CN117477581B

CN117477581B - Power system load balancing control method and power system

Info

Publication number: CN117477581B
Application number: CN202311806405.XA
Authority: CN
Inventors: 洪澄杰; 廖蔚; 陈嘉乐; 齐勇; 杨志鹄; 黄钢忠; 姜春涛; 张清华; 李天宇
Original assignee: Foshan Dayan Data Technology Co ltd
Current assignee: Foshan Dayan Data Technology Co ltd
Priority date: 2023-12-26
Filing date: 2023-12-26
Publication date: 2024-03-26
Anticipated expiration: 2043-12-26
Also published as: CN117477581A

Abstract

The invention discloses a load balancing control method of an electric power system and the electric power system, which are applied to the technical field of electric power regulation and control, wherein the method comprises the following steps: acquiring and indexing relevant parameters at the current moment to obtain load influence characteristics; acquiring and decomposing historical electricity utilization data of a user to obtain historical electricity utilization seasonal indexes, historical electricity utilization wave power rates and historical electricity utilization periodic characteristics; predicting a power consumption seasonal index at a future time by using the seasonal index and the load influence characteristic, predicting a power consumption periodic characteristic at the future time by using the periodic characteristic, and taking the product of the two prediction results as a load prediction value of a user at the future time; obtaining an electric power regulation value of a user at a future moment according to the electric power utilization rate and the membership function; and when the electricity consumption target value of the user is larger than the sum of the predicted value and the regulating value, performing charge increment adjustment. The invention improves the accuracy and efficiency of load prediction, improves the flexibility of load balancing processing, and achieves better power regulation and control effect.

Description

Power system load balancing control method and power system

Technical Field

The invention relates to the technical field of power regulation and control, in particular to a power system load balancing control method and a power system.

Background

In a conventional power system, the balance of supply and demand of electric power is a critical issue. As the scale of power systems continues to expand and the consumer's power demands increase, the problem of unbalanced power loads becomes more pronounced. Such problems can lead to overload or insufficient power supply to certain areas or customers, thereby affecting the stability and reliability of the power system. In this regard, the load balancing technique is used to adjust and distribute the power load in the power system, so that the power load between each area or user is more balanced, and thus the balance of the power supply and demand of the power system is realized. The application of the load balancing technology is beneficial to improving the stability and reliability of the power system, optimizing the energy utilization efficiency, reducing the energy waste and reducing the running cost of the power grid.

At present, the process of realizing load balancing in the related art comprises the following steps: and predicting the load value of the power system in a certain preset time through the power consumption data of the user in the historical time period, comparing the predicted load value of the power system with the power consumption target of the user, and regulating and controlling the power of the power system by using the comparison result. For example, the chinese patent publication No. CN106712040a adopts the above method to realize load balancing. However, the related art has the following problems:

On the one hand, the related art only considers the electricity consumption data of the user in the history period, but does not sufficiently consider the electricity consumption fluctuation condition of the user in the history period and the influence of external index factors such as weather and the like on the power load of the power system. In practical applications, the power load of the power system is often affected by extreme values of some users and external index factors, which may cause the predicted load value of the power system to be far from the expected value, thereby affecting the power regulation effect.

On the other hand, in terms of power scheduling, the related art is generally classified into three levels by two preset thresholds, and power consumption of users is advanced by the classificationAnd (5) row regulation. However, users very close to the threshold value do not necessarily conform to the characteristics of the users in the classification level, which is easy to happen that the power consumption of the users is reversely regulated and controlled, and the elasticity degree and the accuracy of the rigid classification regulation and control mode are required to be improved. Illustratively, 60 kilowatt-hours and 120 kilowatt-hours are used as thresholds, and are divided into [0, 60 ], [60, 120]、(120，]Three intervals, which do not meet the user characteristics of [0, 60 ] for a 59.9 kilowatt-hour user, may result in the situation that electricity consumption of the 59.9 kilowatt-hour user is reversely regulated, and electricity consumption of the user is affected.

Disclosure of Invention

The invention aims to provide a power system load balancing control method and a power system, which are used for solving one or more technical problems in the prior art and at least providing a beneficial selection or creation condition.

The invention solves the technical problems as follows: in one aspect, an embodiment of the present invention provides a method for controlling load balancing in an electric power system, including the following steps:

acquiring related parameters which influence the load of the power system at the current moment, and performing index processing on the related parameters to obtain load influence characteristics;

acquiring historical electricity utilization data of a user, and decomposing the historical electricity utilization data to obtain historical electricity utilization seasonal indexes, historical electricity utilization wave power rates and historical electricity utilization periodic characteristics;

predicting to obtain the electricity consumption seasonal index at the future moment by using the load influence characteristic and the historical electricity consumption seasonal index and combining a support vector regression model, predicting to obtain the electricity consumption periodical characteristic at the future moment by using the historical electricity consumption periodical characteristic and combining a long-short-period memory neural network model, and calculating the product of the electricity consumption seasonal index at the future moment and the electricity consumption periodical characteristic at the future moment as a load predicted value of a user at the future moment;

According to the historical electricity utilization wave rate, combining a membership function, and calculating to obtain an electricity regulation value of a user at a future moment;

and acquiring a power consumption target value of a user, and performing charge increment adjustment on the power consumption target value when the power consumption target value is larger than the sum of the load predicted value and the power regulation value.

