CN115374404B - Method for correcting monthly electric energy proportion deviation of industry based on multi-dimensional data - Google Patents
Method for correcting monthly electric energy proportion deviation of industry based on multi-dimensional data Download PDFInfo
- Publication number
- CN115374404B CN115374404B CN202211288547.7A CN202211288547A CN115374404B CN 115374404 B CN115374404 B CN 115374404B CN 202211288547 A CN202211288547 A CN 202211288547A CN 115374404 B CN115374404 B CN 115374404B
- Authority
- CN
- China
- Prior art keywords
- electric energy
- information
- enterprise
- industry
- dimensional
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention provides a correction method for monthly electric energy proportion deviation of industry based on multi-dimensional data, wherein a server determines historical second electric energy information corresponding to a second enterprise tag in a database, and calculates according to the second electric energy information to obtain one-dimensional electric energy calculation information corresponding to the second enterprise tag; calculating based on the third electric energy information and the first electric energy information to obtain industry electric energy trend information, and performing multidimensional calculation based on the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to the second enterprise label; the server corrects the first electric energy data according to the multi-dimensional electric energy calculation information to obtain second electric energy data, and the monthly electric energy ratio of the industry is obtained according to the first electric energy data and/or the second electric energy data of all the industries. The invention can accurately predict the electric energy data lost by the enterprise and correct the deviation of the monthly electric energy ratio of the industry on the monthly electric energy of the industry.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a correction method for monthly electric energy proportion deviation of an industry based on multi-dimensional data.
Background
In the process of data statistics, industry monthly electric energy duty ratios can be obtained according to the proportion of monthly electric energy of each industry to the sum of the monthly electric energy of all the industries, in the actual duty ratio calculation process, the accuracy of each industry monthly electric energy duty ratio depends on the integrity and the accuracy of the acquired data, but in the actual electric energy data statistics process, certain data cannot be acquired and certain data are lost due to various reasons, and further the industry monthly electric energy duty ratio possibly has errors, so that a technical scheme is urgently needed, relevance calculation can be performed according to the data which cannot be acquired and lost, and deviation correction can be performed on the industry monthly electric energy duty ratio on the industry monthly electric energy.
Disclosure of Invention
The embodiment of the invention provides a method for correcting the monthly electric energy duty ratio deviation of the industry based on multi-dimensional data, which can accurately predict the electric energy data lost by enterprises and correct the monthly electric energy duty ratio deviation of the industry.
The method for correcting the monthly electric energy proportion deviation of the industry based on the multi-dimensional data comprises the following steps:
the server receives first electric energy information sent by all electric energy acquisition ends in a target area at the current moment at intervals of a first preset time period, and calculates the first electric energy information of all the same industries according to industry labels to obtain first electric energy data;
the server takes the enterprise tag receiving the first electric energy information as a first enterprise tag, takes the enterprise tag which does not send the first electric energy information as a second enterprise tag, determines historical second electric energy information corresponding to the second enterprise tag in a database, and calculates according to the second electric energy information to obtain one-dimensional electric energy calculation information corresponding to the second enterprise tag;
the server acquires historical third electric energy information corresponding to a first enterprise tag, calculates based on the third electric energy information and the first electric energy information to obtain industry electric energy trend information, and calculates in a multi-dimensional manner based on the industry electric energy trend information and one-dimensional electric energy calculation information to obtain multi-dimensional electric energy calculation information corresponding to a second enterprise tag;
and the server corrects the first electric energy data according to the multi-dimensional electric energy calculation information to obtain second electric energy data, and obtains the monthly electric energy ratio of the industry according to the first electric energy data and/or the second electric energy data of all industries.
Further, before the step of receiving, by the server, the first electric energy information sent by all the electric energy collection terminals in the target area at the current time at intervals of a first preset time period, the method further includes:
classifying all enterprises in a target area according to corresponding industry types to obtain a plurality of enterprise industry sets, setting an electric energy acquisition end at a corresponding enterprise according to the enterprise industry sets, and adding an industry label and an enterprise label to the electric energy acquisition end;
when the electric energy acquisition end acquires first electric energy information of a corresponding enterprise, corresponding industry labels and enterprise labels are added to the first electric energy information.
Further, the server uses the enterprise tag that receives the first electric energy information as a first enterprise tag, uses the enterprise tag that does not send the first electric energy information as a second enterprise tag, determines historical second electric energy information corresponding to the second enterprise tag in the database, and calculates according to the second electric energy information to obtain one-dimensional electric energy calculation information corresponding to the second enterprise tag, including:
the server compares the information quantity of the first electric energy information with the label quantity of the first enterprise label, and if the information quantity does not correspond to the label quantity, the server selects an enterprise industry set of a corresponding industry;
according to the enterprise industry set, taking an enterprise tag corresponding to an electric energy acquisition end which sends first electric energy information as a first enterprise tag, and taking an enterprise tag which does not send the first electric energy information as a second enterprise tag;
extracting historical second electric energy information corresponding to the second enterprise label in the database, calculating a difference value between the second electric energy information at the later moment and the electric energy information at the previous moment in all the second electric energy information to obtain a change difference value, and obtaining a predicted change rate according to the change difference value;
calculating according to the predicted change rate and second electric energy information closest to the current moment to obtain one-dimensional electric energy calculation information;
calculating to obtain one-dimensional electric energy calculation information corresponding to the second enterprise label through the following formula,
wherein the content of the first and second substances,the information is calculated for the one-dimensional electrical energy,the second power information closest to the current time,is as followsThe second power information of the individual time instant,is a firstThe second power information of the individual time instant,is the upper limit value of the moment corresponding to the second electric energy information,in order to vary the number of the difference values,is a reference weight value.
Further, the step of obtaining, by the server, third historical electric energy information corresponding to the first enterprise tag, performing calculation based on the third electric energy information and the first electric energy information to obtain industry electric energy trend information, and performing multidimensional calculation based on the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to the second enterprise tag includes:
counting historical third electric energy information corresponding to all the first enterprise tags in a second preset time period to obtain a first historical information set;
sequentially traversing each piece of third electric energy information, deleting the non-real third electric energy information to obtain a second historical information set, and selecting all first electric energy information corresponding to each enterprise tag in the second historical information set to obtain a first current information set;
calculating based on the second historical information set and the first current information set to obtain industry electric energy trend information, and performing multidimensional calculation according to the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to the second enterprise label;
industry electric energy trend information is calculated by the following formula,
wherein the content of the first and second substances,as the information on the trend of the electric energy,is the first in the first current information setFirst power information for a first enterprise tag,is the first in the first historical information setThird power information for the first enterprise tag,is an upper limit value for the first number of enterprise tags in the first current set of information,is the quantity value of the first enterprise tag in the first current set of information,to get it towardsA potential weight value.
Further, the obtaining of the multidimensional electric energy calculation information corresponding to the second enterprise tag by performing multidimensional calculation according to the industry electric energy trend information and the one-dimensional electric energy calculation information includes:
if the predicted change rate is larger than the industry electric energy trend information, reducing and adjusting the one-dimensional electric energy calculation information according to the predicted change rate and the industry electric energy trend information to obtain multi-dimensional electric energy calculation information;
if the predicted change rate is smaller than the industry electric energy trend information, increasing and adjusting the one-dimensional electric energy calculation information according to the predicted change rate and the industry electric energy trend information to obtain multi-dimensional electric energy calculation information;
the multidimensional electric energy calculation information is obtained by the following formula,
wherein the content of the first and second substances,for the adjusted multi-dimensional electrical energy calculation information,is a first value of a normalization constant that,in order to reduce the value of the adjustment constant,for the second value of the normalization constant, the value of the constant,to increase the adjustment constant value.
