CN117220417B - Dynamic monitoring method and system for consumer-side electrical load - Google Patents

Abstract

The invention provides a dynamic monitoring method and a system for consumer-side electrical load, wherein the method comprises the following steps: step 1: collecting historical electricity records of a consumption end; step 2: acquiring the load characteristics of each consumption end according to the historical electricity utilization record; step 3: establishing an electric load model of a consumer according to the load characteristics; step 4: acquiring a real-time electricity utilization record of a consumption terminal; step 5: and dynamically monitoring the consumer-side electric load according to the real-time electric record and the electric load model to obtain a dynamic monitoring result. According to the method and the system for dynamically monitoring the electrical load of the consumer, the historical electrical records of the consumer are introduced, the load characteristics of each consumer are determined, and an electrical load model of the consumer is built based on the load characteristics. The consumer-side electric load is dynamically monitored according to the real-time electric record and the electric load model, a dynamic monitoring result is obtained, the suitability of electric load detection is improved, the subsequent realization of load balance is facilitated, and the operation efficiency of the electric power system is improved.

Description

Dynamic monitoring method and system for consumer-side electrical load

Technical Field

The invention relates to the technical field of electric power information acquisition, in particular to a dynamic monitoring method and system for consumer-side electric loads.

Background

An electrical load refers to the amount of electrical energy supply required in an electrical power system, i.e., the consumer's (e.g., home, business, factory, etc.) electricity demand over a specified period of time. The electrical load is usually expressed in the form of power, i.e. the amount of electrical energy consumed per unit time.

The magnitude of the electrical load depends on consumer's electrical equipment, electricity usage habits, and electricity patterns. Different types of consumers have different load characteristics. For example, residential electric loads are typically dominated by household electricity, including household appliances such as lighting, air conditioning, television, refrigerator, and the like; while the electrical load of industrial enterprises may be higher, involving large machinery, production lines, motors, etc.

The power system needs to adjust the power generation capacity according to the requirements of the electric load so as to keep the supply and demand balance. Accurate prediction and management of electrical loads is important for stable operation and energy planning of electrical power systems. By monitoring, analyzing and scheduling the electric load, measures such as load balancing, energy optimization and load response can be realized, and the efficiency and reliability of the electric power system are improved.

The application number is: the invention patent of CN201410209653.0 discloses a power load monitoring system and method based on autonomous wireless networking, wherein the system comprises: set up in the unified monitoring platform, district intelligent management terminal, intelligent power distribution cabinet, district intelligent switch and intelligent domestic protection switch at county level main website center, unified monitoring platform, intelligent power distribution cabinet, district intelligent switch and intelligent domestic protection switch all with district intelligent management terminal communication connection. The invention solves the problem of real-time power load information monitoring such as power load voltage, current and residual current in the existing rural power grid power distribution area, solves the problem of communication data flow rate cost required for monitoring the power load of the rural power grid power distribution area, and solves the problem of rapid positioning and checking of overload, leakage current and other faults of the rural power grid power distribution area.

However, when the prior art detects the electric load, the electric load is not compared with the reasonable electric data of the user, when the abnormal electric consumption of the user exists, the abnormal electric consumption of the user cannot be detected in time, the detection process is unsuitable, further, the change of the electric consumption requirement cannot be perceived in time, and the operation efficiency of the electric power system is low.

In view of the foregoing, there is a need for a method and system for dynamic monitoring of consumer electrical loads that addresses at least the above-mentioned shortcomings.

Disclosure of Invention

Aiming at the defects, the invention provides a dynamic monitoring method and a system for the electric load of a consumer, historical electricity records of the consumer are introduced, the load characteristics of each consumer are determined, and an electric load model corresponding to the consumer is established based on the load characteristics. The consumer electric load is dynamically monitored according to the acquired real-time electric record and electric load model of the consumer, a dynamic monitoring result is acquired, the suitability of electric load detection is improved, further, the subsequent realization of load balance is facilitated, and the operation efficiency of the electric power system is improved.

The invention provides the following technical scheme: the method for dynamically monitoring the electrical load of the consumer end adopts an ETL data acquisition method to acquire the electrical load condition of the consumer end in real time, and comprises the following steps:

Step 1: collecting historical electricity records of a consumption end;

step 2: acquiring the load characteristics of each consumption end according to the historical electricity utilization record;

step 3: establishing an electric load model of a consumer according to the load characteristics;

step 4: acquiring a real-time electricity utilization record of a consumption terminal;

step 5: and dynamically monitoring the consumer-side electric load according to the real-time electric record and the electric load model to obtain a dynamic monitoring result.

Further, the step 1: collecting historical electricity usage records of a consumer, comprising:

acquiring a history electricity utilization record of the intelligent ammeter and uploading the history electricity utilization record;

or alternatively, the first and second heat exchangers may be,

and analyzing the electricity bill of the electric company to obtain the historical electricity record.

Further, the step 2: according to the historical electricity utilization record, acquiring the load characteristics of each consumer, including:

step 2.1: acquiring a first feature type of a load feature, and determining a feature extraction template corresponding to the first feature type; the first feature type includes: peak load, load distribution, load volatility, load curve shape, load duration, and average load;

step 2.2: and determining the load characteristics extracted and summarized by each characteristic extraction template according to the historical electricity utilization record based on the characteristic extraction templates.