Further, in an embodiment of the present invention, the obtaining the relevant parameter that affects the load of the power system at the current time, and performing an indexing process on the relevant parameter to obtain a load affecting feature, includes:

acquiring relevant parameters affecting the load of the power system at the current moment and performing data cleaning treatment;

wherein the relevant parameters include: temperature information, wind speed information, humidity information, cloud cover information, air pressure information and extreme weather event information of the environment where the power system is located at the current moment, date type information, week information and season information of the current moment, user type information and user development status information of the power system used at the current moment;

the date type information is used for representing a holiday type or a workday type of a date to which the current moment belongs, and the user type information is used for representing an industry type of a user using the power system at the current moment;

Performing index processing on the air temperature information, the wind speed information and the humidity information to obtain a first load influence characteristic, wherein the first load influence characteristic meets the following formula:

；

wherein,the first load influence characteristic is used for representing the influence degree of air temperature information, wind speed information and humidity information on the load of the power system; t is air temperature information, v is wind speed information, and h is humidity information; />、/>And->Weights of air temperature information, wind speed information and humidity information respectively, k is a humidity influence factor, ++>The method comprises the steps of carrying out a first treatment on the surface of the b represents an adjustment constant;

and carrying out index processing on the cloud amount information and the air pressure information to obtain a second load influence characteristic, wherein the second load influence characteristic meets the following formula:

；

wherein,for the second load influence characteristic, which is used for representing the influence degree of cloud quantity information and air pressure information on the load of the power system, +.>Cloud amount information is adopted, and P is air pressure information; />、/>、/>And->The influence factors of sunny days, cloudy days and cloudy days on the load of the power system are respectively; />、/>、/>、/>The product of the probability of converting into rain on sunny days, cloudy days and the influence factor of the rain on the load of the power system is respectively;

and carrying out index processing on the extreme weather event information to obtain a third load influence characteristic, wherein the third load influence characteristic meets the following formula:

；

Wherein,a third load influence feature for characterizing the extent of influence of extreme weather event information on the load of the power system; />For extreme weather event information, < >>For normal weather conditions->，/>For other extreme weather conditions +.>Other extreme weather conditions->An impact factor on the load of the power system;

performing index processing on the date type information, the week information and the season information to obtain a fourth load influence characteristic, wherein the fourth load influence characteristic meets the following formula:

；

wherein,a fourth load influence feature for characterizing the degree of influence of the week information, the date type information, and the season information on the load of the power system; w is the influence factor of the week information on the load of the power system, R is the influence factor of the date type information on the load of the power system, Q is the influence factor of the season information on the load of the power system,/A>The influence constant of the region where the power system is located on the date type information, the week information and the season information;

and carrying out index processing on the user type information and the user development status information to obtain a fifth load influence characteristic, wherein the fifth load influence characteristic meets the following formula:

；

Wherein,a fifth load influence feature for characterizing the degree of influence of the user type information and the user development status information on the load of the power system; c is user type information, d is user development status information, < >>The influence constant of the region where the power system is located on the user type information is set;

the first load influencing feature, the second load influencing feature, the third load influencing feature, the fourth load influencing feature and the fifth load influencing feature are taken as load influencing features.

Further, in an embodiment of the present invention, the obtaining historical electricity consumption data of the user, and decomposing the historical electricity consumption data to obtain a historical electricity consumption seasonal index, a historical electricity consumption wave power rate and a historical electricity consumption periodicity feature, includes:

acquiring a power utilization curve of a user in a historical time period corresponding to the current moment as historical power utilization data of the user;

according to the historical electricity consumption data, a moving average method of a multiplication model is combined, and a historical electricity consumption seasonal index is calculated;

and calculating the ratio of the historical electricity consumption data to the historical electricity consumption seasonal index as the historical electricity consumption periodic characteristic after eliminating the seasonal influence.

and calculating the ratio of the standard deviation and the average value of the historical electricity consumption data as the historical electricity consumption wave rate.

Further, in an embodiment of the present invention, the calculating, according to the historical electricity consumption rate and the membership function, the electricity regulation value of the user at the future time includes:

determining a plurality of membership thresholds and a plurality of membership categories, and constructing a membership function of each membership category according to the membership thresholds and the historical electricity consumption wave power, wherein the membership function of the ith membership category satisfies the following formula:

；

wherein,membership value for user belonging to category C, < ->；/>Is a threshold value of the degree of membership,a solving function of the historical electricity utilization wave power in a corresponding interval is adopted;

obtaining a regulation and control coefficient of each membership class, and determining a power regulation and control value of a user at a future moment according to a membership function and the regulation and control coefficient of each membership class, wherein the power regulation and control value of the user at the future moment meets the following formula:

；

Wherein,the control coefficient for the i-th membership class.

Further, in an embodiment of the present invention, the obtaining the electricity target value of the user includes:

determining the membership value of the user to each membership class according to the membership function of each membership class, and selecting the membership class with the largest membership value as the largest membership class;

and acquiring the electricity utilization target value corresponding to the maximum membership type as the electricity utilization target value of the user.

Further, in one embodiment of the present invention, the method further comprises the steps of:

and when the electricity consumption target value is smaller than or equal to the sum of the load predicted value and the electric power regulation value, performing charge down-regulation on the electricity consumption target value.

Further, in one embodiment of the present invention, before the electric power consumption target value is subjected to the electric charge down-regulation or the electric charge up-regulation, the method further includes the steps of:

and acquiring electricity utilization excitation measure information corresponding to the maximum membership type.

Further, in one embodiment of the present invention, when the electricity consumption target value is subjected to charge down-regulation or charge up-regulation, the method further includes the steps of:

and sending electricity utilization incentive measure information to the terminal where the user is located.