Further, the server corrects the first electric energy data according to the multidimensional electric energy calculation information to obtain second electric energy data, and obtains an industry monthly electric energy ratio according to the first electric energy data and/or the second electric energy data of all industries, including:
acquiring first electric energy data and/or second electric energy data of all industries to obtain electric energy data sum;
if all enterprise tags of any one industry are judged to be first enterprise tags, obtaining a first industry monthly electric energy ratio according to the ratio of the first electric energy data and the sum of the electric energy data of the corresponding industry;
if the second enterprise tag exists in all enterprise tags of any one industry, obtaining the monthly electric energy ratio of the second industry according to the ratio of the second electric energy data of the corresponding industry to the sum of the electric energy data;
the first industry monthly electric energy proportion and the second industry monthly electric energy proportion are calculated by the following formula,
wherein the content of the first and second substances,is as followsA first operating month electric energy fraction of the first electric energy data,is as followsThe first power data is transmitted to the first power receiver,is as followsA first power data of an individual industry,is the upper limit value of the industry corresponding to the first electric energy data,is as followsSecond power data of the individual industry,is the upper limit value of the industry corresponding to the second electric energy data,is as followsA second industry monthly power fraction of the second power data,is as followsAnd second power data.
Further, if it is judged that an enterprise of any one industry uploads corresponding first electric energy information after a first preset time period, the first electric energy information is used as training electric energy information;
comparing the training electric energy information with the multi-dimensional electric energy calculation information, and if the difference value between the training electric energy information and the multi-dimensional electric energy calculation information is larger than a first training value, inputting the training electric energy information and the multi-dimensional electric energy calculation information into a training model;
and the training model updates and trains the reference weight value and/or the trend weight value based on the difference between the training electric energy information and the multi-dimensional electric energy calculation information.
Further, the training model performs update training on a reference weight value and/or a trend weight value based on a difference between the training electric energy information and the multidimensional electric energy calculation information, and includes:
if the training electric energy information is judged to be larger than the multi-dimensional electric energy calculation information, increasing and adjusting the reference weight value and/or the trend weight value according to the difference value between the training electric energy information and the multi-dimensional electric energy calculation information;
if the training electric energy information is judged to be smaller than the multi-dimensional electric energy calculation information, reducing and adjusting the reference weight value and/or the trend weight value according to the difference value between the training electric energy information and the multi-dimensional electric energy calculation information;
the training model trains and adjusts the reference weight value and/or the trend weight value through the following formula,
wherein the content of the first and second substances,in order to train the information on the electric energy,to train the adjusted baseline weight values and/or the trend weight values,to train the baseline weight values and/or the trending weight values prior to adjustment,the coefficients are adjusted for the purpose of forward training,the coefficients are adjusted for reverse training.
Further, if it is determined that the historical second electric energy information corresponding to the second enterprise tag in the database is 0, second enterprise attribute information corresponding to the second enterprise tag is acquired;
acquiring first enterprise attribute information corresponding to each first enterprise tag, and acquiring a similarity coefficient based on the similarity of each first enterprise attribute information and the second enterprise attribute information;
determining first electric energy information corresponding to first enterprise attribute information with the highest similarity coefficient as the closest electric energy information;
and obtaining the difference between the first enterprise attribute information with the highest similarity coefficient and the second enterprise attribute information to obtain difference correction information, and correcting the closest electric energy information based on the difference correction information to obtain multi-dimensional electric energy calculation information.
Further, the obtaining of the first enterprise attribute information corresponding to each first enterprise tag and the similarity coefficient based on the similarity between each first enterprise attribute information and the second enterprise attribute information specifically includes:
acquiring a first enterprise number, a first enterprise area and first production value information in the first enterprise attribute information, and acquiring a second enterprise number, a second enterprise area and second production value information which respectively correspond to the second enterprise number, the second enterprise area and the second production value information in the second enterprise attribute information;
calculating according to the number of the first enterprises, the area of the first enterprises, the first production value information, the number of the second enterprises, the area of the second enterprises and the second production value information to obtain a similarity coefficient,
wherein the content of the first and second substances,for the second enterprise attribute information anda similarity coefficient of the first business attribute information,is a weight value of the number of people,is as followsA second number of business persons in the second business attribute information,is a firstA first number of business persons in the first business attribute information,is the weight value of the area, and is,is as followsA second enterprise area in the second enterprise attribute information,is as followsA first business area in the first business attribute information,in order to produce a value weight value,is a firstSecond production value information in the second business attribute information,is a firstFirst production value information of the first business attribute information.
Further, the obtaining of the difference between the first enterprise attribute information with the highest similarity coefficient and the second enterprise attribute information to obtain difference correction information, and correcting the closest electric energy information based on the difference correction information to obtain multidimensional electric energy calculation information includes:
obtaining a people number difference value according to the number of people of the first enterprise and the number of people of the second enterprise, obtaining an area difference value according to the area of the first enterprise and the area of the second enterprise, and obtaining a production value difference value according to the first production value information and the second production value information;
carrying out multi-dimensional fusion calculation on the people number difference value, the area difference value and the output value difference value to obtain difference correction information, calculating the difference correction information through the following formula,
wherein the content of the first and second substances,in order to correct the information for the difference,the first number of persons of the first business attribute information with the highest similarity coefficient,is a weight of the difference in the number of people,the first enterprise area of the first enterprise attribute information with the highest similarity coefficient,is the weight of the area difference,first production value information of the first business attribute information with the highest similarity coefficient,to produce difference weights.
Further, the correcting the closest electric energy information based on the difference correction information to obtain multidimensional electric energy calculation information includes:
if the difference correction information is larger than 0, forward correction is carried out on the nearest electric energy information according to the difference correction information to obtain multidimensional electric energy calculation information;
if the difference correction information is smaller than 0, carrying out negative correction on the nearest electric energy information according to the difference correction information to obtain multi-dimensional electric energy calculation information;
the multidimensional electric energy calculation information is obtained by the following formula,
wherein, the first and the second end of the pipe are connected with each other,the information is calculated for the multi-dimensional electrical energy,in order to be the closest to the power information,constant values are calculated for the multiple dimensions.
The effects are as follows:
1. according to the scheme, when the enterprise electric energy data is lost, one-dimensional electric energy calculation information is obtained by utilizing the enterprise historical data, a corresponding enterprise is found by utilizing an industry label, and the data of the corresponding enterprise is utilized to obtain a corrected value so as to correct the one-dimensional electric energy calculation information and obtain more accurate multi-dimensional electric energy calculation information; in addition, when the corrected value is calculated, the electric energy trend information of the industry is calculated, the data of an enterprise are screened, the unreal data are deleted, and the one-dimensional electric energy calculation information is corrected after the corrected value is more accurate, so that the multi-dimensional electric energy calculation information is more accurate; meanwhile, the monthly electric energy proportion of the enterprise can be calculated;
2. according to the scheme, when one-dimensional electric energy calculation information is calculated, the reference weight value is adjusted, when industry electric energy trend information is calculated, the trend weight value is adjusted, real data uploaded by an enterprise in a supplementing mode is used for adjusting a prediction model of the scheme, the two weights are trained, the one-dimensional electric energy calculation information and the industry electric energy trend information are adjusted more accurately, and therefore more accurate multi-dimensional electric energy calculation information is obtained;
3. according to the scheme, the condition that the enterprise has no historical data is considered, the multidimensional data of enterprise attributes are referred, the data of the closest enterprise are found, the difference correction information between the two enterprises is calculated, the lost data of the enterprises is predicted by using the difference correction information, and the prediction and the supplement of the lost data of the enterprises can be comprehensively realized.