Further, the step 3: according to the load characteristics, establishing an electric load model of the consumer side, which comprises the following steps:

step 3.1: obtaining a visual template corresponding to the load characteristics;

step 3.2: acquiring a first load curve according to the visualization template and the load characteristics; the first load curve is provided with n data acquisition points, each data acquisition point acquires corresponding load characteristics, and a first load curve data set acquired by the n data acquisition points is constructedA，Wherein->Is the ith first load curve data; />；

Step 3.3: acquiring a second characteristic type of the load characteristic corresponding to the first load curve to obtain a second characteristic type data set with n dataB，The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Data for a j-th second feature type; />；

Step 3.4: based on the second characteristic type data set B and the first load curve data setADetermining an analysis curve group W:

wherein->,/>The kth analysis element in the analysis curve set W +.>For the j-th data in the second feature type data set B>Is +/with the ith data in the first load curve data set A>A set of abnormal load characteristic values, wherein K analysis elements in the analysis curve group W form an analysis curve; the analysis curve set W satisfies the following conditions:

1) Boundary conditions:and->；

2) If the kth analysis elementThen the kth analysis element->The method meets the following conditions: 1. e.gtoreq.c.gtoreq.0 and 1.gtoreq.p.gtoreq.0, wherein ≡c>；

Step 3.5: constructing a curve analysis strategy calculation model, and determining a curve analysis strategy;

the curve analysis strategy calculation model is as follows:

；

；

wherein,for the ith first load curve data +.>And j < th > second feature type data->Euclidean distance between +.>；

Performing curve analysis according to the analysis curve group and the curve analysis strategy to determine abnormal load characteristics; the conditions for determining the abnormal load characteristics are as follows: if it isThen the j-th data in the second feature type data set B +.>Is an abnormal load characteristic; otherwise, the j-th data in the second feature type data set B is +.>Is not an abnormal load feature;

step 3.6: removing the abnormal load characteristics to obtain reforming load characteristics;

step 3.7: and establishing an electric load model of the consumer side according to the reforming load characteristics.

Further, the step 5: according to the real-time electricity record and the electricity load model, dynamically monitoring the electricity load of the consumer end to obtain a dynamic monitoring result, comprising:

step 5.1: analyzing the real-time electricity utilization record and determining electricity utilization data;

Step 5.2: determining demonstration data in an electric load model according to the current moment of the real-time electricity utilization record;

step 5.3: determining data differences according to the electricity consumption data and the demonstration data;

step 5.4: if the data difference is greater than or equal to a preset first threshold, determining difference data;

step 5.5: and acquiring a difference data summary table according to the difference data, and displaying a dynamic monitoring result based on the difference data summary table.

Further, the step 3.4: based on the second characteristic type data set B and the first load curve data setADetermining a set of analysis curves W, comprising:

step 3.41: acquiring a first history combination record;

step 3.42: determining a combination type of the first historical combination record; the combination types include: traversing the combination and the basis combination;

step 3.43: if the combination type is traversal combination, the corresponding first history combination record is used as a second history combination record;

step 3.44: obtaining a combination result of the second history combination record;

step 3.45: determining the correlation degree of the combined result according to the result semantic item corresponding to the combined result and a preset correlation semantic item-correlation value library;

step 3.46: if the correlation degree is greater than or equal to a preset second threshold value, the corresponding second history combination record is used as a third history combination record;

Step 3.47: if the combination type is the basis combination, the corresponding first history combination record is used as a fourth history combination record;

step 3.48: acquiring a criterion of a fourth history combination record;

step 3.49: analyzing the basis standard to obtain at least one basis standard item;

step 3.410: determining a standard value according to the standard item according to the fourth history combination record;

step 3.411: acquiring data attributes according to the standard items, and calculating the reference probability of the data attributes in a fourth history combination record;

step 3.412: correspondingly multiplying and summing the reference probability and the standard value to obtain a combined necessary value;

step 3.413: if the combination necessary value is larger than or equal to a preset third threshold value, the corresponding fourth history combination record is used as a fifth history combination record;

step 3.414: integrating the fourth history combination record and the fifth history combination record to obtain a sixth history combination record;

step 3.415: analyzing the sixth history combination record to obtain a feature type set;

step 3.416: traversing the feature type set in sequence, judging whether the second feature type of the first load curve is consistent with the fourth feature type in the feature type set being traversed every time, and if so, taking the first load curve corresponding to the second feature type as the second load curve;

Step 3.417: after traversing one feature type set each time, integrating a second load curve of the corresponding feature type set to obtain an analysis curve group;

step 3.418: and after all the feature type sets are traversed, all the analysis curve groups are obtained.

Further, the step 3.7 establishes an electric load model of the consumer side according to the reforming load characteristics, and comprises the following steps:

step 3.71: acquiring a third load curve according to the visual template and the reforming load characteristics;

step 3.72: based on a preset intercepting period, intercepting a third load curve with the same curve attribute, and determining a plurality of intercepting curve segments;

step 3.73: acquiring the period bit of the interception period of the interception curve segment;

step 3.74: projecting the intercepting curve segments adjacent in the period order to the same coordinate axis, and aligning the starting points of intercepting curves of the intercepting curve segments in the same coordinate axis;

step 3.75: by scanning the coordinate axis perpendicular to the coordinate axis, recording the interception information when the perpendicular is intercepted by the interception curve;

step 3.76: determining future load characteristics according to the intercepted information;

step 3.77: an electrical load model is determined based on the future load characteristics.

Further, the step 3.76: determining future load characteristics according to the intercepted information, including:

Step 3.761: analyzing the interception information to obtain an interception relation, an interception amount and an interception time;

step 3.762: determining a future interception time;

step 3.763: acquiring interception time of a preset time length before future interception time, and taking the interception time as target reference time;

step 3.764: acquiring a target time axis;

step 3.765: according to the target reference time, correspondingly labeling the reference interception relation and the reference interception amount on a target time axis to obtain a labeled image;

step 3.766: according to the annotation image, determining a first change relation of the reference interception relation with time and a second change relation of the reference interception amount with time;

step 3.767: determining a future interception relation and a future interception amount corresponding to the future interception time according to the first change relation and the second change relation based on the labeling relation of the future interception time corresponding to the target time axis;

step 3.768: and determining future load characteristics according to the future interception relation and the future interception amount.

Further, the method further comprises the following steps:

step 6: if at least one monitoring reminding event exists in the dynamic monitoring result, carrying out corresponding processing;

wherein, step 6: if at least one monitoring reminding event exists in the dynamic monitoring result, corresponding processing is carried out, and the method comprises the following steps:

Step 6.1: determining a verification consumption end according to the monitoring reminding event;

step 6.2: acquiring the electricity utilization type of the verification consumption terminal;

step 6.3: according to the electricity utilization type, carrying out electricity utilization rationality analysis on the verification consumption terminal;

step 6.4: if the electricity rationality analysis passes, adjusting the default electric load quantity of the corresponding verification consumer;

step 6.5: if the electricity utilization rationality analysis does not pass, carrying out manual electricity utilization verification on the corresponding verification consumer based on a preset manual verification rule, and obtaining a verification result;

step 6.6: if the verification result is that verification passes, adjusting the default electric load quantity of the corresponding verification consumption terminal; otherwise, based on the verification result, a treatment plan is determined.

The invention also provides a consumer electric load dynamic monitoring system adopting the method, which comprises the following steps:

the historical electricity utilization record collection subsystem is used for collecting historical electricity utilization records of a consumption end;

the load characteristic acquisition subsystem is used for acquiring the load characteristic of each consumption end according to the historical electricity utilization record;

the electric load model modeling subsystem is used for establishing an electric load model of the consumption end according to the load characteristics;

the real-time electricity utilization record acquisition subsystem is used for acquiring the real-time electricity utilization record of the consumption terminal;

And the dynamic monitoring subsystem is used for dynamically monitoring the consumer-side electric load according to the real-time electric record and the electric load model to obtain a dynamic monitoring result.

The beneficial effects of the invention are as follows:

1. the method introduces historical electricity utilization records of the consumers, determines the load characteristics of each consumer, and establishes an electric load model corresponding to the consumer based on the load characteristics. The consumer electric load is dynamically monitored according to the acquired real-time electric record and electric load model of the consumer, a dynamic monitoring result is acquired, the suitability of electric load detection is improved, further, the subsequent realization of load balance is facilitated, and the operation efficiency of the electric power system is improved.

2. The invention acquires the load characteristics acquired by n data acquisition points in a first load curve through the step 3.2 to form a first load curve data set A, and further acquires a second characteristic data set B corresponding to the first load curve to further construct an analysis curve groupBy analysing the kth analysis element in the set of curves WAs j-th data in the second feature type data set B +.>Is +/with the ith data in the first load curve data set A>Abnormal load characteristic values between the first and second data, and constructing a curve analysis strategy calculation model by considering the distance between the data of the first load curve and the second characteristic data of the nth data acquisition point >And the distance between the data of the first load curve of the nth data acquisition point and the surrounding critical point and the shortest difference measure value between the data of the first load curve next to each other and the second characteristic data +.>、/>Etc., and further judging the j-th data in the second characteristic data type data set B +.>Whether the abnormal load characteristics are in accordance with the characteristics or not is judged, and then the abnormal load characteristics are effectively screened out and removed; the identification performance of the abnormal load characteristics and the accuracy of dynamic monitoring of the consumer-side electrical load are effectively improved.

3. The analysis curve set W constructed by the invention meets the convenience condition and the analysis element if the kth analysis elementThen the kth analysis element->The method meets the following conditions: 1. not less than 0 of e-c and not less than 1 of q-p not less than 0, whereinAnd further, the deviation of each data in the second characteristic type data set B and the first load curve data is ensured to be in a controllable range, and the abnormal load characteristic value in the second characteristic type data set B can be effectively identified and screened through a curve analysis strategy.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.

The technical scheme of the invention is further described in detail through the drawings and the embodiments.

Drawings

The invention will be described in more detail hereinafter on the basis of embodiments and with reference to the accompanying drawings. Wherein:

FIG. 1 is a schematic diagram of a method for dynamically monitoring consumer electrical load provided by the present invention;

fig. 2 is a schematic diagram of a consumer electrical load dynamic monitoring system provided by the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

The embodiment of the invention provides a dynamic monitoring method for consumer electric load, which adopts an ETL data acquisition method to acquire the electric load condition of a consumer in real time, as shown in figure 1, and comprises the following steps:

step 1: collecting historical electricity records of a consumption end; the historical electricity consumption records collected in the step 1 are data sets recorded in a historical electricity consumption load condition data warehouse, which are collected by adopting an ETL data collection method disclosed in the prior art of Chinese patents CN201811454518.7 and CN 201410682033.9;

Wherein, the consumption end is: home, business, factory, etc.; historical electricity usage record: the consumer uses the process record of each powered device historically, and different consumers have different electricity habits, such as: the electric load of residential areas is mainly household electricity, including household appliances such as illumination, air conditioner, television, refrigerator and the like, while the electric load of industrial enterprises is generally higher, and large-scale mechanical equipment, production lines, motors and the like are involved;

step 2: acquiring the load characteristics of each consumption end according to the historical electricity utilization record; wherein, the load characteristic is: how much peak load is, how much load is distributed, how much load fluctuates, how much load curve shape is, how much load duration is, how much average load is, etc.;

step 3: establishing an electric load model of a consumer according to the load characteristics; wherein, the electric load model is: a mathematical model for predicting and analyzing the electrical energy demand of a consumer in an electrical power system;

step 4: acquiring a real-time electricity utilization record of a consumption terminal; the real-time electricity consumption record is a process record of the consumer real-time electricity consumption condition;

step 5: and dynamically monitoring the consumer-side electric load according to the real-time electric record and the electric load model to obtain a dynamic monitoring result. Wherein, consumer side electrical loads are: the real-time electricity consumption of consumer, carry out dynamic monitoring to consumer electricity load and be: monitoring whether the electricity consumption of the consumption terminal is abnormal in real time; the dynamic monitoring result is as follows: normal electricity consumption and abnormal electricity consumption.

The working principle and the beneficial effects of the technical scheme are as follows:

the method and the device introduce historical electricity utilization records of the consumers, determine the load characteristics of each consumer, and establish an electric load model corresponding to the consumer based on the load characteristics. The consumer electric load is dynamically monitored according to the acquired real-time electric record and electric load model of the consumer, a dynamic monitoring result is acquired, the suitability of electric load detection is improved, further, the subsequent realization of load balance is facilitated, and the operation efficiency of the electric power system is improved.

In one embodiment, step 1: collecting historical electricity usage records of a consumer, comprising:

acquiring a history electricity utilization record of the intelligent ammeter and uploading the history electricity utilization record; the intelligent ammeter records are obtained by establishing a communication link with the intelligent ammeter based on the internet of things technology;

or alternatively, the first and second heat exchangers may be,

and analyzing the electricity bill of the electric company to obtain the historical electricity record. The electricity bill is a record of electricity consumption of the service user recorded by a local server of the electric company.

The working principle and the beneficial effects of the technical scheme are as follows:

according to the method, the intelligent electric meter and the electric company are introduced to acquire the historical electricity utilization record, so that the comprehensiveness of the historical electricity utilization record is improved.

In one embodiment, step 2: according to the historical electricity utilization record, acquiring the load characteristics of each consumer, including:

step 2.1: acquiring a first feature type of a load feature, and determining a feature extraction template corresponding to the first feature type; wherein the first feature type is: peak load, load distribution, load volatility, load curve shape, load duration, and average load; wherein, peak load is: a highest load value; the load distribution is: distribution of load over different time periods, such as: the ratio of peak time, valley time and flat peak time; load fluctuation reflects the fluctuation degree of load; the shape of the load curve is that the shape and the frequency spectrum characteristics of the load curve are analyzed by drawing the load curve or using methods such as Fourier transformation; the load duration is: calculating the duration of the load exceeding a certain threshold value, and measuring the duration and the stability of the load; the average load is: calculating an average value of the load for knowing the base load level of the consumer; the feature extraction template constraint corresponding to the first feature type only extracts the load features of the corresponding first feature type, and does not extract other contents;

Step 2.2: and determining the load characteristics extracted and summarized by each characteristic extraction template according to the historical electricity utilization record based on the characteristic extraction templates.

The working principle and the beneficial effects of the technical scheme are as follows:

according to the method and the device, the first characteristic type of the load characteristic is introduced, the load characteristic in the historical electricity record is extracted through the characteristic extraction template corresponding to the first characteristic type, and the pertinence of the load characteristic extraction process is improved.

In one embodiment, step 3: according to the load characteristics, establishing an electric load model of the consumer side, which comprises the following steps:

step 3.1: obtaining a visual template corresponding to the load characteristics; the visual template is as follows: the constraint only visualizes the load characteristics in a certain visualization form, and other forms are not adopted;

step 3.2: acquiring a first load curve according to the visualization template and the load characteristics; the first load curve is provided with n data acquisition points, and each data acquisition point acquires corresponding loadCharacterized in that a first load curve data set acquired by n data acquisition points is constructedA，Wherein->Is the ith first load curve data; />；

Step 3.3: acquiring a second characteristic type of the load characteristic corresponding to the first load curve to obtain a second characteristic type data set with n data B，The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Data for a j-th second feature type; />The method comprises the steps of carrying out a first treatment on the surface of the Wherein the second feature type is: peak load, load distribution, load volatility, load curve shape, load duration, and average load;

step 3.4: based on the second characteristic type data set B and the first load curve data setADetermining an analysis curve group W:

wherein->,/>The kth analysis element in the analysis curve set W +.>For the j-th data in the second feature type data set B>Is +/with the ith data in the first load curve data set A>A set of abnormal load characteristic values, wherein K analysis elements in the analysis curve group W form an analysis curve; the analysis curve set W satisfies the following conditions:

1) Boundary conditions:and->；

2) If the kth analysis elementThen the kth analysis element->The method meets the following conditions: 1. e.gtoreq.c.gtoreq.0 and 1.gtoreq.p.gtoreq.0, wherein ≡c>；

Step 3.5: constructing a curve analysis strategy calculation model, and determining a curve analysis strategy;

the curve analysis strategy calculation model is as follows:

；

；

wherein,for the ith first load curve data +.>And j < th > second feature type data->Euclidean distance between +.>；

Performing curve analysis according to the analysis curve group and the curve analysis strategy to determine abnormal load characteristics; the conditions for determining the abnormal load characteristics are as follows: if it is Then the j-th data in the second feature type data set B +.>Is an abnormal load characteristic; otherwise, the j-th data in the second feature type data set B is +.>Is not an abnormal load feature;

wherein, abnormal load characteristics are: analyzing load characteristics of abnormal results according to curve analysis strategies, such as: false peak load;

step 3.6: removing the abnormal load characteristics to obtain reforming load characteristics;

step 3.7: and establishing an electric load model of the consumer side according to the reforming load characteristics.

The working principle and the beneficial effects of the technical scheme are as follows:

the method comprises the steps of introducing a visualization template to visualize load characteristics, acquiring a first load curve, and determining an analysis curve group for combined analysis according to a second characteristic type of the first load curve. According to the third characteristic type of each analysis curve group and the introduced curve analysis strategy library, determining a curve analysis strategy, performing curve analysis, acquiring abnormal load characteristics determined by curve analysis, removing the abnormal load characteristics, determining reforming load characteristics, and establishing an electric load model of a consumer according to the reforming load characteristics, thereby improving the modeling accuracy of the electric load model.

The load characteristics acquired by n data acquisition points in the first load curve are acquired through the step 3.2 to form a first load curve data set A, a second characteristic data set B corresponding to the first load curve is further acquired, and then an analysis curve group is constructed By analyzing the kth analysis element +.in curve set W>As j-th data in the second feature type data set B +.>Is +/with the ith data in the first load curve data set A>Abnormal load characteristic values between the first and second data, and constructing a curve analysis strategy calculation model by considering the distance between the data of the first load curve and the second characteristic data of the nth data acquisition point>And the distance between the data of the first load curve of the nth data acquisition point and the surrounding critical point and the shortest difference measure value between the data of the first load curve next to each other and the second characteristic data +.>、/>Etc., and further judging the j-th data in the second characteristic data type data set B +.>Whether the abnormal load characteristics are in accordance with the characteristics or not is judged, and then the abnormal load characteristics are effectively screened out and removed; the identification performance of the abnormal load characteristics and the accuracy of dynamic monitoring of the consumer-side electrical load are effectively improved.

In one embodiment, step 5: according to the real-time electricity record and the electricity load model, dynamically monitoring the electricity load of the consumer end to obtain a dynamic monitoring result, comprising:

step 5.1: analyzing the real-time electricity utilization record and determining electricity utilization data; wherein, the electricity consumption data is: when and how much electricity is used;

Step 5.2: determining demonstration data in an electric load model according to the current moment of the real-time electricity utilization record; wherein, the demonstration data is: the electricity load model presumes the current-moment electricity consumption data of the consumption terminal according to the historical electricity consumption habit of the consumption terminal;

step 5.3: determining data differences according to the electricity consumption data and the demonstration data; wherein, the data difference is: what data is bad, such as: the electricity consumption difference is small at peak;

step 5.4: if the data difference is greater than or equal to a preset first threshold, determining difference data; wherein the preset first threshold is preset manually;

step 5.5: and acquiring a difference data summary table according to the difference data, and displaying a dynamic monitoring result based on the difference data summary table. The difference data summary table includes difference data of different difference types including: peak value difference, valley value difference, electricity consumption period difference, and electricity consumption total amount difference.

The working principle and the beneficial effects of the technical scheme are as follows:

according to the method and the device, the demonstration data in the electric load model are determined according to the current moment of the real-time electric record, the data difference is determined according to the electric data and the demonstration data, and the difference data with the screening data difference larger than the first threshold are summarized in the difference data summary table to be displayed, so that the intuitiveness of displaying the dynamic monitoring result is improved.

In one embodiment, step 3.4: based on the second characteristic type data set B and the first load curve data setADetermining a set of analysis curves W, comprising:

step 3.41: acquiring a first history combination record; wherein, the first history combination record is: historically, recordings of combinatorial analyses of electrical load curves of different load characteristics, such as: recording the combination analysis of the load fluctuation curve and the load curve shape curve;

step 3.42: determining a combination type of the first historical combination record; the combination types include: traversing the combination and the basis combination; wherein, the traversal combination is: combinations that do not repeat the electrical load profile of the load signature; the basis combination is as follows: combining the electrical load curves of the load characteristics according to a certain selection basis;

step 3.43: if the combination type is traversal combination, the corresponding first history combination record is used as a second history combination record;

step 3.44: obtaining a combination result of the second history combination record; wherein, the combination result is: analysis process results in the second historical combined record for the electrical load curve of the joint analysis;

step 3.45: determining the correlation degree of the combined result according to the result semantic item corresponding to the combined result and a preset correlation semantic item-correlation value library; the result semantic item is: analyzing semantics of the obtained combined result based on semantic analysis technology; the preset related semantic item-related value library is as follows: the database comprises a plurality of one-to-one correlation semantic items and correlation value correspondence relations, such as: what the relevant values of the semantic items describe the load volatility and the load duration are; the degree of correlation is: a sum of correlation values for each of the related semantic items;

Step 3.46: if the correlation degree is greater than or equal to a preset second threshold value, the corresponding second history combination record is used as a third history combination record; wherein the preset second threshold value is preset manually;

step 3.47: if the combination type is the basis combination, the corresponding first history combination record is used as a fourth history combination record;

step 3.48: acquiring a criterion of a fourth history combination record; wherein, the standard is as follows: combining load characteristics depending on what;

step 3.49: analyzing the basis standard to obtain at least one basis standard item; wherein, according to standard terms, the method comprises the following steps: according to a single data item in the standard;

step 3.410: determining a standard value according to the standard item according to the fourth history combination record; wherein, the standard value is: the fourth history combination record corresponds to the matching degree of the record item according to the standard item and the record item according to the standard item;

step 3.411: acquiring data attributes according to the standard items, and calculating the reference probability of the data attributes in a fourth history combination record; wherein, the data attribute is: the fourth history combination record corresponds to a record type of the record item according to the standard item; the reference probabilities are: dividing the reference number of a certain data attribute by the total reference number;

Step 3.412: correspondingly multiplying and summing the reference probability and the standard value to obtain a combined necessary value; the larger the combination necessary value is, the more reasonable the combination selection of the load characteristics corresponding to the combination in the fourth history combination record is;

step 3.413: if the combination necessary value is larger than or equal to a preset third threshold value, the corresponding fourth history combination record is used as a fifth history combination record; wherein the preset third threshold value is preset manually;

step 3.414: integrating the fourth history combination record and the fifth history combination record to obtain a sixth history combination record;

step 3.415: analyzing the sixth history combination record to obtain a feature type set; the feature type set includes a plurality of fourth feature types, the fourth feature types being: a sixth history record of the corresponding peak load, load distribution, load volatility, load curve shape, load duration, and average load;

step 3.416: traversing the feature type set in sequence, judging whether the second feature type of the first load curve is consistent with the fourth feature type in the feature type set being traversed every time, and if so, taking the first load curve corresponding to the second feature type as the second load curve;

Step 3.417: after traversing one feature type set each time, integrating a second load curve of the corresponding feature type set to obtain an analysis curve group;

step 3.418: and after all the feature type sets are traversed, all the analysis curve groups are obtained.

The working principle and the beneficial effects of the technical scheme are as follows:

in general, a single analysis is not performed when the curve analysis of the load features is performed, and in order to better understand the correlation between the load features, the curve simultaneous analysis of the load features with different feature types is performed, so that the combination type of the first history combination record is introduced, and the combination type includes a traversal combination and a basis combination.

When the combination type is traversal combination, obtaining result semantic items of a combination result of the second history combination record, determining the correlation degree of the combination result according to the introduced correlation semantic item-correlation value library, and screening a third history combination record with the correlation degree being more than or equal to a second threshold value.

When the combination type is the basis combination, the basis standard of the fourth history combination record is obtained, the combination necessary value is calculated according to the standard value of each basis standard item and the reference probability of the data attribute of the basis standard item of the fourth history combination record, and when the combination necessary value is more than or equal to a preset third threshold value, the fifth history combination record is determined.

Integrating the fourth history combination record and the fifth history combination record to obtain a sixth history combination record, obtaining a feature type set according to the sixth history combination record, traversing the feature type set in sequence, determining a first load curve corresponding to a second feature type consistent with a fourth feature type in the feature type set as a second load curve each time, integrating the second load curve after traversing one feature type set each time to obtain an analysis curve group, and obtaining all analysis curve groups after traversing the feature type set. And the sixth history combination record for acquiring the subsequent feature type set is extracted according to the combination type classification, so that the acquisition efficiency of the sixth history combination record is improved, and further, the rationality and the comprehensiveness of acquiring the analysis curve set are improved.

In one embodiment, the step 3.7 establishes an electrical load model of the consumer according to the reforming load characteristics, including:

step 3.71: acquiring a second load curve according to the visualized template and the reforming load characteristics; wherein the generation principle of the second load curve is the same as that of the first load curve;

step 3.72: based on a preset intercepting period, intercepting a second load curve with the same curve attribute, and determining a plurality of intercepting curve segments; wherein, the intercepting period is preset by manual work, such as: one day, another example is: a quarter of a year; the curve attributes are: the type of descriptive variable of the load curve;

Step 3.73: acquiring the period bit of the interception period of the interception curve segment; wherein, the cycle rank is: the intercepting period of which is intercepted;

step 3.74: projecting the intercepting curve segments adjacent in the period order to the same coordinate axis, and aligning the starting points of intercepting curves of the intercepting curve segments in the same coordinate axis; wherein, the rules of adjacent intercepting curve segment projections are consistent and are all set manually; the coordinate axes are: two-dimensional rectangular coordinate axes;

step 3.75: by scanning the coordinate axis perpendicular to the coordinate axis, recording the interception information when the perpendicular is intercepted by the interception curve; wherein, the interception curve is: projecting the curve segments in the same coordinate axis; intercepting information is, for example: which curve is up, which curve is down, what the cut length is, what the meaning of the unit cut length representation is;

step 3.76: determining future load characteristics according to the intercepted information; wherein, future load characteristics are: how much the future peak load is, how much the future load is distributed, how much the future load fluctuates, how much the future load curve shape is, how much the future load duration is, how much the future average load is, etc.;

step 3.77: an electrical load model is determined based on the future load characteristics. When an electric load model is built according to the future load characteristics, the method is realized based on a deep learning technology.

The working principle and the beneficial effects of the technical scheme are as follows:

generally, when future trends are predicted, historical data are learned, but learning all the historical data has high requirement on system calculation force, so that the application introduces a interception period, intercepts a second load curve with the same curve attribute, and determines a plurality of interception curve segments. The method comprises the steps of projecting the intercepting curve segments adjacent in period order to the same coordinate axis and aligning, determining intercepting information when the perpendicular line of the perpendicular line scanning coordinate axis perpendicular to the coordinate axis is intercepted by the intercepting curve, determining future load characteristics through the intercepting information and determining an electric load model, and because the intercepting period is manually set, the intercepting information can learn tiny changes of historical electric records so as to achieve the effect of outputting a more accurate electric load model.

In one embodiment, the step 3.76: determining future load characteristics according to the intercepted information, including:

step 3.761: analyzing the interception information to obtain an interception relation, an interception amount and an interception time; wherein, the interception relation is: which curve is up and which curve is down; the interception amount is as follows: what the length of the cut is and what the meaning of the unit cut length representation is; the intercepting time is as follows: intercepting the recording time of the curve segment corresponding to the second load curve;

Step 3.762: determining a future interception time; wherein, the future interception time is: the prediction time at which the electric load prediction is required;

step 3.763: acquiring interception time of a preset time length before future interception time, and taking the interception time as target reference time; wherein the preset time length is preset manually;

step 3.764: acquiring a target time axis;

step 3.765: according to the target reference time, correspondingly labeling the reference interception relation and the reference interception amount on a target time axis to obtain a labeled image;

step 3.766: according to the annotation image, determining a first change relation of the reference interception relation with time and a second change relation of the reference interception amount with time; wherein, the first change relation is: referring to the change trend of the interception relation along with the change of time; the second variation relationship is: referring to the change trend of the interception amount along with the change of time;

step 3.767: determining a future interception relation and a future interception amount corresponding to the future interception time according to the first change relation and the second change relation based on the labeling relation of the future interception time corresponding to the target time axis; wherein, the future interception relation is: which curve is in the future is on the top and which curve is in the future is on the bottom; the future interception amount is as follows: what the future intercept length is and what the meaning of the unit intercept length representation is;

Step 3.768: and determining future load characteristics according to the future interception relation and the future interception amount. From the future intercept relationship and the future intercept amount, a future curve may be determined, and further, from the future curve, a future load characteristic may be determined.

The working principle and the beneficial effects of the technical scheme are as follows:

according to the method and the device, the interception time with the preset time length before the future interception time is introduced as the target reference time, so that the reference interception relation and the reference timeliness of the reference interception amount are improved. According to the annotation image, a first change relation of the reference interception relation changing along with time and a second change relation of the reference interception amount changing along with time are determined, a future interception relation and a future interception amount are determined according to the first change relation and the second change relation, a future load characteristic is determined by further determining a future curve, and accuracy of future load characteristic extraction is improved.

The embodiment of the invention provides a dynamic monitoring method for consumer-side electrical loads, which further comprises the following steps:

step 6: if at least one monitoring reminding event exists in the dynamic monitoring result, carrying out corresponding processing; wherein, the monitoring reminding event is: reminding events corresponding to abnormal consumption end points;

Wherein, step 6: if at least one monitoring reminding event exists in the dynamic monitoring result, corresponding processing is carried out, and the method comprises the following steps:

step 6.1: determining a verification consumption end according to the monitoring reminding event; wherein, the verification consumption end is: a consumer with abnormal electricity consumption;

step 6.2: acquiring the electricity utilization type of the verification consumption terminal; the electricity consumption type is as follows: verifying the type of the consumer-side electricity consumption main body, such as: civil, such as: industrial use;

step 6.3: according to the electricity utilization type, carrying out electricity utilization rationality analysis on the verification consumption terminal; the method comprises the steps of analyzing the rationality of abnormal electricity consumption when the electricity consumption rationality analysis is carried out on a verification consumption end, and comprehensively analyzing by combining the electricity consumption type, the electricity consumption scale and the uploaded electricity consumption range change record of the verification consumption end during analysis;

step 6.4: if the electricity rationality analysis passes, adjusting the default electric load quantity of the corresponding verification consumer; the default electric load is the electricity consumption of the verification consumer estimated by the power plant, and when the electricity consumption of the verification consumer is reasonable and the default electric load is not matched, the corresponding power supply strategy needs to be adjusted in time so as to ensure the stability of the power system;

step 6.5: if the electricity utilization rationality analysis does not pass, carrying out manual electricity utilization verification on the corresponding verification consumer based on a preset manual verification rule, and obtaining a verification result; wherein, the manual verification rule is: the method comprises the steps of manually checking the electric field area of a checking consumption end in a manual checking mode;

Step 6.6: if the verification result is that verification passes, adjusting the default electric load quantity of the corresponding verification consumption terminal;

otherwise, based on the verification result, a treatment plan is determined. The treatment scheme comprises the following steps: and (3) processing strategies for illegal electricity utilization behaviors (such as illegal coin digging and other behaviors).

The working principle and the beneficial effects of the technical scheme are as follows:

the method and the device determine the verification consumption terminal according to the monitoring reminding event, and based on the electricity utilization type of the verification consumption terminal, conduct electricity utilization rationality analysis on the verification consumption terminal, and when the electricity utilization rationality analysis passes, consider the stability of the power system and timely adjust the power supply strategy. When the electricity utilization rationality analysis is not passed, an artificial verification rule is introduced to carry out the artificial electricity utilization verification of the verification consumer end, and a verification result is obtained, when the verification result is passed, the power supply strategy is adjusted in time, otherwise, based on the verification result, a processing scheme is determined, and the monitoring result is fed back in time, so that the stability of the power system is improved.

The embodiment of the invention also provides a consumer electric load dynamic monitoring system adopting the method, as shown in fig. 2, comprising:

the historical electricity utilization record collecting subsystem 1 is used for collecting historical electricity utilization records of a consumption end;

The load characteristic acquisition subsystem 2 is used for acquiring the load characteristic of each consumption end according to the historical electricity utilization record;

the electric load model building subsystem 3 is used for building an electric load model of a consumption end according to load characteristics;

the real-time electricity utilization record acquisition subsystem 4 is used for acquiring the real-time electricity utilization record of the consumption terminal;

and the dynamic monitoring subsystem 5 is used for dynamically monitoring the consumer-side electric load according to the real-time electric record and the electric load model to obtain a dynamic monitoring result.

It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.

Claims (7)

1. The method for dynamically monitoring the electrical load of the consumer end adopts an ETL data acquisition method to acquire the electrical load condition of the consumer end in real time and is characterized by comprising the following steps:

step 1: collecting historical electricity records of a consumption end;

step 2: acquiring the load characteristics of each consumption end according to the historical electricity utilization record;

step 3: establishing an electric load model of a consumer according to the load characteristics;

Step 4: acquiring a real-time electricity utilization record of a consumption terminal;

step 5: dynamically monitoring the consumer-side electric load according to the real-time electric record and the electric load model to obtain a dynamic monitoring result;

the step 3: according to the load characteristics, establishing an electric load model of the consumer side, which comprises the following steps:

step 3.1: obtaining a visual template corresponding to the load characteristics;

step 3.2: acquiring a first load curve according to the visualization template and the load characteristics; the first load curve is provided with n data acquisition points, each data acquisition point acquires corresponding load characteristics, and a first load curve data set acquired by the n data acquisition points is constructedA，Wherein->Is the ith first load curve data;；

step 3.3: acquiring a second characteristic type of the load characteristic corresponding to the first load curve to obtain a second characteristic type data set with n dataB，The method comprises the steps of carrying out a first treatment on the surface of the Wherein (1)>Data for a j-th second feature type;；

step 3.4: based on the second characteristic type data set B and the first load curve data setADetermining an analysis curve group W:

wherein->,/>The kth analysis element in the analysis curve set W +.>For the j-th data in the second feature type data set B >Is +/with the ith data in the first load curve data set A>A set of abnormal load characteristic values, wherein K analysis elements in the analysis curve group W form an analysis curve; the analysis curve set W satisfies the following conditions:

1) Boundary conditions:and->；

2) If the kth analysis elementThen the kth analysis element->The method meets the following conditions: 1. e.gtoreq.c.gtoreq.0 and 1.gtoreq.p.gtoreq.0, wherein ≡c>；

Step 3.5: constructing a curve analysis strategy calculation model, and determining a curve analysis strategy;

the curve analysis strategy calculation model is as follows:

；

；

wherein,for the ith first load curve data +.>And j < th > second feature type data->Euclidean distance between +.>；

Performing curve analysis according to the analysis curve group and the curve analysis strategy to determine abnormal load characteristics; the conditions for determining the abnormal load characteristics are as follows: if it isThen the j-th data in the second feature type data set B +.>Is an abnormal load characteristic; otherwise, the j-th data in the second feature type data set B is +.>Is not an abnormal load feature;

step 3.6: removing the abnormal load characteristics to obtain reforming load characteristics;

step 3.7: establishing an electric load model of a consumption end according to the reforming load characteristics;

Step 3.7 establishes an electric load model of the consumer according to the reforming load characteristics, and comprises the following steps:

step 3.71: acquiring a third load curve according to the visual template and the reforming load characteristics;

step 3.72: based on a preset intercepting period, intercepting a third load curve with the same curve attribute, and determining a plurality of intercepting curve segments;

step 3.73: acquiring the period bit of the interception period of the interception curve segment;

step 3.74: projecting the intercepting curve segments adjacent in the period order to the same coordinate axis, and aligning the starting points of intercepting curves of the intercepting curve segments in the same coordinate axis;

step 3.75: by scanning the coordinate axis perpendicular to the coordinate axis, recording the interception information when the perpendicular is intercepted by the interception curve;

step 3.76: determining future load characteristics according to the intercepted information;

step 3.77: determining an electrical load model according to future load characteristics;

the step 3.76: determining future load characteristics according to the intercepted information, including:

step 3.761: analyzing the interception information to obtain an interception relation, an interception amount and an interception time;

step 3.762: determining a future interception time;

step 3.763: acquiring interception time of a preset time length before future interception time, and taking the interception time as target reference time;

Step 3.764: acquiring a target time axis;

step 3.765: according to the target reference time, correspondingly labeling the reference interception relation and the reference interception amount on a target time axis to obtain a labeled image;

step 3.766: according to the annotation image, determining a first change relation of the reference interception relation with time and a second change relation of the reference interception amount with time;

step 3.767: determining a future interception relation and a future interception amount corresponding to the future interception time according to the first change relation and the second change relation based on the labeling relation of the future interception time corresponding to the target time axis;

step 3.768: and determining future load characteristics according to the future interception relation and the future interception amount.

2. A method for dynamically monitoring consumer electrical load according to claim 1, wherein said step 1: collecting historical electricity usage records of a consumer, comprising:

acquiring a history electricity utilization record of the intelligent ammeter and uploading the history electricity utilization record;

or alternatively, the first and second heat exchangers may be,

and analyzing the electricity bill of the electric company to obtain the historical electricity record.

3. A method for dynamically monitoring consumer electrical load according to claim 1, wherein said step 2: according to the historical electricity utilization record, acquiring the load characteristics of each consumer, including:

Step 2.1: acquiring a first feature type of a load feature, and determining a feature extraction template corresponding to the first feature type; the first feature type includes: peak load, load distribution, load volatility, load curve shape, load duration, and average load;

step 2.2: and determining the load characteristics extracted and summarized by each characteristic extraction template according to the historical electricity utilization record based on the characteristic extraction templates.

4. A method for dynamic monitoring of consumer electrical loads according to claim 1, characterized in that said step 5: according to the real-time electricity record and the electricity load model, dynamically monitoring the electricity load of the consumer end to obtain a dynamic monitoring result, comprising:

step 5.1: analyzing the real-time electricity utilization record and determining electricity utilization data;

step 5.2: determining demonstration data in an electric load model according to the current moment of the real-time electricity utilization record;

step 5.3: determining data differences according to the electricity consumption data and the demonstration data;

step 5.4: if the data difference is greater than or equal to a preset first threshold, determining difference data;

step 5.5: and acquiring a difference data summary table according to the difference data, and displaying a dynamic monitoring result based on the difference data summary table.

5. A method for dynamic monitoring of consumer electrical loads according to claim 1, wherein said step 3.4: determining an analysis curve set W based on the second feature type data set B and the first load curve data set a, comprising:

step 3.41: acquiring a first history combination record;

step 3.42: determining a combination type of the first historical combination record; the combination types include: traversing the combination and the basis combination;

step 3.43: if the combination type is traversal combination, the corresponding first history combination record is used as a second history combination record;

step 3.44: obtaining a combination result of the second history combination record;

step 3.45: determining the correlation degree of the combined result according to the result semantic item corresponding to the combined result and a preset correlation semantic item-correlation value library;

step 3.46: if the correlation degree is greater than or equal to a preset second threshold value, the corresponding second history combination record is used as a third history combination record;

step 3.47: if the combination type is the basis combination, the corresponding first history combination record is used as a fourth history combination record;

step 3.48: acquiring a criterion of a fourth history combination record;

step 3.49: analyzing the basis standard to obtain at least one basis standard item;

Step 3.410: determining a standard value according to the standard item according to the fourth history combination record;

step 3.411: acquiring data attributes according to the standard items, and calculating the reference probability of the data attributes in a fourth history combination record;

step 3.412: correspondingly multiplying and summing the reference probability and the standard value to obtain a combined necessary value;

step 3.413: if the combination necessary value is larger than or equal to a preset third threshold value, the corresponding fourth history combination record is used as a fifth history combination record;

step 3.414: integrating the fourth history combination record and the fifth history combination record to obtain a sixth history combination record;

step 3.415: analyzing the sixth history combination record to obtain a feature type set;

step 3.416: traversing the feature type set in sequence, judging whether the second feature type of the first load curve is consistent with the fourth feature type in the feature type set being traversed every time, and if so, taking the first load curve corresponding to the second feature type as the second load curve;

step 3.417: after traversing one feature type set each time, integrating a second load curve of the corresponding feature type set to obtain an analysis curve group;

step 3.418: and after all the feature type sets are traversed, all the analysis curve groups are obtained.

6. The consumer electrical load dynamic monitoring method of claim 1, further comprising:

step 6: if at least one monitoring reminding event exists in the dynamic monitoring result, carrying out corresponding processing;

wherein, step 6: if at least one monitoring reminding event exists in the dynamic monitoring result, corresponding processing is carried out, and the method comprises the following steps:

step 6.1: determining a verification consumption end according to the monitoring reminding event;

step 6.2: acquiring the electricity utilization type of the verification consumption terminal;

step 6.3: according to the electricity utilization type, carrying out electricity utilization rationality analysis on the verification consumption terminal;

step 6.4: if the electricity rationality analysis passes, adjusting the default electric load quantity of the corresponding verification consumer;

step 6.5: if the electricity utilization rationality analysis does not pass, carrying out manual electricity utilization verification on the corresponding verification consumer based on a preset manual verification rule, and obtaining a verification result;

step 6.6: if the verification result is that verification passes, adjusting the default electric load quantity of the corresponding verification consumption terminal; otherwise, based on the verification result, a treatment plan is determined.

7. A consumer electrical load dynamic monitoring system employing the method of any one of claims 1-6, comprising:

The historical electricity utilization record collection subsystem is used for collecting historical electricity utilization records of a consumption end;

the load characteristic acquisition subsystem is used for acquiring the load characteristic of each consumption end according to the historical electricity utilization record;

the electric load model modeling subsystem is used for establishing an electric load model of the consumption end according to the load characteristics;

the real-time electricity utilization record acquisition subsystem is used for acquiring the real-time electricity utilization record of the consumption terminal;

and the dynamic monitoring subsystem is used for dynamically monitoring the consumer-side electric load according to the real-time electric record and the electric load model to obtain a dynamic monitoring result.