On the other hand, the embodiment of the invention also provides a power system which is used for providing power for a plurality of users, and the power system adopts the power system load balancing control method to regulate the power load.

The beneficial effects of the invention are as follows: the load balancing control method and the power system for the power system effectively improve the accuracy and efficiency of load prediction, can more flexibly and individually allocate proper power consumption values for each user, and meet the elastic requirement of a user demand side on power consumption adjustment, so that the flexibility of load balancing processing is improved, and better power regulation and control effects can be achieved.

Additional features and advantages of the application 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 application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

Fig. 1 is a flowchart of a load balancing control method of an electric power system according to an embodiment of the present invention;

fig. 2 is a schematic diagram of a load balancing control method of an electric power system according to an embodiment of the present invention;

FIG. 3 is a flow chart of seasonal decomposition provided by an embodiment of the invention;

FIG. 4 is a flow chart of constructing membership functions provided by an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.

The present application is further described below with reference to the drawings and specific examples. The described embodiments should not be construed as limitations on the present application, and all other embodiments, which may be made by those of ordinary skill in the art without the exercise of inventive faculty, are intended to be within the scope of the present application.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with one another without conflict.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the present application.

Aiming at the problems and defects existing in the related art, the embodiment of the invention provides a load balancing control method of a power system and the power system, wherein the power system adopts the control method to perform power regulation and control, and fully considers the related parameters influencing the load of the power system, load influencing factors of multiple dimensions such as power consumption data of users in a historical time period, and the like, thereby improving the load prediction efficiency and the load prediction accuracy, distributing proper power consumption values for each user more flexibly and more personally, meeting the requirement of a user demand side on elastic power distribution, and further achieving a better power regulation and control effect.

First, the implementation steps of a load balancing control method for an electric power system according to an embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

The method provided by the embodiment of the invention can be applied to the terminal, the server, software running in the terminal or the server and the like. The terminal may be, but is not limited to, a tablet computer, a notebook computer, a desktop computer, etc. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content distribution networks, basic cloud computing services such as big data and artificial intelligent platforms.

Referring to fig. 1 and 2, the method mainly includes the following steps S100 to S600.

S100, acquiring related parameters which influence the load of the power system at the current moment, and performing index processing on the related parameters to obtain load influence characteristics.

It should be noted that the relevant parameters mainly include parameters in weather, time, and user.

This step is intended to present the extent to which the respective relevant parameter affects the load of the power system. Meanwhile, for discrete data such as weather parameters, the indexing processing can play a certain role in coding and dimension reduction, and the prediction efficiency in the subsequent steps is improved.

And S200, acquiring historical electricity utilization data of a user, and decomposing the historical electricity utilization data to obtain historical electricity utilization seasonal indexes, historical electricity utilization wave rates and historical electricity utilization periodic characteristics.

It should be noted that, the seasonal index of the historical electricity consumption is used for characterizing the influence degree of seasonal features in the historical electricity consumption data on the power load, namely, influence factors; the historical electricity utilization periodicity characteristic is data with periodicity characteristic in the historical electricity utilization data.

In this step, the electricity consumption data is generally periodic, so that the electricity consumption data in the historical period is extremely important for load prediction. In order to extract data with periodic characteristics and seasonal influence factors from the historical electricity consumption data, the step carries out seasonal decomposition on the historical electricity consumption data so as to obtain the periodic characteristics of the historical electricity consumption and the seasonal index of the historical electricity consumption. In addition, in order to facilitate the regulation and control of the load of the power system, the step also carries out numerical processing on the historical electricity consumption data, and aims to obtain the historical electricity consumption wave dynamic rate for representing the electricity consumption fluctuation degree in the historical time period, and the relative size of load dispersion can be reflected.

S300, predicting to obtain the electricity consumption seasonal index at the future time by using the load influence characteristic and the historical electricity consumption seasonal index and combining a support vector regression model, predicting to obtain the electricity consumption periodic characteristic at the future time by using the historical electricity consumption periodic characteristic and combining a long-short-period memory neural network model, and calculating the product of the electricity consumption seasonal index at the future time and the electricity consumption periodic characteristic at the future time as a load predicted value of a user at the future time.

In this step, the historical electricity consumption seasonal index and the load influence characteristic are used as inputs of an SVR (Support Vector Regression ) model, and the seasonal index at the future time is predicted by the SVR model, so as to obtain the electricity consumption seasonal index at the future time. Meanwhile, the historical electricity utilization periodic characteristics are used as input of an LSTM (Long Short-Term Memory neural network) model, and the periodic characteristics at the future moment are predicted through the LSTM model, so that the electricity utilization periodic characteristics at the future moment are obtained. Then, the product of these two predictions is calculated and taken as the load prediction value of the user at the future time.

S400, according to the historical electricity utilization wave rate and the membership function, calculating to obtain the electricity regulation value of the user at the future moment.

In the step, the historical electricity consumption wave dynamic rate reflects the relative magnitude of load dispersion in a historical time period, and the users using the electric power system are classified by combining the historical electricity consumption wave dynamic rate with the fuzzy set theory, so that each user can have an own electricity consumption target and an electric power regulation value for regulating and controlling the electricity consumption target, and the requirement of load elastic regulation is met.

S500, acquiring a power consumption target value of a user, and judging whether the power consumption target value is larger than the sum of a load predicted value and a power regulation value. If yes, go to step S600.