Drawings
Fig. 1 is a schematic diagram of an application scenario provided in an embodiment of the present invention;
FIG. 2 is a schematic flow chart of the correction method for monthly electric energy percentage deviation of industry based on multi-dimensional data.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein.
It should be understood that, in the various embodiments of the present invention, the sequence numbers of the processes do not mean the execution sequence, and the execution sequence of the processes should be determined by the functions and the internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
It should be understood that in the present application, "comprising" and "having" and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements explicitly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that, in the present invention, "a plurality" means two or more. "and/or" is merely an association describing an associated object, meaning that three relationships may exist, e.g., a and/or B, may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "comprises A, B and C" and "comprises A, B, C" means that A, B, C all comprise, "comprises A, B or C" means that one of three A, B, C is comprised, "comprises A, B and/or C" means that any 1 or any 2 or 3 of three A, B, C are comprised.
It should be understood that in the present invention, "B corresponding to a", "a corresponds to B", or "B corresponds to a" means that B is associated with a, and B can be determined from a. Determining B from a does not mean determining B from a alone, but may be determined from a and/or other information. And the matching of A and B means that the similarity of A and B is greater than or equal to a preset threshold value.
As used herein, "if" can be interpreted as "at … …" or "at … …" or "in response to a determination" or "in response to a detection", depending on context.
The technical solution of the present invention will be described in detail below with specific examples. These several specific embodiments may be combined with each other below, and details of the same or similar concepts or processes may not be repeated in some embodiments.
As mentioned in the background art, in the prior art, in the actual power data statistics process, some data cannot be collected and some data are lost due to various reasons, for example, power data is lost and cannot be recovered due to network transmission, for example, the power collection end is damaged and cannot be normally measured and cannot be found in time, and the data cannot be recovered due to insufficiencies, the scheme performs correlation calculation on the data, and obtains predicted data to supplement the missing data.
Referring to fig. 1, a schematic view of an application scenario provided by an embodiment of the present invention includes a server and an electric energy collection end, where the electric energy collection end is connected to the server, the electric energy collection end sends collected data to the server, and the server processes the data. The electric energy acquisition terminals are arranged in a plurality of numbers, and in practical application, the electric energy acquisition terminals can be arranged in each enterprise of a monitoring area and used for acquiring electric energy data of each enterprise.
Referring to fig. 2, it is a schematic flow chart of the method for correcting industry monthly electric energy percentage deviation based on multidimensional data according to the present invention, and an execution subject of the method shown in fig. 2 may be a software and/or hardware device. The execution subject of the present application may include, but is not limited to, at least one of: user equipment, network equipment, etc. The user equipment may include, but is not limited to, a computer, a smart phone, a Personal Digital Assistant (PDA), the above mentioned electronic equipment, and the like. The network device may include, but is not limited to, a single network server, a server group of multiple network servers, or a cloud of numerous computers or network servers based on cloud computing, wherein cloud computing is one type of distributed computing, a super virtual computer consisting of a cluster of loosely coupled computers. The present embodiment does not limit this. The method for correcting the monthly electric energy proportion deviation of the industry based on the multi-dimensional data comprises the following steps of S1 to S4:
s1, a server receives first electric energy information sent by all electric energy acquisition ends in a target area at the current moment at intervals of a first preset time period, and calculates the first electric energy information of all the same industries according to industry labels to obtain first electric energy data.
In the scheme, the server can receive the electric energy data collected by the electric energy collection end in the target area at regular time, and then the industry labels are utilized to calculate the first electric energy information of all the same industries to obtain the first electric energy data.
The industry label may refer to the type of industry to which an enterprise belongs, for example, if enterprise a is an enterprise that manufactures lamps, then its industry label may be "lamp manufacturing"; the first preset time period may be monthly, and the target area may be, for example, for a town or a city, which is not limited by the present embodiment.
It can be understood that the scheme utilizes the industry to perform classification calculation on the electric energy data in the target area to obtain a plurality of first electric energy data. It should be noted that the first power data and the industry tag are in one-to-one correspondence.
In order to classify the enterprise, in some embodiments, before the step of receiving, by the server, the first power information sent by all power collection terminals in the target area at the current time at intervals of a first preset time period, the method further includes:
classifying all enterprises in a target area according to corresponding industry types to obtain a plurality of enterprise industry sets, setting an electric energy acquisition end at a corresponding enterprise according to the enterprise industry sets, and adding an industry label and an enterprise label to the electric energy acquisition end;
when the electric energy acquisition end acquires first electric energy information of a corresponding enterprise, corresponding industry labels and enterprise labels are added to the first electric energy information.
According to the scheme, enterprises can be classified according to industry types to obtain a plurality of enterprise industry sets, corresponding industry labels and enterprise labels are added to the electric energy collection ends of the enterprises, and therefore the electric energy collection ends can directly mark first electric energy information when the first electric energy information is uploaded, namely the first electric energy information can contain the industry labels and the enterprise labels.
According to the scheme, after the first electric energy information is collected, the industry label and the enterprise label in the first electric energy information can be directly analyzed, the first electric energy information is directly and accurately classified into the corresponding first electric energy data, and the first electric energy information does not need to be traced and inquired.
And S2, the server takes the enterprise label receiving the first electric energy information as a first enterprise label, takes the enterprise label not sending the first electric energy information as a second enterprise label, determines historical second electric energy information corresponding to the second enterprise label in the database, and calculates according to the second electric energy information to obtain one-dimensional electric energy calculation information corresponding to the second enterprise label.
It can be understood that, for example, there are 100 enterprises in the target area, 99 enterprises upload corresponding first power information, and 1 enterprise does not upload the first power information, and then the 99 enterprises are used as the first enterprise tags, and the 1 enterprise is used as the second enterprise tag.
After the second enterprise tag is determined, the scheme finds historical second electric energy information corresponding to the second enterprise tag from the database, and then calculates one-dimensional electric energy calculation information predicted by the enterprise by using the second electric energy information.
For example, when the current month is month 4, and enterprise a does not upload the energy data of month 4, the present solution marks enterprise a as a second enterprise tag, and then finds historical second energy information of enterprise a from the database, for example, the second energy information of month 1, month 2, and month 3, and calculates the predicted one-dimensional energy calculation information using the second energy information of month 1, month 2, and month 3.
In some embodiments, the step of using, by the server, the enterprise tag that receives the first electric energy information as a first enterprise tag, using the enterprise tag that does not send the first electric energy information as a second enterprise tag, determining historical second electric energy information corresponding to the second enterprise tag in the database, and performing calculation according to the second electric energy information to obtain one-dimensional electric energy calculation information corresponding to the second enterprise tag includes:
the server compares the information quantity of the first electric energy information with the label quantity of the first enterprise label, and if the information quantity does not correspond to the label quantity, the server selects an enterprise industry set of the corresponding industry. According to the scheme, the information quantity of the first electric energy information and the label quantity of the first enterprise label are utilized to judge whether the enterprise does not upload data, if the quantity does not correspond to the label quantity, the fact that the enterprise does not upload data is indicated, the enterprise has a corresponding industry label, and at the moment, the server selects an enterprise industry set of a corresponding industry according to the information quantity.
