of intersection points corresponding to the Y th second data segment.

Step S6: Determine whether a load mode change event occurs according to the difference in the number of intersections between a plurality of adjacent second data segments, specifically including:

；如果交点数量差值△S _Y大于

，则说明事件为负荷模式变动事件，并执行“步骤S62”；如果交点数量差值△S _Y小于或等于

，则令Y=Y+1，判断Y是否小于或等于M，如果Y小于或等于M，则返回“步骤S61”，如果Y大于M，则结束；其中，

为大于0小于1的有理数，S _Y-1表示第Y-1个第二数据段对应的交点个数，△S _Y表示第Y个第二数据段与第Y-1个第二数据段之间的交点数量差值；本实施例中，

取0.5。Step S61: Determine whether the difference in the number of intersections ΔS _Y is greater than

;If the difference in the number of intersections △ S _Y is greater than

, then the event is a load pattern change event, and "step S62" is executed; if the difference in the number of intersections △ S _Y is less than or equal to

, then let Y=Y+1, judge whether Y is less than or equal to M, if Y is less than or equal to M, return to "step S61", if Y is greater than M, end; where,

is a rational number greater than 0 and less than 1, S _Y-1 represents the number of intersections corresponding to the Y-1 second data segment, △ S _Y represents the difference between the Y second data segment and the Y-1 second data segment. The difference in the number of intersection points between; in this embodiment,

Take 0.5.

Step S62: Determine whether the second event ending condition is met. If the second event ending condition is met, the load mode change event has ended; if the second event ending condition is not met, continue to execute "Step S62".

The end conditions of the second event are as follows:

(11)

(11)

为大于0小于1的有理数，S _n-1表示第n-1个第二数据段对应的交点个数，K ₂ 表示设定的整数，△S _n表示第n个第二数据段与第n-1个第二数据段之间的交点数量差值。本实施例中，

取0.5，K ₂取5。Wherein, Y ₁ and Y ₂ represent the sequence numbers of two non-adjacent second data segments, Y ₁ is less than Y ₂ ,

is a rational number greater than 0 and less than 1, S _n-1 represents the number of intersections corresponding to the n-1 th second data segment, K ₂ represents the set integer, △ S _n represents the n th second data segment and the n th -1 difference in the number of intersections between the second data segments. In this embodiment,

Take 0.5 and K2 take ₅ .

Step S7: Determine the load action corresponding to the load state change event and/or the load mode change event, specifically including:

Step S71: Determine the load action corresponding to the load state change event, which specifically includes:

Step S711 : Extract the steady-state current waveform before the load state change event occurs, the steady-state current waveform after, and the voltage waveform corresponding to each steady-state current waveform to obtain the steady-state waveform of the event.

The specific determination formula of the steady-state current waveform is:

(12)

(12)

表示设定的稳态电流波形判断阈值，K ₃表示设定的整数，△P _n表示第n个第一数据段与第n-1个第一数据段之间平均功率差值；本实施例中，

为500瓦，K ₃为2。Wherein, X ₁ and X ₂ represent the sequence numbers of two non-adjacent first data segments, X ₁ is less than X ₂ ,

Represents the preset steady-state current waveform judgment threshold, K ₃ represents the set integer, ΔP _n represents the average power difference between the nth first data segment and the n-1th first data segment; this embodiment middle,

is 500 watts, and K3 is ₂ .

The specific formulas of the steady-state current waveform and voltage waveform of the event are:

(13)

(13)

(14)

(14)

Among them, I _o,r represents the steady state current waveform of the detected state change event, I _r represents the steady state current waveform after the load state change event, and I _r-1 represents the steady state current waveform before the load state change event Current waveform, U _{o,r represents} the voltage waveform of the detected state change event, Ur represents the voltage waveform after the load state change event, and Ur _-1 represents the voltage waveform before the load state change event.

Step S712: Determine the electrical characteristics corresponding to the load state change event according to the steady-state waveform of the event corresponding to the load state change event, and the specific formula is:

(15)

(15)

(16)

(16)

Among them, I _s is the rms value of the event current, I _o,r,z is the sampling current of the zth sampling point of the event current I _o,r , P _p is the event power, U _o,r,z is the event voltage U _{o, r} is the sampling voltage of the zth sampling point, T is the sampling period, and T is 100 in this embodiment.

Step S713: Compare the electrical characteristics corresponding to the load state change event with reference information to determine the load action corresponding to the load state change event. The reference information is at least one of parameters of the load equipment and production control information.

Step S72: Determine the load action corresponding to the load mode change event, which specifically includes:

Step S721: Extract the voltage and current waveforms of the whole process of the load mode change event to obtain the transient waveform of the event.

The specific formula of the transient waveform of the event is:

(17)

(17)

Among them, P _v,A,i , P _v,B,i , P _v,C,i respectively represent the extracted power value of the ith voltage cycle of A, B, and C phases during the load mode change event, i ∈ [1 ,TN ₂ L]; I _v,z,A , I _v,z,B , I _v,z,C represent the zth sampling current of the ith voltage cycle of phase A, B, and C during the load mode change event, respectively ; U _{v, z, A} , U _{v, z, B} , U _{v, z, C} represent the z-th sampled voltage of the i -th voltage cycle of A, B, and C phases during the load mode change event, respectively; T represents the voltage cycle The number of sampling points; L represents the number of the second data segment in which the load pattern change event lasts; I _v,h,A , I _v,h,B , I _v,h,C respectively represent A, The h-th sampled current of B and C phases; U _v,h,A , U _v,h,B , U _v,h,C respectively represent the h-th sampled voltage of A, B, and C phases during the load mode change event; z∈[1,T] represents the zth sampling point in the ith voltage cycle of A, B, and C phases during the load pattern change event; h ∈[1,TN ₂ L] represents the A, B, and C phase change event during the load pattern change event. , B, and C phase h sampling point; in this embodiment, T is 100.

Step S722: Determine the time characteristic corresponding to the load pattern change event according to the transient waveform of the load pattern change event, and the specific formula is:

T _v =TN ₂ L( 18)

Wherein, T _v represents the duration of the load pattern change event, T represents the number of sampling points of the voltage cycle, L represents the number of the second data segment during which the load pattern change event lasts; in this embodiment, T is 100.

Step S723: Compare the time characteristic corresponding to the load pattern change event with reference information to determine the load action corresponding to the load pattern change event. The reference information is at least one of parameters of the load equipment and production control information.

Example 2

As shown in FIG. 2, the present invention also provides a non-intrusive industrial load event detection system, the system includes:

The first electricity consumption data set construction module 201 is used for synchronously collecting electricity consumption data from the power metering place of the industrial user through the non-intrusive collection device, so as to constitute the first electricity consumption data set.

an average power difference calculation module 202, configured to intercept N adjacent first data segments from the first power consumption data set, and calculate the average power difference between any two adjacent first data segments; Among them, N is a positive integer greater than 2.

The load state change event determining module 203 is configured to determine whether a load state change event occurs according to the average power difference corresponding to a plurality of adjacent first data segments.

The second power consumption data set construction module 204 is configured to construct a second power consumption data set according to the detected load state change event.

The intersection number difference calculation module 205 is used to intercept M adjacent second data segments from the second power consumption data set, and calculate the intersection number difference between any two adjacent second data segments; Among them, M is a positive integer greater than 2.

The load pattern change event determination module 206 is configured to determine whether a load pattern change event occurs according to the difference in the number of intersections between a plurality of adjacent second data segments.

As an optional implementation manner, the average power difference calculation module 202 of the present invention specifically includes:

The initial point determination unit is used for taking the collection point corresponding to the voltage value from negative to positive as the initial point of the voltage cycle.

A first intercepting unit, configured to take an initial point as a critical point, intercepting N adjacent first data segments from the first power consumption data set; each of the _first data segments includes N1 voltage period collections The electricity consumption data of ; wherein, N ₁ is a positive integer greater than 1.

The first load electric power calculation unit is configured to calculate the load electric power of each voltage period in each of the first data sections according to the electric power consumption data collected in each voltage period in each of the first data sections.

The first load average power consumption value calculation unit is configured to calculate the load average power consumption value in each of the first data sections according to the load power consumption of each voltage period in each of the first data sections.

The average power difference calculation unit is configured to calculate the average power difference between the two adjacent first data segments according to the load average electric power values of the two adjacent first data segments.

As an optional implementation manner, the load state change event determination module 203 of the present invention specifically includes:

A load electric power variance calculation unit, configured to calculate the load corresponding to each of the first data segments according to the load electric power of each voltage period in each of the first data segments and the average load electric power value of the load in each of the first data segments power consumption variance;

；如果△P _X 大于

，则说明事件为负荷状态变动事件，并执行“第二判断单元”；如果△P _X 小于或等于

，则令X=X+1，判断X是否小于或等于N，如果X小于或等于N，则执行“第一判断单元”，如果X大于N，则结束；其中，

均为大于0小于1的有理数，P' _X 表示第X个第一数据段对应的负荷用电功率方差，P _X-1 表示第X-1个第一数据段内负荷平均用电功率值；The first judgment unit is used to judge whether the average power difference Δ P _X is greater than

; if △ P _X is greater than

, then the event is a load state change event, and the "second judgment unit" is executed; if △ P _X is less than or equal to

, then let X=X+1, judge whether X is less than or equal to N, if X is less than or equal to N, execute the "first judgment unit", if X is greater than N, end; where,

All are rational numbers greater than 0 and less than 1, P' _X represents the variance of the load power consumption corresponding to the Xth first data segment, and P _X-1 represents the load average power consumption value in the X-1th first data segment;

The second judging unit is used to judge whether the first event ending condition is met. If the first event ending condition is met, it means that the load state change event has ended. ".

As an optional implementation manner, the load pattern change event determination module 206 of the present invention specifically includes:

；如果交点数量差值△S _Y大于

，则说明事件为负荷模式变动事件，并执行“第四判断单元”；如果交点数量差值△S _Y小于或等于

，则令Y=Y+1，判断Y是否小于或等于M，如果Y小于或等于M，则返回“第三判断单元”，如果Y大于M，则结束；其中，

为大于0小于1的有理数，S _Y-1表示第Y-1个第二数据段对应的交点个数，△S _Y表示第Y个第二数据段与第Y-1个第二数据段之间的交点数量差值。The third judging unit is used to judge whether the difference in the number of intersections △ S _Y is greater than

;If the difference in the number of intersections △ S _Y is greater than

, then the event is a load pattern change event, and the “fourth judgment unit” is executed; if the difference in the number of intersections △ S _Y is less than or equal to

, then let Y=Y+1, judge whether Y is less than or equal to M, if Y is less than or equal to M, return to the "third judgment unit", if Y is greater than M, end; where,

is a rational number greater than 0 and less than 1, S _Y-1 represents the number of intersections corresponding to the second data segment Y-1, △ S _Y represents the difference between the second data segment Y and the second data segment Y-1 The difference in the number of intersections between.

The fourth judging unit is used to judge whether the second event ending condition is met. If the second event ending condition is met, it means that the load mode change event has ended; ".

As an optional implementation manner, the system of the present invention further includes:

The load action determination module is configured to determine the load action corresponding to the load state change event and/or the load mode change event.

As an optional implementation manner, the load action determination module of the present invention specifically includes:

The first load action determination unit is configured to determine the load action corresponding to the load state change event.

The second load action determination unit is configured to determine the load action corresponding to the load pattern change event.

Example 3

As shown in Figure 3, a non-intrusive acquisition device is installed at the low-voltage side power measurement of a steel plant to collect voltage waveform and current waveform signal data. The sampling rate is 5000Hz and the acquisition time is 60 minutes.

Experiment according to the method disclosed in the present invention:

1) According to the user's production process classification event type, taking the steel plant in this embodiment as an example, according to the production process, in addition to the ironmaking blast furnace and its supporting loads, there will be open and closed state changes in the steel plant, among which the hot rolling mill, Loads such as cold rolling mills, coilers, uncoilers, flying shears, continuous casting machines, etc., due to the gap between the workpieces, the load will repeatedly fluctuate between whether or not to output force to the workpiece, and a mode change event will occur.

2) Collect the total current waveform and voltage waveform of the industrial user.

3) Calculate the average power consumption of each first data segment.