In the step, the electricity consumption target value is compared with a predicted value corresponding to a certain time in the future and an adjustable value corresponding to the time, and a corresponding electricity consumption adjustment scheme is adopted according to the comparison result.

And S600, performing charge increment adjustment on the electricity consumption target value.

In this step, when the target value of electricity consumption of the user is greater than the sum of the power regulation value and the load prediction value, it is indicated that the power supplied to the user is insufficient, and then the charge increment adjustment is required to be performed on the electricity consumption of the user, so as to balance the load of the power system.

In some embodiments of the present invention, referring to fig. 1, in the control method, when the electricity consumption target value is less than or equal to the sum of the load predicted value and the electricity regulation value of the user at a future time in step S500, step S700 is entered.

And S700, performing charge down-regulation on the electricity consumption target value.

In this step, when the electricity consumption target value is less than or equal to the sum of the power regulation value and the load prediction value, it is indicated that the power supplied to the user is overloaded, and then the electricity consumption of the user needs to be subjected to charge down-regulation, so that the load of the power system is balanced.

In some embodiments of the present invention, step S100 may include, but is not limited to, steps S110-S170 as follows.

S110, acquiring relevant parameters affecting the load of the power system at the current moment, and performing data cleaning treatment.

It should be noted that, the relevant parameters include: date type information, day of the week information, and season information to which the current time belongs.

Optionally, the date type information is used to characterize a holiday type or a workday type of a date to which the current time belongs. Illustratively, the holiday type may include a mid-autumn festival, an end-noon festival, etc., and the weekday type may include a weekday and a non-weekday.

Furthermore, the relevant parameters include: the method comprises the steps of air temperature information, wind speed information, humidity information, cloud cover information, air pressure information and extreme weather event information of an environment where a power system is located at the current moment, and user type information and user development status information of the power system at the current moment.

Optionally, the user type information is used to characterize the industry type of the user using the power system at the current moment, and these types may be set according to the actual situation, which is not particularly limited by the embodiment of the present invention. Illustratively, the user types may include industries such as forestry, catering, animal husbandry, and the like.

In this step, there may be a case where data is missing due to a problem of a malfunction of the apparatus or the like, and thus it is necessary to perform data cleaning after obtaining the relevant parameters. Specifically, for the time series data, if the data is cleaned by deleting the missing data, the missing of the effective information may be caused, so that the time series data is filled by a filling method, and the filling method may be a K-nearest neighbor missing value filling method, a random forest missing value filling method, or the like.

And S120, indexing the air temperature information, the humidity information and the wind speed information to obtain a first load influence characteristic.

The first load influence feature is used for characterizing the influence degree of air temperature information, humidity information and wind speed information on the load of the electric power system, and the influence degree satisfies the following formula (1):

（1）；

in the method, in the process of the invention,representing a first load influencing featureT represents air temperature information, and the unit is DEG C; v represents wind speed information in m/s; h represents humidity information in units of; />Weight representing air temperature information +.>Weight representing wind speed information +.>A weight representing humidity information; k represents a humidity influence factor; b represents the tuning constant.

Alternatively, weights of the air temperature information, the humidity information, and the wind speed information, and humidity influence factors and adjustment constants may be set according to actual situations, which is not particularly limited in the embodiment of the present invention. The humidity influencing factor is preferably。

In this step, the influence of the air temperature on the power load is larger in general, when the air temperature is lower, the user can start the thermal insulation equipment such as the ground heating equipment to keep warm, when the air temperature is higher, the user can start the cooling equipment such as the air conditioner to remove heat, so that the power consumption is increased due to the excessively high air temperature and the excessively low air temperature, and the load of the power system is increased. In this step, the air temperature is converted into index data by adopting a quadratic function mode, the air temperature which is more comfortable on the skin surface of the human body is generally 23-25 ℃, and the average value of the air temperature is taken in the step, namely 24 ℃, so that an air temperature model is obtained, and the air temperature model maps the influence degree of the air temperature on the load of the power system.

In addition, the wind speed and the relative humidity have a certain influence on the cold and hot feeling of the human body, for example, the human body feeling is more comfortable under the weather conditions of 30 ℃ and the wind speed of 3m/s and the relative humidity of 70 percent. The prior knowledge shows that the wind speed is linearly related to the electric load, and the relative humidity is exponentially related to the electric load, so that the influence of the wind speed and the relative humidity is added on the basis of the air temperature model in the step, so that a weather model is obtained, and as shown in a formula (1), the weather model can feed back the influence of the air temperature, the relative humidity and the wind speed on the electric load, and the air temperature information, the humidity information and the wind speed information at the current moment are substituted into the weather model, so that the first load influence characteristic can be obtained.

And S130, performing index processing on the air pressure information and the cloud amount information to obtain a second load influence characteristic.

The second load influence feature is used for characterizing the influence degree of the air pressure information and the cloud amount information on the load of the power system, and the second load influence feature satisfies the following formula (2):

（2）；

in the method, in the process of the invention,representing a second load influencing feature; />Cloud amount information is represented in units of; p represents air pressure information, which is in units of kPa; />An influence factor indicating the load of the power system on a sunny day; / >Representing an influence factor of the cloud on the load of the power system; />Representing factors of influence of clouds on loads of the power system; />An influence factor of cloudy days on the load of the power system is represented; />A product of a probability of expressing rain by converting on a sunny day and an influence factor of the rain on the load of the power system; />Representing the product of the probability that the few cloud transitions represent rain and the impact factor of the rain on the load of the power system; />Representing the product of the probability that the multi-cloud conversion represents rain and the influence factor of the rain on the load of the power system; />The representation of the cloudy day transition represents the product of the probability of rain and the impact factor of rain on the load of the power system.

Alternatively, the factors of influence of cloudy, sunny, cloudy and cloudy on the load of the power system and the product of the probability of converting cloudy, sunny, cloudy and cloudy into rain and the factors of influence of rain on the load of the power system may be determined according to practical situations, and the embodiment of the present invention is not limited in particular. By way of example only, and in an illustrative,and->The value is smaller, and a numerical value between 0.02 and 0.08 is recommended to be selected; />And->The value is larger, and a value between 0.2 and 0.4 is recommended to be selected.

In this step, cloud amount information refers to the area of the cloud and the percentage of the cloud occupying the sky, and is often used by weather forecast to judge the basis of weather, and the cloud amount information can be obtained in real time by crawling local weather data websites. When the cloud cover is 0% -10%, the weather is a sunny day with a high probability; when the cloud cover is 10% -30%, the weather is in a small cloud with a high probability; when the cloud cover is 30% -70%, the weather is cloudy with high probability; when the cloud cover is more than 70%, the weather is a cloudy day with a high probability. In the weather described above, if the air pressure is reduced, there is a possibility of rain. For sunny days and cloudless conditions, the probability of rain is small and is generally small; for cloudy and overcast conditions, the probability of rain is high and is typically medium or heavy. Priori knowledge indicates that the power load is relatively low on sunny days, cloudy and cloudy conditions, and relatively high on cloudy and rainy conditions. Based on the above, the cloud amount air pressure model is built in the step, and as shown in the formula (2), the cloud amount air pressure model can feed back the influence of cloud amount and air pressure on the power load, and the cloud amount information and the air pressure information at the current moment are input into the cloud amount air pressure model, so that the second load influence characteristic can be obtained.

And S140, indexing the polar weather event information to obtain a third load influence characteristic.

It should be noted that the third load influence feature is used to characterize the influence degree of the extreme weather event information on the load of the power system, which satisfies the following formula (3):

（3）；

in the method, in the process of the invention,representing a third load influencing feature,/->Representing extreme weather event information, < >>Indicating normal weather conditions->。/>Indicating other extreme weather conditions +.>Respectively indicate->Impact factors on the load of the power system.

Alternatively, other extreme weather conditions and factors of influence of other extreme weather conditions on the load of the power system may be set according to actual conditions, which is not particularly limited by the embodiment of the present invention.

In this step, the extreme weather event information refers to some sudden or upcoming extreme weather event. Because no priori knowledge exists at present to indicate the influence degree of the extreme weather event information on the load of the power system, the dictionary mapping is adopted in the step to describe the influence degree of the extreme weather event information on the load of the power system, and as shown in a formula (3), the extreme weather event information at the current moment is input into the dictionary mapping, so that the third load influence characteristic can be obtained.

And S150, indexing the week information, the date type information and the season information to obtain a fourth load influence characteristic.

The fourth load influence feature is used to characterize the influence degree of the week information, the date type information, and the season information on the load of the power system, and satisfies the following formula (4):

（4）；

in the method, in the process of the invention,the fourth load influence characteristic is that w represents an influence factor of the day information on the load of the power system, R represents an influence factor of the date type information on the load of the power system, and Q represents an influence factor of the season information on the load of the power system. />The regional constant, i.e., the influence constant of the region in which the power system is located on the date type information, the day of the week information, and the season information, is expressed.

Alternatively, the region constant may be set according to the influence of each region, which is not particularly limited by the present invention.

In the step, according to priori knowledge, the power load on the working day is always higher than that on the non-working day, and the power load difference between the working day and the non-working day is about 20% -25%; in contrast, during major legal holidays such as spring festival or holidays, the power load is often lower than that of working day, for example, the load difference between spring festival and working day is as high as 76%. Also, the electrical load tends to be seasonal, e.g., about 20% load spread in spring versus summer and about 15% load spread in winter versus autumn. Considering that holidays and seasons have a great influence on the load of the power system, this step considers that the three are expressed in a product manner to construct a date model as shown in formula (4).

After obtaining the week information, the date type information and the season information at the current moment, mapping the influence degree of different weeks, different holidays and different seasons on the load of the power system according to the actual situation of the region where the power system is located, further obtaining the influence factors of the date type information, the week information and the season information on the load of the power system, substituting the influence factors into a date model, and obtaining the fourth load influence characteristic.

And S160, indexing processing is carried out on the user development status information and the user type information, and a fifth load influence characteristic is obtained.

It should be noted that the fifth load influence feature is used to characterize the degree of influence of the user development status information and the user type information on the load of the power system, and satisfies the following formula (5):

（5）；

in the method, in the process of the invention,representing a fifth load influencing feature, < >>The influence constant of the region where the power system is located on the user type information is represented, c represents the user type information, and d represents the user development status information.

Alternatively, for the user type information, users may be classified into first industry, second industry, third industry, etc. according to their industry properties and assigned corresponding values. For the user development status information, if the user has no data source, the information can be set. For the influence constant, if there is no user type information, the information may not be set, that is, the influence constant belongs to optional data, and as for the specific value thereof, the influence constant may be set according to the actual situation, which is not particularly limited in the embodiment of the present invention.

In this step, according to a priori knowledge, the economic condition of the user has a certain influence on the prediction of the load of the power system. For industrial type users such as textile, metal smelting, etc., the power load is often larger than that of agricultural type users such as forestry, agriculture, etc. For users who are developing economically and at high speed, the production is often required to be enlarged, so that the electric load is often increased than ever. Based on this, the user influence model is constructed in this step, and as shown in formula (5), the user influence model can feed back the influence of the type and the development status of the user on the power load, and the user development status information and the user type information at the current time are substituted into the user influence model, so that the fifth load influence characteristic can be obtained.

S170, taking the first load influence characteristic, the second load influence characteristic, the third load influence characteristic, the fourth load influence characteristic and the fifth load influence characteristic as load influence characteristics.

In some embodiments of the present invention, it is critical to extract seasonal load data that is rich in seasonal periods. Therefore, the embodiment of the invention adopts the x-11 seasonal adjustment method to decompose the historical electricity consumption data of the user into a trend component, a seasonal index, a seasonal adjustment sequence and a random component, and the seasonal adjustment sequence can be understood as the data after the influence of seasons is removed. Considering that the trend component and the random component can be obtained through calculation of the data and the seasonal index after the seasonal influence is removed, the embodiment of the invention only extracts the data and the seasonal index after the seasonal influence is removed from the historical electricity consumption data, and the data and the seasonal index are used for predicting the electric load.

Specifically, referring to fig. 2 and 3, step S200 corresponding to the seasonal decomposition process may include, but is not limited to, the steps of:

firstly, acquiring a power consumption curve of a user in a historical time period corresponding to the current moment as historical power consumption data of the user.

And then, according to the historical electricity consumption data, calculating to obtain the seasonal index of the historical electricity consumption by combining a moving average method of the multiplication model.

And finally, calculating the ratio of the historical electricity consumption data to the historical electricity consumption seasonal index as the historical electricity consumption periodic characteristic after eliminating the seasonal influence.

In the above steps, the seasonal decomposition process includes two steps of calculation of the seasonal index and seasonal adjustment, wherein the seasonal index is used for measuring the deviation degree of each time point from the average seasonal pattern, and the seasonal adjustment is used for removing the part affected by the seasonality in the original data. For calculation of seasonal indexes, the embodiment of the invention adopts a moving average method of a multiplication model to calculate, and the obtained historical electricity consumption seasonal indexes meet the following formula (6):

（6）；

in the method, in the process of the invention,indicating a seasonal index of electricity usage corresponding to the i-th moment; />Data corresponding to the ith moment in the electricity consumption curve of the user, < > >Representing data corresponding to the i-1 th moment in the electricity utilization curve of the user，/>And the data corresponding to the (i+1) th moment in the electricity utilization curve of the user is represented.

For season adjustment, the embodiment of the invention divides the original data by the seasonal index to obtain the data after eliminating the influence of the season, namely the periodic characteristic of the historical electricity consumption, as shown in the following formula (7):

（7）；

in the method, in the process of the invention,and the historical electricity utilization periodic characteristic corresponding to the ith moment is represented.

In some embodiments of the present invention, step S200 may further include, but is not limited to, the following steps:

the ratio of the average value and the standard deviation of the historical electricity data is calculated as the historical electricity consumption rate.

In the step, the load fluctuation rate refers to the load fluctuation degree in the user electricity consumption period, and can reflect the relative magnitude of load dispersion, and the load fluctuation rate can be obtained by the following formula (8):

（8）；

where s represents the standard deviation of the electricity usage curve over the historical period of time,the average of the electricity usage profile over the historical time period is shown.

In some embodiments of the present invention, in step S300, the SVR model and the LSTM model are pre-trained prediction models, the SVR model is used to predict the seasonal index of the same period according to the load influence characteristic at the current time and the seasonal index of the past year, and the LSTM model is used to predict the current year periodicity according to the periodic part of the user load.

Specifically, for the prediction process of the SVR model, the load influence characteristic at the current time and the historical electricity consumption seasonal index are used as input characteristics of the SVR model, the SVR model learns in advance a mapping relation between the historical electricity consumption seasonal index and the load influence characteristic at the current time and the electricity consumption seasonal index at the future time, and the SVR model can predict the electricity consumption seasonal index at the future time through the mapping relation.

Alternatively, a gaussian radial basis function with high versatility is selected as a kernel function of the SVR model, and an indirect loss function is employed to measure the difference between the predicted and actual values of the SVR model and minimize this difference. In particular, use is made ofAn intrinsic Loss function, intended to avoid the occurrence of training oscillations due to being too sensitive when the absolute Loss function is small in deviation, defined as: />Y represents the actual value, i.e. the seasonal index corresponding to the point in time; />Is a predictive value->Is a parameter controlling the tolerance of the model.

Specifically, for the LSTM model, the historical electricity periodic feature is used as an input feature of the LSTM model, the LSTM model learns in advance a mapping relationship between the historical electricity periodic feature and the electricity periodic feature at a future time, and the LSTM model can predict the electricity periodic feature at the future time through the mapping relationship.

Alternatively, regarding the setting of super parameters such as bias and weights thereof of the LSTM model, a random search method may be used for selection.

In some embodiments of the present invention, in step S400, users are classified using fuzzy theory, and different power regulation values are provided according to users of different membership categories, so that corresponding power regulation schemes are adopted according to correlations between the power regulation values of the users and the load prediction values of the users in subsequent steps.

Specifically, referring to fig. 2 and 4, step S400 may include, but is not limited to, the following steps S410-S420.

S410, determining a plurality of membership thresholds and a plurality of membership categories, and constructing a membership function of each membership category according to the membership thresholds and the historical electricity consumption wave rate.

It should be noted that, the membership function means that if a number a belongs to [0,1] corresponding to any element x in the range U, a is called a fuzzy set on U, and a (x) is the membership of x to a. When x varies in U, A (x) is a function called the membership function of A. In practice, the membership function may be understood as a degree of correlation, the closer the membership value is to 0, the less relevant a and x are explained; the closer the membership value is to 1, the more relevant a is to x.

In this step, first, a plurality of membership thresholds are set，/>And setting a plurality of membership categories, wherein the membership categories refer to electricity utilization categories to which users belong, and each two membership threshold values form a section. Then, utilizing the historical electricity utilization wave rate and combining the membership threshold value to construct membership functions of a plurality of membership categories. In general, users are classified into n classes according to practical situations, a corresponding membership function needs to be constructed for each class, and the sum of membership functions of all classes is 1.

Specifically, the membership function for the ith membership class is shown in the following formula (9):

（9）；

in the method, in the process of the invention,，/>a membership value representing the membership of the user to category C; />Representing membership threshold, +.>And a solving function representing the historical electricity consumption wave power in the corresponding section.

S420, obtaining the regulation and control coefficient of each membership category, and determining the power regulation and control value of the user at the future moment according to the membership function and the regulation and control coefficient of each membership category.

In this step, for each membership class, a corresponding adjustable constant is assigned, which is called an adjustment coefficient, and this adjustment coefficient may be set according to the actual regional situation. Assuming that users in a region are classified into n membership categories, the power regulation values that can be allocated to the users at future time points are shown in the following formula (10):

（10）；

In the method, in the process of the invention,the regulatory coefficients of the ith membership class are shown.

In some embodiments of the present invention, in step S500, the implementation process of obtaining the electricity consumption target value of the user mainly includes the following steps S510 to S520.

S510, determining a membership value of a user to each membership class according to a membership function of each membership class, and selecting the membership class with the largest membership value as the largest membership class;

s520, acquiring the electricity utilization target value corresponding to the maximum membership type as the electricity utilization target value of the user.

In the above steps, for each membership category, a corresponding power scheduling scheme may be set. The power scheduling scheme can be decomposed into a power target value and power excitation measure information of the users belonging to the class. When the maximum membership type is determined, acquiring an electricity quantity scheduling scheme of the maximum membership type, extracting a corresponding electricity utilization target value from the electricity quantity scheduling scheme, and taking the electricity utilization target value as an electricity utilization target value of a user.

In some embodiments of the present invention, the method may further include the steps of, before the charge up-regulation in step S600 and before the down-regulation in step S700:

Firstly, determining a membership value of a user belonging to each membership class according to a membership function of each membership class, and selecting the membership class with the largest membership value as the largest membership class;

and then, obtaining the electricity utilization excitation measure information corresponding to the maximum membership type.

In the above steps, for each membership category, a corresponding power scheduling scheme may be set. The power scheduling scheme can be decomposed into a power target value and power excitation measure information of the users belonging to the class. When the maximum membership type is determined, acquiring an electricity quantity scheduling scheme of the maximum membership type, and extracting corresponding electricity utilization excitation measure information from the electricity quantity scheduling scheme.

In some embodiments of the present invention, when the charge up-regulation is performed in step S600 and when the charge down-regulation is performed in step S700, the method may further include the steps of:

and sending electricity consumption excitation measure information to a terminal where the user is located so as to excite the user to consume electricity or to recommend the user to reduce the electricity consumption.

The principle of the method provided by the invention will be illustrated by way of an example. Referring to fig. 2, in this example:

The method comprises the steps of firstly, collecting relevant parameters of a power system in weather, time and user at the current moment, and obtaining a power utilization curve of a user in a historical time period.

And secondly, decomposing the electricity utilization curve by using an x-11 seasonal decomposition method to obtain the periodic characteristics of the historical electricity utilization and the seasonal index of the historical electricity utilization, and carrying out numerical processing on the electricity utilization curve to obtain the historical electricity utilization rate. At the same time, the relevant parameters are converted into a plurality of characteristic indexes, and further load influence characteristics are obtained.

Thirdly, taking the load influence characteristic and the historical electricity consumption seasonal index as the input of an SVR model, predicting the SVR model to obtain the electricity consumption seasonal index at the future time, taking the historical electricity consumption periodic characteristic as the input of an LSTM model, predicting the LSTM to obtain the electricity consumption periodic characteristic at the future time, and calculating the product of the two prediction results to be taken as the load predicted value of the user at the future time.

Fourth, according to the historical electricity consumption wave rate, combining the membership function of each membership class to obtain the electricity regulation value of the user at the future moment.

And fifthly, determining a membership class with the largest membership value as the largest membership class according to the membership value of each membership class of the user, and then acquiring an electricity quantity scheduling scheme of the largest membership class, wherein the electricity quantity scheduling scheme can be decomposed into electricity consumption target values and electricity consumption excitation measure information.

Sixthly, judging whether the electricity target value is larger than the sum of the electric power regulation value and the load predicted value; if yes, entering a sixth step; if not, entering a seventh step;

and sixthly, performing charge increment adjustment on the electricity consumption target value, and sending electricity consumption excitation measure information to a user.

And seventhly, performing charge down regulation on the electricity consumption target value, and sending electricity consumption excitation measure information to a user.

In summary, the method provided by the embodiment of the invention can provide the following technical effects:

on the one hand, the data such as weather, date and user type are subjected to index processing, and an index function which accords with the actual situation is built according to the mutual influence relation among all the characteristics, so that the dimension of the prediction model is reduced, and the prediction efficiency of the prediction model is improved. In addition, the indexing processing can extract the influence condition of each data on the user load, so that the user can check the influence coefficient of each data conveniently, the accuracy of load prediction is improved, and meanwhile, the user can adjust the corresponding influence coefficient according to the regional requirement, so that the manpower and material resources in the aspect of model prediction are greatly reduced.

On the other hand, different components of the user load are predicted by using the long-short-period neural network and the support vector regression prediction model, so that different models can predict characteristic data with different characteristics in the respective good fields, the advantages of each prediction model are exerted to the greatest extent, and the accuracy of load prediction is improved.

On the other hand, the fuzzy set theory is used for classifying the electricity utilization users, compared with the traditional charge classification scheme, the fuzzy set classification can be used for configuring the corresponding adjustable quantity and electricity utilization target for each user more flexibly and more personally, and the electric power system can perform load balancing processing according to the preset electricity utilization scheme and corresponding measures, so that the elastic requirement of the user on the demand side is better met, and the flexibility of the load balancing processing system is greatly improved.

In addition, the embodiment of the invention also provides a power system, which is used for providing power for a plurality of users and adjusting the power load by adopting the power system load balancing control method.

Similarly, the content in the above method embodiment is applicable to the system embodiment, and the functions specifically implemented by the system embodiment are the same as those of the above method embodiment, and the beneficial effects achieved by the method embodiment are the same as those achieved by the above method embodiment.

In some alternative embodiments, the functions/acts noted in the block diagrams may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Furthermore, the embodiments presented and described in the flowcharts of the present invention are provided by way of example in order to provide a more thorough understanding of the technology. The disclosed methods are not limited to the operations and logic flows presented herein. Alternative embodiments are contemplated in which the order of various operations is changed, and in which sub-operations described as part of a larger operation are performed independently.

Furthermore, while the invention is described in the context of functional modules, it should be appreciated that, unless otherwise indicated, one or more of the functions and/or features may be integrated in a single physical device and/or software module or may be implemented in separate physical devices or software modules. It will also be appreciated that a detailed discussion of the actual implementation of each module is not necessary to an understanding of the present invention. Rather, the actual implementation of the various functional modules in the apparatus disclosed herein will be apparent to those skilled in the art from consideration of their attributes, functions and internal relationships. Accordingly, one of ordinary skill in the art can implement the invention as set forth in the claims without undue experimentation. It is also to be understood that the specific concepts disclosed are merely illustrative and are not intended to be limiting upon the scope of the invention, which is to be defined in the appended claims and their full scope of equivalents.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in the form of a software product stored in a storage medium, including several programs for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable programs for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with a program execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the programs from the program execution system, apparatus, or device and execute the programs. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the program execution system, apparatus, or device.

More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

It is to be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented in hardware, as in another embodiment, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.

In the foregoing description of the present specification, reference has been made to the terms "one embodiment/example", "another embodiment/example", "certain embodiments/examples", and the like, means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

While embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that: many changes, modifications, substitutions and variations may be made to the embodiments without departing from the spirit and principles of the invention, the scope of which is defined by the claims and their equivalents.

While the preferred embodiment of the present invention has been described in detail, the present invention is not limited to the embodiments described above, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the present invention, and these equivalent modifications and substitutions are intended to be included in the scope of the present invention as defined in the appended claims.

Claims

1. The power system load balancing control method is characterized by comprising the following steps of:

acquiring a power consumption target value of a user, and performing charge increment adjustment on the power consumption target value when the power consumption target value is larger than the sum of the load predicted value and the power regulation value;

the method for obtaining the relevant parameters influencing the load of the power system at the current moment comprises the steps of:

；

2. The method for controlling load balancing of a power system according to claim 1, wherein the step of obtaining the historical power consumption data of the user, and performing a decomposition process on the historical power consumption data to obtain a historical power consumption seasonal index, a historical power consumption wave power rate and a historical power consumption periodicity feature comprises:

3. The method for controlling load balancing of a power system according to claim 1, wherein the step of obtaining the historical power consumption data of the user, and performing a decomposition process on the historical power consumption data to obtain a historical power consumption seasonal index, a historical power consumption wave power rate and a historical power consumption periodicity feature comprises:

4. The method for controlling load balancing of an electric power system according to claim 3, wherein said calculating an electric power regulation value of a user at a future time according to said historical electric power consumption rate in combination with a membership function comprises:

determining a plurality of membership thresholds and a plurality of membership categories, and constructing a membership function of each membership category according to the membership thresholds and the historical electricity consumption wave power;

and acquiring a regulation and control coefficient of each membership class, and determining the power regulation and control value of the user at the future moment according to the membership function and the regulation and control coefficient of each membership class.

5. The method for controlling load balancing of an electric power system according to claim 1, wherein said obtaining the electricity consumption target value of the user comprises:

6. The method for controlling load balancing of an electric power system according to claim 1, further comprising the steps of:

7. The power system load balancing control method according to claim 6, characterized in that before the electric power consumption target value is subjected to electric charge down-regulation or electric charge up-regulation, the method further comprises the steps of:

8. The power system load balancing control method according to claim 7, wherein when the electricity consumption target value is subjected to charge down-regulation or charge up-regulation, the method further comprises the steps of:

9. An electrical power system for providing electrical power to a plurality of consumers, wherein the electrical power system employs an electrical power system load balancing control method as claimed in any one of claims 1 to 8 for regulating electrical power load.