And according to the enterprise industry set, taking the enterprise label corresponding to the electric energy acquisition end which sends the first electric energy information as a first enterprise label, and taking the enterprise label which does not send the first electric energy information as a second enterprise label. The scheme marks the enterprise uploading data in the enterprise industry set as a first enterprise label and marks the enterprise not uploading data as a second enterprise label.
And extracting historical second electric energy information corresponding to the second enterprise label in the database, calculating a difference value between the second electric energy information at the later moment and the electric energy information at the previous moment in all the second electric energy information to obtain a change difference value, and obtaining a predicted change rate according to the change difference value. For example, the second power information may be power data of 1 month, 2 months and 3 months, and the scheme may calculate a difference between 3 months and 2 months, further calculate a difference between 2 months and 1 month, calculate a variation difference, and sum and average the variation differences to obtain the predicted variation rate.
And calculating according to the predicted change rate and the second electric energy information closest to the current moment to obtain one-dimensional electric energy calculation information. According to the scheme, after the predicted change rate is obtained, the second electric energy information (the electric energy information in 3 months) closest to the current moment is calculated by using the predicted change rate, and one-dimensional electric energy calculation information is obtained. It should be noted that, in the scheme, the second electric energy information closest to the current time is used for calculation, and is closer to the current time, so that the reference degree is higher, and the prediction accuracy can be improved.
Calculating to obtain one-dimensional electric energy calculation information corresponding to the second enterprise label through the following formula,
wherein, the first and the second end of the pipe are connected with each other,the information is calculated for the one-dimensional electrical energy,the second power information closest to the current time,is as followsThe second power information of the individual time instant,is as followsThe second power information of the individual time instant,is the upper limit value of the moment corresponding to the second electric energy information,in order to vary the number of difference values,is a reference weight value.
In the above-mentioned formula,representing the difference in the variation of the phase,representing predicted rate of change, reference weight valueThe method is used for calibrating the calculated one-dimensional electric energy calculation information, so that the method is more accurate.
And S3, the server acquires third historical electric energy information corresponding to the first enterprise label, calculates based on the third electric energy information and the first electric energy information to obtain industry electric energy trend information, and calculates in a multi-dimensional mode based on the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multi-dimensional electric energy calculation information corresponding to the second enterprise label.
In the scheme, the one-dimensional electric energy calculation information calculated in the step S2 is not accurate enough, and the one-dimensional electric energy calculation information is calibrated again to obtain more accurate multi-dimensional electric energy calculation information.
According to the scheme, historical third electric energy information corresponding to a normal enterprise (an enterprise which uploads data normally) is obtained, for example, electric energy data of 3 months, then, electric energy data (first electric energy information) of 4 months of the normal enterprise and electric energy data (third electric energy information) of 3 months are used for calculating industry electric energy trend information, and industry electric energy trend information is used for calibrating one-dimensional electric energy calculation information to obtain more accurate multi-dimensional electric energy calculation information.
In some embodiments, the obtaining, by the server, third historical electric energy information corresponding to a first enterprise tag, performing calculation based on the third electric energy information and the first electric energy information to obtain industry electric energy trend information, and performing multidimensional calculation based on the industry electric energy trend information and one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to a second enterprise tag includes:
and counting historical third electric energy information corresponding to all the first enterprise tags in a second preset time period to obtain a first historical information set. For example, according to the scheme, historical third electric energy information corresponding to all first enterprise tags (enterprises which upload data normally), for example, electric energy data in 3 months, is acquired, and all data is collected into a first historical information set.
And traversing each piece of third electric energy information in sequence, deleting the non-real third electric energy information to obtain a second historical information set, and selecting all first electric energy information corresponding to each enterprise label in the second historical information set to obtain a first current information set. According to the scheme, the third non-true electric energy information also exists in the first historical information set, and can be deleted, so that the data in the second historical information set are guaranteed to be true and accurate, and the subsequent data can be accurately predicted. For example, if the 3 month power data for enterprise a is also predicted in the first historical information set, it is the third non-real power information, which the present scheme deletes from the first historical information set.
And calculating based on the second historical information set and the first current information set to obtain industry electric energy trend information, and performing multidimensional calculation according to the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to the second enterprise label. The scheme can calculate industry electric energy trend information by utilizing the electric energy data (first electric energy information) of a plurality of normal enterprises in month 4 and the electric energy data (third electric energy information) of month 3.
Industry electric energy trend information is calculated by the following formula,
wherein the content of the first and second substances,as the information on the trend of the electric energy,is the first in the first current information setFirst power information for a first enterprise tag,is the first in the first historical information setThird power information for the first enterprise tag,is an upper limit value for the first number of enterprise tags in the first current set of information,is the quantity value of the first enterprise tag in the first current set of information,is a trend weight value.
In the above-mentioned formula,representing a difference between the first power information and the third power information for an enterprise,representing industry electric energy trend information and trend weighted valueFor obtaining electric energy trend informationCalibrating to obtain more accurate electric energy trend information。
In order to calibrate the one-dimensional electric energy calculation information by using the industry electric energy trend information, in some embodiments, the performing multidimensional calculation according to the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to the second enterprise tag includes:
and if the predicted change rate is greater than the industry electric energy trend information, reducing and adjusting the one-dimensional electric energy calculation information according to the predicted change rate and the industry electric energy trend information to obtain multi-dimensional electric energy calculation information. The predicted change rate of the scheme is calculated according to the data of the dimensionality of the enterprise, the industry electric energy trend information is calculated according to the data of all enterprises of corresponding industries in the target area, and when the predicted change rate is larger than the industry electric energy trend information, the predicted change rate in the scheme is larger, and smaller adjustment needs to be carried out on the predicted change rate.
And if the predicted change rate is smaller than the industry electric energy trend information, increasing and adjusting the one-dimensional electric energy calculation information according to the predicted change rate and the industry electric energy trend information to obtain multi-dimensional electric energy calculation information. The predicted change rate of the scheme is calculated according to the data of the dimension of the enterprise, the industry electric energy trend information is calculated according to the data of all enterprises of the corresponding industry in the target area, and when the predicted change rate is smaller than the industry electric energy trend information, the predicted change rate in the scheme is smaller and needs to be adjusted.
The multidimensional electric energy calculation information is obtained by the following formula,
wherein the content of the first and second substances,for the adjusted multi-dimensional electrical energy calculation information,is a first value of a normalization constant that,in order to reduce the value of the adjustment constant,for the second value of the normalization constant, the value of the constant,to increase the adjustment constant value.
In the above-mentioned formula,the representative predicted change rate is larger than the industry electric energy trend information, which shows that the predicted change rate in the scheme is larger, so that the information is calculated for the one-dimensional electric energyThe adjustment is small, and the operation is simple,representing the amplitude that needs to be turned down, wherein,representing the turndown factor.
Similarly, in the above formula,the representative predicted change rate is smaller than the industry electric energy trend information, which shows that the predicted change rate in the scheme is smaller, therefore, the information is calculated for the one-dimensional electric energyThe size of the powder is adjusted to be large,representing the amplitude that needs to be adjusted, wherein,representing the upscaling factor.
And S4, the server corrects the first electric energy data according to the multi-dimensional electric energy calculation information to obtain second electric energy data, and the monthly electric energy ratio of the industry is obtained according to the first electric energy data and/or the second electric energy data of all the industries.