4) Detecting and extracting state change events through transient waveform features, FIG. 4 shows the state change event detected at time t ₁ and the current and voltage waveforms extracted by the present invention.

5) The collected signal is segmented by state change events.

6) Calculate the number S _Y of intersections between the power curve of each second data segment and its mean horizontal line.

7) The mode change event is detected by the transient waveform feature. Figure 5 shows the mode change event detected at time t ₂ and the power waveform extracted by the present invention.

8) Calculate the electrical and time characteristics of the event, compare it with the reference information in Example 1, identify the load type of the event, and restore its load action.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant part can be referred to the description of the method.

In this paper, specific examples are used to illustrate the principles and implementations of the present invention. The descriptions of the above embodiments are only used to help understand the method and the core idea of the present invention; meanwhile, for those skilled in the art, according to the present invention There will be changes in the specific implementation and application scope. In conclusion, the contents of this specification should not be construed as limiting the present invention.

Claims (10)

1. A method of non-intrusive industrial load event detection, the method comprising:

synchronously acquiring electricity utilization data from an electric power metering position of an industrial user through a non-invasive acquisition device to form a first electricity utilization data set;

intercepting N adjacent first data segments from the first power data set, and calculating the average power difference between any two adjacent first data segments; wherein N is a positive integer greater than 2;

determining whether a load state change event occurs according to the average power difference value corresponding to a plurality of adjacent two first data segments;

constructing a second electrical data set according to the load state change event obtained by detection;

intercepting M adjacent second data segments from the second electrical data set, and calculating the intersection number difference between any two adjacent second data segments; wherein M is a positive integer greater than 2;

and determining whether a load mode change event occurs according to the difference of the number of the intersection points between a plurality of adjacent two second data segments.

2. The method of non-intrusive industrial load event detection as defined in claim 1, further comprising:

and determining the load action corresponding to the load state change event and/or the load mode change event.

3. The method according to claim 1, wherein the step of extracting N adjacent first data segments from the first electrical data set and calculating the average power difference between any two adjacent first data segments comprises:

taking a collecting point corresponding to the change of the voltage value from negative to positive as an initial point of a voltage cycle;

intercepting N adjacent first data segments from the first electrical data set by taking an initial point as a critical point; each of the first data segments includesN ₁Electricity consumption data collected in each voltage period; wherein,N ₁is a positive integer greater than 1;

calculating the load power consumption of each voltage period in each first data segment according to the power consumption data acquired by each voltage period in each first data segment;

calculating the average power consumption value of the load in each first data segment according to the power consumption of the load in each voltage period in each first data segment;

and calculating the average power difference value between two adjacent first data segments according to the load average power utilization values of the two adjacent first data segments.

4. The method according to claim 3, wherein the determining whether the load state change event occurs according to the average power difference corresponding to two adjacent first data segments specifically comprises:

step S31: calculating load power consumption variance corresponding to each first data segment according to the load power consumption of each voltage period in each first data segment and the load average power consumption value in each first data segment;

step S32: judging the average power difference deltaP _X Whether or not greater than

(ii) a If ΔP _X Is greater than

If yes, the event is a load state change event, and step S33 is executed; if ΔP _X Is less than or equal to

If yes, let X = X +1, determine whether X is less than or equal to N, if X is less than or equal to N, execute "step S32", if X is greater than N, end; wherein,

are all rational numbers greater than 0 and less than 1,P' _X the power consumption variance of the load corresponding to the Xth first data segment is shown,P _X-1 representing the average power consumption value of the load in the X-1 th first data segment;

step S33: it is determined whether the first event end condition is satisfied, and if the first event end condition is satisfied, it indicates that the load-state variation event has ended, and if the first event end condition is not satisfied, the process proceeds to "step S33".

5. The method according to claim 4, wherein the first event ending condition is:

wherein, X₁And X₂Indicating two non-adjacent first data segment sequence numbers, X₁Less than X₂，P' _nThe power consumption variance of the load corresponding to the nth first data segment is shown,P _n-1represents the average power consumption value of the load in the (n-1) th first data segment,K ₁denotes a set integer,. DELTAP _nRepresents the average power difference between the nth first data segment and the (n-1) th first data segment.

6. The method according to claim 1, wherein the determining whether the load pattern change event occurs according to the difference between the number of intersections between two adjacent second data segments specifically comprises:

step S61: the difference value delta of the number of the intersection points is judgedS _YWhether or not greater than

(ii) a If the number of intersections is different by ΔS _YIs greater than

If yes, the event is a load mode change event, and step S62 is executed; if the number of intersections is different by ΔS _YIs less than or equal to

If yes, let Y = Y +1, determine whether Y is less than or equal to M, if Y is less than or equal to M, return to "step S61", if Y is greater than M, end; wherein,

is a rational number greater than 0 and less than 1,S _Y-1the number of the intersection points corresponding to the Y-1 th second data segment is represented;

step S62: judging whether a second event ending condition is met, if so, indicating that the load mode change event is ended; if the second event end condition is not satisfied, "step S62" is continuously performed.

7. The method according to claim 6, wherein the second event ending condition is:

wherein, Y₁And Y₂Indicating two non-adjacent second data segment sequence numbers, Y₁Less than Y₂，S _n-1Indicates the number of the intersection points corresponding to the (n-1) th second data segmentNumber of，K ₂ Denotes a set integer,. DELTAS _nIndicating the difference in the number of intersections between the nth second data segment and the (n-1) th second data segment.

8. A non-intrusive industrial load event detection system, the system comprising:

the first electricity data set construction module is used for synchronously acquiring electricity data from the electric power metering position of an industrial user through a non-invasive acquisition device to form a first electricity data set;

the average power difference calculation module is used for intercepting N adjacent first data segments from the first power data set and calculating the average power difference between any two adjacent first data segments; wherein N is a positive integer greater than 2;

the load state change event determining module is used for determining whether a load state change event occurs according to the average power difference value corresponding to the two adjacent first data segments;

the second electrical data set building module is used for building a second electrical data set according to the load state change event obtained by detection;

the intersection number difference calculation module is used for intercepting M adjacent second data segments from the second electrical data set and calculating the intersection number difference between any two adjacent second data segments; wherein M is a positive integer greater than 2;

and the load mode change event determining module is used for determining whether a load mode change event occurs according to the difference value of the number of the intersection points between the plurality of adjacent second data segments.

9. The system of claim 8, wherein the mean power difference calculation module comprises:

the initial point determining unit is used for taking the acquisition point corresponding to the voltage value from negative to positive as the initial point of the voltage cycle;

a first cut-off unit for cutting off the first light beam from the initial point as the critical pointIntercepting N adjacent first data segments on the electricity utilization data set; each of the first data segments includesN ₁Electricity consumption data collected in each voltage period; wherein N is₁Is a positive integer greater than 1;

the first load electricity consumption power calculation unit is used for calculating the load electricity consumption power of each voltage period in each first data segment according to the electricity consumption data acquired by each voltage period in each first data segment;

the first load average power consumption value calculation unit is used for calculating the load average power consumption value in each first data segment according to the load power consumption of each voltage period in each first data segment;

and the average power difference calculation unit is used for calculating the average power difference between the two adjacent first data segments according to the load average power utilization values of the two adjacent first data segments.

10. The system of claim 9, wherein the load-state-change event determination module comprises:

the load electricity consumption power variance calculating unit is used for calculating load electricity consumption power variances corresponding to the first data sections according to the load electricity consumption power of each voltage period in each first data section and the load average electricity consumption power value in each first data section;

a first judgment unit for judging the average power difference value DeltaP _X Whether or not greater than

(ii) a If ΔP _X Is greater than

If yes, the event is a load state change event, and a second judgment unit is executed; if a is equalP _X Is less than or equal to

If yes, let X = X +1, determine whether X is less than or equal to N, if X is less than or equal to N, execute "first determination unit", if X is greater than N, end; wherein,

are all rational numbers greater than 0 and less than 1,P' _X the power consumption variance of the load corresponding to the Xth first data segment is shown,P _X-1 representing the average power consumption value of the load in the X-1 th first data segment;

and a second judging unit for judging whether the first event ending condition is satisfied, if the first event ending condition is satisfied, indicating that the load state change event has ended, and if the first event ending condition is not satisfied, continuing to execute the second judging unit.

Images