According to the scheme, after the more accurate multi-dimensional electric energy calculation information is obtained in the steps S1 to S3, the first electric energy data can be corrected by using the multi-dimensional electric energy calculation information to obtain the second electric energy data, and the missing data can be supplemented by using the second electric energy data. In addition, the scheme can also calculate the monthly electric energy ratio of the industry.
In some embodiments, the server corrects the first electric energy data according to the multidimensional electric energy calculation information to obtain second electric energy data, and obtains an industry monthly electric energy ratio according to the first electric energy data and/or the second electric energy data of all industries, including:
and acquiring first electric energy data and/or second electric energy data of all industries to obtain the sum of the electric energy data. In the scheme, in order to calculate the monthly electric energy ratio of the industry, the electric energy data sum is calculated firstly.
And if all enterprise tags of any one industry are judged to be the first enterprise tags, obtaining the monthly electric energy proportion of the first industry according to the ratio of the first electric energy data and the sum of the electric energy data of the corresponding industry. According to the scheme, the condition that all enterprises upload data is considered, and under the condition, the first industry monthly electric energy ratio can be obtained by utilizing the ratio of the first electric energy data and the sum of the electric energy data of the corresponding industries.
And if the second enterprise label exists in all the enterprise labels of any one industry, obtaining the monthly electric energy ratio of the second industry according to the ratio of the second electric energy data of the corresponding industry to the sum of the electric energy data. According to the scheme, the condition that the enterprise does not upload data is considered, and under the condition, the monthly electric energy ratio of the second industry can be obtained according to the ratio of the second electric energy data of the corresponding industry to the sum of the electric energy data.
The first industry monthly electric energy proportion and the second industry monthly electric energy proportion are calculated by the following formula,
wherein the content of the first and second substances,is as followsA first operating month electric energy fraction of the first electric energy data,is a firstThe first power data is transmitted to the first power receiver,is as followsThe first power data of an individual industry,is the upper limit value of the industry corresponding to the first electric energy data,is as followsSecond power data of the individual industry,is the upper limit value of the industry corresponding to the second electric energy data,is as followsA second industry monthly power fraction of the second power data,is as followsThe second power data is stored in a memory,represents the sum of the electric energy data,
on the basis of the embodiment, the situation that the data which is possibly lost by the enterprise after the first preset time period is found again is considered, at this time, the enterprise can supplement and upload the found data, and under the scene, the prediction model of the scheme can be adjusted by using the supplemented and uploaded real data, so that the prediction result of the prediction model of the scheme is more accurate, therefore, steps S5 to S7 are provided on the basis of the embodiment, and the method specifically comprises the following steps:
and S5, if the enterprise of any industry uploads corresponding first electric energy information after a first preset time period, the first electric energy information is used as training electric energy information.
It can be understood that, when it is determined that the enterprise uploads the first electric energy information later, the first electric energy information is used as training electric energy information.
S6, comparing the training electric energy information with the multi-dimensional electric energy calculation information, and if the difference value between the training electric energy information and the multi-dimensional electric energy calculation information is larger than a first training value, inputting the training electric energy information and the multi-dimensional electric energy calculation information into a training model.
According to the scheme, after the training electric energy information is obtained, the training electric energy information and the multi-dimensional electric energy calculation information are compared, if the difference value between the training electric energy information and the multi-dimensional electric energy calculation information is larger than a first training value, the fact that the error of the multi-dimensional electric energy calculation information predicted by the scheme is larger than a preset value is shown, and at the moment, the training electric energy information and the multi-dimensional electric energy calculation information are input into a training model for training.
And S7, updating and training a reference weight value and/or a trend weight value by the training model based on the difference between the training electric energy information and the multi-dimensional electric energy calculation information.
It can be understood that, in the above embodiment, when calculating the one-dimensional electric energy calculation information, the reference weight value is used for adjustment, and when calculating the industry electric energy trend information, the trend weight value is used for adjustment.
In some embodiments, the training model is used for performing update training on a reference weight value and/or a trend weight value based on a difference value between the training electric energy information and the multi-dimensional electric energy calculation information, and the update training comprises the following steps:
and if the training electric energy information is judged to be larger than the multi-dimensional electric energy calculation information, increasing and adjusting the reference weight value and/or the trend weight value according to the difference value between the training electric energy information and the multi-dimensional electric energy calculation information. When the training electric energy information is determined to be larger than the multi-dimensional electric energy calculation information, the predicted multi-dimensional electric energy calculation information is smaller, the corresponding reference weight value and/or trend weight value needs to be adjusted to be larger, and the subsequently obtained multi-dimensional electric energy calculation information is adjusted to be larger, so that the multi-dimensional electric energy calculation information is more accurate.
And if the training electric energy information is judged to be smaller than the multi-dimensional electric energy calculation information, reducing and adjusting the reference weight value and/or the trend weight value according to the difference value between the training electric energy information and the multi-dimensional electric energy calculation information. When the training electric energy information is determined to be smaller than the multi-dimensional electric energy calculation information, the predicted multi-dimensional electric energy calculation information is larger, the corresponding reference weight value and/or trend weight value needs to be adjusted smaller, and the subsequently obtained multi-dimensional electric energy calculation information is adjusted smaller, so that the multi-dimensional electric energy calculation information is more accurate.
The training model trains and adjusts the reference weight value and/or the trend weight value through the following formula,
wherein the content of the first and second substances,in order to train the information on the electric energy,to train the adjusted baseline weight values and/or the trend weight values,the baseline weight values and/or the trending weight values prior to training adjustment,the coefficients are adjusted for the purpose of forward training,the coefficients are adjusted for reverse training.
In the above-mentioned formula,which is representative of the adjustment parameter(s),representing the coefficients that need to be scaled up,representing coefficients to be scaled down, forward training scaling coefficientsAnd reverse training adjustment factorMay be manually set such that adjusted baseline and/or trend weights are trainedIs more accurate.
On the basis of the above embodiment, the present solution further considers a case that an enterprise may not have historical data to refer to, for example, the enterprise just stands this month and has no historical data to refer to, and in order to predict the enterprise, the present solution refers to dimensional data of enterprise attributes, and provides steps S8 to S11, specifically as follows:
and S8, if the historical second electric energy information corresponding to the second enterprise label in the database is judged to be 0, acquiring second enterprise attribute information corresponding to the second enterprise label. It can be understood that, when the historical second electric energy information is 0, it indicates that the enterprise does not have corresponding historical data, and the scheme may acquire corresponding second enterprise attribute information, where the second enterprise attribute information may be, for example, the number of second enterprises, the area of the second enterprise, and the second production value information.
And S9, acquiring first enterprise attribute information corresponding to each first enterprise tag, and obtaining a similarity coefficient based on the similarity of each first enterprise attribute information and the second enterprise attribute information.
According to the scheme, the enterprises with the most similar attribute dimensions can be found, and the similar enterprises are utilized to carry out data prediction on the enterprises.
In some embodiments, the obtaining of the first enterprise attribute information corresponding to each first enterprise tag and the obtaining of the similarity coefficient based on the similarity between each first enterprise attribute information and the second enterprise attribute information specifically includes:
the method comprises the steps of obtaining a first enterprise number, a first enterprise area and first production value information in first enterprise attribute information, and obtaining a second enterprise number, a second enterprise area and second production value information which respectively correspond to the second enterprise number, the second enterprise area and the second production value information in second enterprise attribute information.
Calculating according to the number of the first enterprises, the area of the first enterprises, the first production value information, the number of the second enterprises, the area of the second enterprises and the second production value information to obtain a similarity coefficient,
wherein the content of the first and second substances,for the second enterprise attribute information anda similarity coefficient of the first business attribute information,is a weight value of the number of people,is as followsA second number of business persons in the second business attribute information,is as followsA first number of business persons in the first business attribute information,is the weight value of the area, and is,is as followsA second enterprise area in the second enterprise attribute information,is as followsA first business area in the first business attribute information,in order to produce a value weight value,is as followsSecond production value information in the second business attribute information,is as followsFirst production value information of the first business attribute information.
In the above-mentioned formula,a similar value representing the dimension of the number of people,the larger the number of people is, the larger the difference is,a similarity value representing the dimension of the area,the larger the difference in the area of the enterprise,a similarity value representing a production value dimension,the larger the difference between production values, the weighted value of peopleArea weight valueProduction value weightMay be manually set.
And S10, determining first electric energy information corresponding to the first enterprise attribute information with the highest similarity coefficient as the closest electric energy information. For example, if the similarity coefficient of enterprise a is the highest, the scheme will take the first power information of enterprise a as the closest power information.
S11, obtaining the difference between the first enterprise attribute information with the highest similarity coefficient and the second enterprise attribute information to obtain difference correction information, and correcting the closest electric energy information based on the difference correction information to obtain multi-dimensional electric energy calculation information.
According to the scheme, the difference between the first enterprise attribute information and the second enterprise attribute information is utilized to obtain difference correction information, and then the nearest electric energy information is corrected to obtain the multi-dimensional electric energy calculation information predicted by the enterprise.
In some embodiments, the obtaining a difference between the first enterprise attribute information and the second enterprise attribute information with the highest similarity coefficient to obtain difference correction information, and correcting the closest electric energy information based on the difference correction information to obtain multidimensional electric energy calculation information includes:
obtaining a number difference value according to the number of people of the first enterprise and the number of people of the second enterprise, obtaining an area difference value according to the area of the first enterprise and the area of the second enterprise, and obtaining a production value difference value according to the first production value information and the second production value information.
Carrying out multi-dimensional fusion calculation on the people number difference value, the area difference value and the output value difference value to obtain difference correction information, calculating the difference correction information through the following formula,
wherein the content of the first and second substances,in order to correct the information for the difference,the number of first enterprises of the first enterprise attribute information with the highest similarity coefficient,is a weight of the difference in the number of people,the first enterprise area of the first enterprise attribute information with the highest similarity coefficient,as a weight of the area difference,first production value information of the first business attribute information with the highest similarity coefficient,to produce difference weights.
In the above-mentioned formula,the number of the representative persons is different from the number of the representative persons,represents the value of the difference in area,representing the difference value of the output value, wherein the larger the difference value is, the larger the difference correction information F needs to be, and the weight of the people number difference isArea difference weightProduction variance weightMay be manually set.
It should be noted that, in the present scheme, the comprehensive calculation is performed by synthesizing the conditions of multiple dimensions, and in practical application, the difference value of the number of people, the difference value of the area, and the difference value of the yield value also have negative conditions. For example, although the number of persons in the enterprise a is smaller than that in the enterprise B, the production value may be larger than that in the enterprise B, and in this case, a negative value may be merged with other values.
In some embodiments, the correcting the closest electric energy information based on the difference correction information to obtain multidimensional electric energy calculation information includes:
and if the difference correction information is larger than 0, forward correction is carried out on the nearest electric energy information according to the difference correction information to obtain multi-dimensional electric energy calculation information. If the difference correction information is larger than 0, the electric energy data of the enterprise is larger than that of the reference enterprise.
And if the difference correction information is less than 0, performing negative correction on the nearest electric energy information according to the difference correction information to obtain multi-dimensional electric energy calculation information. If the difference correction information is less than 0, the electric energy data of the enterprise is smaller than that of the reference enterprise.
The multidimensional electric energy calculation information is obtained by the following formula,
wherein, the first and the second end of the pipe are connected with each other,information is calculated for the multi-dimensional electrical energy,in order to be the closest to the power information,constant values are calculated for the multiple dimensions.
In the above-mentioned formula,representing the electric energy adjustment amplitude, it can be understood that when the difference correction information F is a negative number, the electric energy data of the enterprise is smaller than that of the reference enterprise,also negative number, for nearest power informationCarrying out size-reduction treatment; when the difference correction information F is positive, the electric energy data of the enterprise is larger than that of the reference enterprise,also positive number, for nearest power informationAnd (5) carrying out enlargement processing.
The present invention also provides a storage medium having a computer program stored therein, the computer program being executable by a processor to implement the methods provided by the various embodiments described above.
The storage medium may be a computer storage medium or a communication medium. Communication media includes any medium that facilitates transfer of a computer program from one place to another. Computer storage media may be any available media that can be accessed by a general purpose or special purpose computer. For example, a storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. Of course, the storage medium may also be integral to the processor. The processor and the storage medium may reside in an Application Specific Integrated Circuits (ASIC). Additionally, the ASIC may reside in user equipment. Of course, the processor and the storage medium may reside as discrete components in a communication device. The storage medium may be read-only memory (ROM), random-access memory (RAM), CD-ROMs, magnetic tapes, floppy disks, optical data storage devices, and the like.
The present invention also provides a program product comprising execution instructions stored in a storage medium. The at least one processor of the device may read the execution instructions from the storage medium, and the execution of the execution instructions by the at least one processor causes the device to implement the methods provided by the various embodiments described above.
In the embodiment of the terminal or the server, it should be understood that the Processor may be a Central Processing Unit (CPU), other general-purpose processors, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), and the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (12)
1. The method for correcting the monthly electric energy proportion deviation of the industry based on the multidimensional data is characterized by comprising the following steps:
the server receives first electric energy information sent by all electric energy acquisition ends in a target area at the current moment at intervals of a first preset time period, and calculates the first electric energy information of all the same industries according to industry labels to obtain first electric energy data;
the server takes the enterprise tag which receives the first electric energy information as a first enterprise tag, takes the enterprise tag which does not send the first electric energy information as a second enterprise tag, determines historical second electric energy information corresponding to the second enterprise tag in a database, and calculates according to the second electric energy information to obtain one-dimensional electric energy calculation information corresponding to the second enterprise tag;
the server acquires historical third electric energy information corresponding to a first enterprise tag, calculates based on the third electric energy information and the first electric energy information to obtain industry electric energy trend information, and calculates in a multi-dimensional manner based on the industry electric energy trend information and one-dimensional electric energy calculation information to obtain multi-dimensional electric energy calculation information corresponding to a second enterprise tag;
and the server corrects the first electric energy data according to the multi-dimensional electric energy calculation information to obtain second electric energy data, and obtains the monthly electric energy ratio of the industry according to the first electric energy data and/or the second electric energy data of all industries.
2. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 1,
before the step of receiving, by the server, first electric energy information sent by all electric energy collection terminals in a target area at a current moment at a first preset time interval, the method includes:
classifying all enterprises in a target area according to corresponding industry types to obtain a plurality of enterprise industry sets, setting an electric energy acquisition end at a corresponding enterprise according to the enterprise industry sets, and adding an industry label and an enterprise label to the electric energy acquisition end;
when the electric energy acquisition end acquires first electric energy information of a corresponding enterprise, corresponding industry labels and enterprise labels are added to the first electric energy information.
3. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 2,
the server takes the enterprise label receiving the first electric energy information as a first enterprise label, takes the enterprise label not sending the first electric energy information as a second enterprise label, determines historical second electric energy information corresponding to the second enterprise label in a database, and calculates according to the second electric energy information to obtain one-dimensional electric energy calculation information corresponding to the second enterprise label, and the method comprises the following steps:
the server compares the information quantity of the first electric energy information with the label quantity of the first enterprise label, and if the information quantity does not correspond to the label quantity, the server selects an enterprise industry set of a corresponding industry;
according to the enterprise industry set, taking an enterprise tag corresponding to an electric energy acquisition end which sends first electric energy information as a first enterprise tag, and taking an enterprise tag which does not send the first electric energy information as a second enterprise tag;
extracting historical second electric energy information corresponding to the second enterprise label in the database, calculating a difference value between the second electric energy information at the later moment and the electric energy information at the previous moment in all the second electric energy information to obtain a change difference value, and obtaining a predicted change rate according to the change difference value;
calculating according to the predicted change rate and second electric energy information closest to the current moment to obtain one-dimensional electric energy calculation information;
calculating to obtain one-dimensional electric energy calculation information corresponding to the second enterprise label through the following formula,
wherein, the first and the second end of the pipe are connected with each other,information is calculated for the one-dimensional electrical energy,the second power information closest to the current time,is as followsThe second power information of the individual time instant,is as followsThe second power information of the individual time instant,is the upper limit value of the moment corresponding to the second electric energy information,in order to vary the number of difference values,is a reference weight value.
4. The method for correcting industry monthly electric energy proportion deviation based on multi-dimensional data as claimed in claim 3,
the server acquires historical third electric energy information corresponding to a first enterprise tag, calculates based on the third electric energy information and the first electric energy information to obtain industry electric energy trend information, and calculates multidimensional based on the industry electric energy trend information and one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to a second enterprise tag, wherein the method comprises the following steps:
counting historical third electric energy information corresponding to all the first enterprise tags in a second preset time period to obtain a first historical information set;
sequentially traversing each piece of third electric energy information, deleting the non-real third electric energy information to obtain a second historical information set, and selecting all first electric energy information corresponding to each enterprise tag in the second historical information set to obtain a first current information set;
calculating based on the second historical information set and the first current information set to obtain industry electric energy trend information, and performing multidimensional calculation according to the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to the second enterprise label;
industry electric energy trend information is calculated by the following formula,
wherein the content of the first and second substances,as the information on the trend of the electric energy,is the first in the first current information setFirst power information for a first enterprise tag,is the first in the first historical information setThird power information for the first enterprise tag,is an upper limit value for the first number of enterprise tags in the first current set of information,is the quantity value of the first enterprise tag in the first current set of information,is a trend weight value.
5. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 3,
the multidimensional calculation is performed according to the industry electric energy trend information and the one-dimensional electric energy calculation information to obtain multidimensional electric energy calculation information corresponding to the second enterprise label, and the multidimensional electric energy calculation information comprises:
if the predicted change rate is larger than the industry electric energy trend information, reducing and adjusting the one-dimensional electric energy calculation information according to the predicted change rate and the industry electric energy trend information to obtain multi-dimensional electric energy calculation information;
if the predicted change rate is smaller than the industry electric energy trend information, increasing and adjusting the one-dimensional electric energy calculation information according to the predicted change rate and the industry electric energy trend information to obtain multi-dimensional electric energy calculation information;
the multidimensional electric energy calculation information is obtained by the following formula,
wherein, the first and the second end of the pipe are connected with each other,computing information for the adjusted multidimensional electrical energy,Is a first value of a normalization constant that,in order to reduce the value of the adjustment constant,is the value of a second normalizing constant and,to increase the adjustment constant value.
6. The method for correcting industry monthly electric energy proportion deviation based on multi-dimensional data as claimed in claim 5,
the server corrects the first electric energy data according to the multi-dimensional electric energy calculation information to obtain second electric energy data, and obtains the monthly electric energy ratio of the industries according to the first electric energy data and/or the second electric energy data of all industries, and the method comprises the following steps:
acquiring first electric energy data and/or second electric energy data of all industries to obtain an electric energy data sum;
if all enterprise tags of any one industry are judged to be first enterprise tags, obtaining a first industry monthly electric energy ratio according to the ratio of the first electric energy data of the corresponding industry to the sum of the electric energy data;
if the second enterprise tag exists in all enterprise tags of any one industry, obtaining the monthly electric energy ratio of the second industry according to the ratio of the second electric energy data of the corresponding industry to the sum of the electric energy data;
the first industry monthly electric energy proportion and the second industry monthly electric energy proportion are calculated by the following formula,
wherein the content of the first and second substances,is as followsA first operating month electric energy fraction of the first electric energy data,is a firstThe first power data is transmitted to the first power receiver,is as followsThe first power data of an individual industry,is the upper limit value of the industry corresponding to the first electric energy data,is as followsSecond power data of the individual industry,is the upper limit value of the industry corresponding to the second electric energy data,is as followsA second industry monthly power fraction of the second power data,is as followsAnd second power data.
7. The method for correcting industry monthly electric energy proportion deviation based on multi-dimensional data as claimed in claim 5,
if the enterprise of any one industry uploads corresponding first electric energy information after a first preset time period is judged, taking the first electric energy information as training electric energy information;
comparing the training electric energy information with the multi-dimensional electric energy calculation information, and if the difference value between the training electric energy information and the multi-dimensional electric energy calculation information is larger than a first training value, inputting the training electric energy information and the multi-dimensional electric energy calculation information into a training model;
and the training model updates and trains the reference weight value and/or the trend weight value based on the difference between the training electric energy information and the multi-dimensional electric energy calculation information.
8. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 7,
the training model performs update training on a reference weight value and/or a trend weight value based on a difference between the training electric energy information and the multi-dimensional electric energy calculation information, and includes:
if the training electric energy information is judged to be larger than the multi-dimensional electric energy calculation information, increasing and adjusting the reference weight value and/or the trend weight value according to the difference value between the training electric energy information and the multi-dimensional electric energy calculation information;
if the training electric energy information is judged to be smaller than the multi-dimensional electric energy calculation information, reducing and adjusting the reference weight value and/or the trend weight value according to the difference value between the training electric energy information and the multi-dimensional electric energy calculation information;
the training model trains and adjusts the reference weight value and/or the trend weight value through the following formula,
wherein the content of the first and second substances,in order to train the information on the electric energy,to train the adjusted baseline weight values and/or the trend weight values,to train the baseline weight values and/or the trending weight values prior to adjustment,the coefficients are adjusted for the purpose of forward training,the coefficients are adjusted for reverse training.
9. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 1,
if the historical second electric energy information corresponding to the second enterprise label in the database is judged to be 0, acquiring second enterprise attribute information corresponding to the second enterprise label;
acquiring first enterprise attribute information corresponding to each first enterprise tag, and acquiring a similarity coefficient based on the similarity of each first enterprise attribute information and the second enterprise attribute information;
determining first electric energy information corresponding to first enterprise attribute information with the highest similarity coefficient as the closest electric energy information;
and acquiring the difference between the first enterprise attribute information with the highest similarity coefficient and the second enterprise attribute information to obtain difference correction information, and correcting the closest electric energy information based on the difference correction information to obtain multidimensional electric energy calculation information.
10. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 9,
the obtaining of the first enterprise attribute information corresponding to each first enterprise tag and the obtaining of the similarity coefficient based on the similarity between each first enterprise attribute information and the second enterprise attribute information includes:
acquiring a first enterprise number, a first enterprise area and first production value information in the first enterprise attribute information, and acquiring a second enterprise number, a second enterprise area and second production value information which respectively correspond to the second enterprise attribute information;
calculating according to the number of the first enterprises, the area of the first enterprises, the first production value information, the number of the second enterprises, the area of the second enterprises and the second production value information to obtain a similarity coefficient,
wherein the content of the first and second substances,for the second business attribute information and the firstA similarity coefficient of the first business attribute information,is a weight value of the number of people,is as followsA second number of business persons in the second business attribute information,is a firstA first number of business persons in the first business attribute information,is the weight value of the area, and is,is as followsA second enterprise area in the second enterprise attribute information,is as followsA first business area in the first business attribute information,in order to produce a value weight value,is as followsSecond production value information in the second enterprise attribute information,is as followsFirst production value information in the first business attribute information.
11. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 10,
the obtaining of the difference between the first enterprise attribute information and the second enterprise attribute information with the highest similarity coefficient to obtain difference correction information, and correcting the closest electric energy information based on the difference correction information to obtain multidimensional electric energy calculation information includes:
obtaining a people number difference value according to the number of people of the first enterprise and the number of people of the second enterprise, obtaining an area difference value according to the area of the first enterprise and the area of the second enterprise, and obtaining a production value difference value according to the first production value information and the second production value information;
carrying out multi-dimensional fusion calculation on the people number difference value, the area difference value and the output value difference value to obtain difference correction information, calculating the difference correction information through the following formula,
wherein the content of the first and second substances,in order to correct the information for the difference,the number of first enterprises of the first enterprise attribute information with the highest similarity coefficient,is a weight for the difference in the number of people,the first enterprise area of the first enterprise attribute information with the highest similarity coefficient,as a weight of the area difference,first production value information of the first business attribute information with the highest similarity coefficient,to produce difference weights.
12. The method for correcting industry monthly electric energy ratio deviation based on multi-dimensional data as claimed in claim 11,
the correcting the nearest electric energy information based on the difference correction information to obtain multidimensional electric energy calculation information comprises the following steps:
if the difference correction information is larger than 0, forward correction is carried out on the nearest electric energy information according to the difference correction information to obtain multi-dimensional electric energy calculation information;
if the difference correction information is smaller than 0, carrying out negative correction on the nearest electric energy information according to the difference correction information to obtain multi-dimensional electric energy calculation information;
the multidimensional electric energy calculation information is obtained by the following formula,
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211288547.7A CN115374404B (en) | 2022-10-20 | 2022-10-20 | Method for correcting monthly electric energy proportion deviation of industry based on multi-dimensional data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211288547.7A CN115374404B (en) | 2022-10-20 | 2022-10-20 | Method for correcting monthly electric energy proportion deviation of industry based on multi-dimensional data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115374404A CN115374404A (en) | 2022-11-22 |
CN115374404B true CN115374404B (en) | 2023-01-24 |
Family
ID=84073257
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211288547.7A Active CN115374404B (en) | 2022-10-20 | 2022-10-20 | Method for correcting monthly electric energy proportion deviation of industry based on multi-dimensional data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115374404B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116434425B (en) * | 2023-06-09 | 2023-08-18 | 国网浙江省电力有限公司营销服务中心 | Electricity charge data acquisition method, acquisition equipment and storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022141883A1 (en) * | 2020-12-31 | 2022-07-07 | 平安科技(深圳)有限公司 | Enterprise revenue trend prediction method and apparatus, and computer device and storage medium |
CN114780860A (en) * | 2022-05-30 | 2022-07-22 | 国网浙江省电力有限公司杭州供电公司 | Autonomous decision-making method based on multi-dimensional big data fusion and convergence |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11216718B2 (en) * | 2015-10-27 | 2022-01-04 | Yardi Systems, Inc. | Energy management system |
CN112529301B (en) * | 2020-12-14 | 2023-11-14 | 南京中兴力维软件有限公司 | Power consumption prediction method, device and storage medium |
CN113011774A (en) * | 2021-04-06 | 2021-06-22 | 新奥数能科技有限公司 | Power scheduling method and device, electronic equipment and computer readable medium |
CN114638425B (en) * | 2022-03-24 | 2023-05-23 | 国网江苏省电力有限公司南通供电分公司 | Historical data-based large-user monthly electricity consumption prediction method and system |
CN115062872B (en) * | 2022-08-11 | 2022-11-08 | 国网(宁波)综合能源服务有限公司 | Industry energy consumption prediction method and prediction system based on electric power big data |
-
2022
- 2022-10-20 CN CN202211288547.7A patent/CN115374404B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022141883A1 (en) * | 2020-12-31 | 2022-07-07 | 平安科技(深圳)有限公司 | Enterprise revenue trend prediction method and apparatus, and computer device and storage medium |
CN114780860A (en) * | 2022-05-30 | 2022-07-22 | 国网浙江省电力有限公司杭州供电公司 | Autonomous decision-making method based on multi-dimensional big data fusion and convergence |
Non-Patent Citations (1)
Title |
---|
Analysis of energy consumption of Hebei based on energy consumption coefficient of elasticity;Yonggui He;《IEEE》;20110922;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN115374404A (en) | 2022-11-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112069398A (en) | Information pushing method and device based on graph network | |
US10013636B2 (en) | Image object category recognition method and device | |
US20210342643A1 (en) | Method, apparatus, and electronic device for training place recognition model | |
CN112200211B (en) | Small sample fish identification method and system based on residual network and transfer learning | |
CN108228684B (en) | Method and device for training clustering model, electronic equipment and computer storage medium | |
CN115374404B (en) | Method for correcting monthly electric energy proportion deviation of industry based on multi-dimensional data | |
CN110765882B (en) | Video tag determination method, device, server and storage medium | |
CN112785441B (en) | Data processing method, device, terminal equipment and storage medium | |
CN111008726B (en) | Class picture conversion method in power load prediction | |
CN114330583B (en) | Abnormal electricity utilization identification method and abnormal electricity utilization identification system | |
CN107590460A (en) | Face classification method, apparatus and intelligent terminal | |
CN110634060A (en) | User credit risk assessment method, system, device and storage medium | |
CN107645493B (en) | IP group similarity calculation method | |
CN114119989A (en) | Training method and device for image feature extraction model and electronic equipment | |
CN110942090A (en) | Model training method, image processing method, device, electronic equipment and storage medium | |
EP4285281A1 (en) | Annotation-efficient image anomaly detection | |
CN114925286B (en) | Public opinion data processing method and device | |
WO2022237065A1 (en) | Classification model training method, video classification method, and related device | |
CN115730152A (en) | Big data processing method and big data processing system based on user portrait analysis | |
Lipor et al. | Margin-based active subspace clustering | |
CN115062686A (en) | Multi-KPI (Key performance indicator) time sequence abnormity detection method and system based on multi-angle features | |
CN114839586A (en) | Low-voltage transformer area metering device misalignment calculation method based on EM algorithm | |
CN112699908B (en) | Method for labeling picture, electronic terminal, computer readable storage medium and equipment | |
CN114202694A (en) | Small sample remote sensing scene image classification method based on manifold mixed interpolation and contrast learning | |
CN113360512A (en) | Model updating method and device based on user feedback and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |