CN116167581A

CN116167581A - Battery demand estimation method and device, scheduling method and computer equipment

Info

Publication number: CN116167581A
Application number: CN202310125504.5A
Authority: CN
Inventors: 请求不公布姓名
Original assignee: Fujian Ningde Zhixiang Infinite Technology Co ltd
Current assignee: Fujian Ningde Zhixiang Infinite Technology Co ltd
Priority date: 2023-02-16
Filing date: 2023-02-16
Publication date: 2023-05-26

Abstract

The invention discloses a battery demand pre-estimating method, a device, a scheduling method and computer equipment, which comprise the following steps: acquiring dot information of a target dot; extracting first power conversion information and first charging depth information of the target mesh point according to mesh point information of the target mesh point; according to the first power conversion information and the first charging depth information, calculating by adopting a trained prediction model to obtain a predicted value; and taking the predicted value as the battery demand of the target network point. According to the method and the device, data processing is carried out depending on big data, portrait description of the target network points is formed, the number of battery assets required by the target network points is accurately quantized, efficient configuration of the power conversion assets is achieved, and the asset utilization rate is improved.

Description

Battery demand estimation method and device, scheduling method and computer equipment

Technical Field

The invention relates to the technical field of battery changing cabinets, in particular to a battery demand prediction method, a device, a scheduling method and computer equipment.

Background

With the increase of environmental awareness, electric vehicles are becoming popular with more and more consumers. The electric vehicle power conversion service is used for providing leasing and power conversion service for electric vehicle users, and the power conversion cabinet is used for enabling the users to easily continue to travel only by changing the batteries to the power conversion network point before the electric quantity of the batteries is exhausted.

With the development of the power conversion service scale, large-scale battery and electric cabinet assets are required to be put in, and the existing service operation usually carries out the dispatching in or out of the electric cabinets and batteries of all the network points in a manual assignment dispatching mode, but the manual assignment dispatching mode is low in efficiency and poor in accuracy, unreasonable in battery quantity proportion in all the network points is easy to cause, and the problems of mismatching of the available battery quantity and the electric cabinet quantity with the actual demand are caused.

Disclosure of Invention

Based on the above, it is necessary to provide a method, a device, a scheduling method and a computer device for estimating the battery demand, which are used for solving the problems that in the prior art, the manual dispatch mode is low in efficiency and poor in accuracy, and unreasonable in proportion of the number of batteries in each network point is easy to cause, so that the number of available batteries, the number of electric cabinets and the actual demand are not matched.

In a first aspect, the present application provides a method for estimating a battery demand. The method comprises the following steps:

acquiring dot information of a target dot;

extracting first power conversion information and first charging depth information of the target mesh point according to mesh point information of the target mesh point;

according to the first power conversion information and the first charging depth information, calculating by adopting a trained prediction model to obtain a predicted value;

And taking the predicted value as the battery demand of the target network point.

In one embodiment, the trained prediction model includes:

acquiring dot information of a plurality of dots in a preset area;

extracting training data according to the grid point information of the multiple grid points, wherein the training data comprises second electricity conversion information and second charging depth information of each grid point;

constructing an initial prediction model;

and based on the training data, iteratively training the initial prediction model to obtain a trained prediction model.

In one of the embodiments of the present invention,

iteratively training the initial predictive model based on the training data, comprising:

dividing the training data into a training set and an evaluation set according to a preset proportion;

inputting the training set into the initial prediction model for calculation to obtain an output predicted value;

and comparing the output predicted value with the battery demand corresponding to the evaluation set, and if the difference value of the output predicted value and the battery demand is larger than a preset threshold value, correcting the initial prediction model until the difference value of the output predicted value and the battery demand is smaller than the preset threshold value.

In one embodiment, the initial prediction model includes a battery demand quantum model and a battery growth quantum model, where the modifying the initial prediction model includes: and correcting the sum of the battery demand quantum model and the battery growth quantum model.

In one embodiment, constructing the battery demand quantum model includes:

calculating to obtain the power conversion success rate in the first preset time and the power conversion times in the peak time according to the second power conversion information;

according to the second charging depth information, calculating to obtain the average charging depth in a first preset time;

constructing the battery demand quantum model according to the power conversion success rate, the power conversion times in the peak time and the average charging depth;

wherein the peak time is less than or equal to the first preset time.

In one embodiment, the calculating the number of power changes in the peak time includes:

setting a plurality of time intervals according to preset intervals in the first preset time;

calculating the number of power conversion times in each time interval;

and taking the power change frequency with the largest value as the power change frequency in the peak time.

In one embodiment, the setting a plurality of time intervals within the first preset time according to a preset interval includes:

and combining the time intervals according to the preset quantity to obtain a plurality of combined time intervals.

In one embodiment, constructing the battery growth quantum model includes:

Acquiring the electricity taking quantity of new users of each website in a second preset time according to the second electricity changing information;

calculating to obtain the average demand increment of each website according to the electricity taking quantity of the new user;

and constructing the battery growth quantum model according to the average demand growth amount.

In a second aspect, the present application also provides a battery demand amount calculation device. The device comprises:

the acquisition module is used for acquiring the dot information of the target dots;

the extraction module is used for extracting first power conversion information and first charging depth information of the target mesh point according to mesh point information of the target mesh point;

and the calculation module is used for calculating by adopting a trained prediction model according to the first power conversion information and the first charging depth information to obtain a predicted value, and taking the predicted value as the battery demand of the target network point.

In a third aspect, the present application further provides a scheduling method. The method comprises the following steps:

calculating to obtain the battery demand of the network points to be scheduled by adopting the method of any one of the first aspect;

acquiring current inventory information of the network points to be scheduled, wherein the inventory information comprises the number of batteries and the number of electric cabinets;

And determining the quantity of the transferred batteries and the quantity of the electric cabinets, and the quantity of the transferred batteries and the quantity of the electric cabinets according to the current inventory information and the battery demand.

In a fourth aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of the methods in the first and third aspects when the processor executes the computer program.

The battery demand estimating method, the device, the scheduling method and the computer equipment have at least the following advantages:

according to the method, based on the mesh point information of the target mesh point, the first power conversion information and the first charging depth information of the target mesh point are extracted, then a trained prediction model is adopted to calculate the first power conversion information and the first charging depth information, a predicted value is obtained, and the predicted value is used as the battery demand of the target mesh point. According to the method and the device, data processing is carried out depending on big data, portrait description of the target network points is formed, the number of battery assets required by the target network points is accurately quantized, efficient configuration of the power conversion assets is achieved, and the asset utilization rate is improved.

The conception, specific structure, and technical effects of the present invention will be further described with reference to the accompanying drawings to fully understand the objects, features, and effects of the present invention.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a diagram of an application environment for a battery demand estimation method and a scheduling method in one embodiment;

FIG. 2 is a flowchart of a method for estimating a battery demand according to an embodiment;

FIG. 3 is a schematic diagram of a path for acquiring dot information in one embodiment;

FIG. 4 is a flow diagram of training a predictive model in one embodiment;

FIG. 5 is a schematic view of a scene of a predictive model in one embodiment;

FIG. 6 is a schematic diagram of a business scenario in one embodiment;

FIG. 7 is a variance distribution diagram of scene 1 in one embodiment;

FIG. 8 is a statistical schematic of the order growth of scenario 2 in one embodiment;

FIG. 9 is a schematic diagram of dot distribution of scene 3 in one embodiment;

FIG. 10 is a schematic diagram of dot heat distribution of scene 4 in one embodiment;

FIG. 11 is a block diagram showing a battery demand estimating apparatus according to an embodiment;

FIG. 12 is a flow diagram of a scheduling method in one embodiment;

FIG. 13 is a block diagram of a scheduling system in one embodiment;

FIG. 14 is a schematic diagram of the processing of the task analysis module in one embodiment;

FIG. 15 is a block diagram of the task management module in one embodiment;

fig. 16 is an internal structural view of a computer device in one embodiment.

Detailed Description

Other advantages and effects of the present invention will become apparent to those skilled in the art from the following disclosure, which describes the embodiments of the present invention with reference to specific examples. The invention may be practiced or carried out in other embodiments that depart from the specific details, and the details of the present description may be modified or varied from the spirit and scope of the present invention. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict.

It should be noted that the illustrations provided in the following embodiments merely illustrate the basic concept of the present invention by way of illustration, and only the components related to the present invention are shown in the illustrations, not according to the number, shape and size of the components in actual implementation, and the form, number and proportion of each component in actual implementation may be arbitrarily changed, and the layout of the components may be more complex.

Some exemplary embodiments of the invention have been described for illustrative purposes, it being understood that the invention may be practiced otherwise than as specifically shown in the accompanying drawings.

Referring to fig. 1, the battery demand estimating method and the scheduling method provided in the embodiments of the present application may be applied to the application environment shown in fig. 1. Wherein the terminal communicates with the server through a network. The data storage system may store data that the server needs to process. The data storage system may be integrated on a server or may be placed on a cloud or other network server.

The terminal can send the website information of the target website to the server so that the server processes the website information of the target website, for example, the server acquires the website information from a historical order of the target website and extracts first power conversion information and first charging depth information of the target website from the website information; and calculating by adopting a trained prediction model according to the first power conversion information and the first charging depth information to obtain a predicted value. Meanwhile, the server also determines the quantity of the transferred batteries and the quantity of the electric cabinets, and the quantity of the transferred batteries and the quantity of the electric cabinets according to the predicted battery demand and the current inventory information of the target network point, and feeds the information back to the terminal.

The terminal can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, internet of things equipment and portable wearable equipment, and the internet of things equipment can be smart speakers, smart televisions, smart air conditioners, smart vehicle-mounted equipment and the like. The portable wearable device may be a smart watch, smart bracelet, headset, or the like. The server may be implemented as a stand-alone server or as a server cluster composed of a plurality of servers.

In a possible embodiment, a method for estimating a battery demand is provided in the embodiment of the present application, and the method is described below by taking the application of the method to the server in fig. 1 as an example.

Referring to fig. 2, fig. 2 is a flowchart of a battery demand estimating method according to the present embodiment, which specifically includes:

step S202, acquiring the dot information of the target dot.

Referring to fig. 3, specifically, the embodiment of the present application relies on the IOT platform and the platform cloud gateway to obtain the website information of the target website, where the website information includes the number of electrical cabinets, battery data, application-end embedded point data, population heat data obtained based on the map platform, and POI data (Point of Information, interest point data). And carrying out aggregation processing on the data by depending on a big data platform to form the portrait description of the target network point. The portrait information can accurately describe the information such as the historical operation condition of the target network point, the order period distribution information, the surrounding adjacent network point condition, the surrounding area heat data, the number of new users, the number of network point in-cabinet batteries and the like. After the processing of the website portrait data is finished, the website portrait data is stored in a database of a big data platform, and daily level/hour level updating is achieved.

Step S204, extracting power conversion information and charging depth information of the target mesh point according to mesh point information of the target mesh point.

Specifically, the first electricity exchange behavior of the target network point represents the electricity exchange behavior of the user in the target network point within a period of time, and the electricity exchange times of the user in unit time, the successful electricity exchange times in unit time and the distribution of the service peak time of the target network point can be obtained by analyzing the electricity exchange behavior.

The first charging depth information of the target network point represents the charging depth (depth of charge) of the battery in the target network point, and it is understood that the charging depth is the ratio of the electric quantity received by the battery from the external circuit in the charging process to the electric quantity in the full charging state of the battery, and the charging period and the charging electric quantity of the battery can be obtained through the charging depth, so that the available battery quantity can be obtained.

And S206, calculating by adopting a trained prediction model according to the first power conversion information and the first charging depth information to obtain a predicted value, and taking the predicted value as the battery demand of the target network point.

Specifically, the prediction model predicts the battery demand of the target network node at the time of t+1 according to the first power conversion information and the first charging depth information of the target network node at the time of T. The prediction model is trained based on dot information of a plurality of dots.

Referring to fig. 4, the step of training the prediction model includes:

step S402, acquiring dot information of a plurality of dots in a preset area.

Specifically, in order to obtain more accurate image description, in the embodiment of the present application, according to information such as operation conditions of each website, traffic, and the like, an area to be analyzed is divided in advance to obtain a plurality of areas, and then a preset area is selected from the plurality of areas to analyze.

Step S404, training data is extracted according to the dot information of the plurality of dots.

Step S406, an initial prediction model is constructed.

Step S408, based on the training data, iteratively training the initial prediction model to obtain a trained prediction model.

Specifically, the training data comprises second power conversion information and second charging depth information of each website, the training data is divided into a training set and an evaluation set according to a preset proportion, and the training set is input into an initial prediction model for calculation to obtain a predicted value output by the initial prediction model; and comparing the predicted value output by the initial predicted model with the battery demand corresponding to the evaluation set, and if the difference value of the predicted value and the battery demand is larger than a preset threshold value, correcting the initial predicted model until the difference value of the predicted value output by the initial predicted model and the battery demand corresponding to the evaluation set is smaller than the preset threshold value. It should be understood that the data type and dimension of the training data adopted in training the prediction model are the same as the data type and dimension of the first power conversion information and the first charging depth information of the target network point, the only difference is that the number of the training data is large, and a large amount of training data is adopted to carry out iterative training on the prediction model, so that a prediction result with higher precision can be obtained.

Referring to fig. 5, the training objective of the prediction model is to roughly estimate the upper limit of the number of batteries required by a website, so as to improve the prediction accuracy, in this embodiment, when the prediction model is trained, besides taking the conventional electric field scene at the time of the peak of service, the additional electric field scene when a new user appears is considered, and based on this, the prediction model in this embodiment includes a battery demand quantum model and a battery growth quantum model. The battery demand quantum model is the battery demand quantity of peak time obtained according to the second power conversion information and the second charging depth information, and the battery growth quantum model is the average power taking quantity of new users in a past period of time. In one possible embodiment, the initial predictive model is the sum of a battery demand quantum model and a battery growth quantum model. The modifying of the initial predictive model includes modifying a sum of a battery demand quantum model and a battery growth quantum model. Specifically, the correction of the sum of the battery demand quantum model and the battery growth quantum model includes: and comparing the predicted value output by the initial predicted model with the battery demand corresponding to the evaluation set, and correcting the sum of the battery demand quantum model and the battery growth quantum model until the difference value is smaller than a preset threshold value if the difference value is larger than the preset threshold value. It should be understood that the training data is obtained based on the actual operation condition of each website, so that the evaluation set has the actual battery demand of each website and the actual average power-taking quantity of the new user, and the prediction value output by the initial prediction model is compared with the actual value, so that the prediction precision of the initial prediction model can be obtained.

In one possible embodiment, constructing the battery demand quantum model includes:

according to the second power conversion information, calculating to obtain the power conversion success rate in the first preset time and the power conversion times in the peak time; according to the second charging depth information, calculating to obtain the average charging depth in the first preset time; according to the power conversion success rate, the power conversion times in the peak time and the average charging depth, constructing a battery demand quantum model; wherein the peak time is less than or equal to a first preset time.

Specifically, the expression of the battery demand quantum model is:

wherein n is _chg (t) is the number of times of power conversion in unit time, n _success And (t) is the number of successful power conversion in unit time, and the ratio of the number of successful power conversion in unit time to the number of power conversion in unit time is the power conversion success rate. In this embodiment, the statistics of the power change success rate is (0, 24]Data within an hour, therefore, a first preset time t e (0, 24)]。

depth _avg The average charging depth of the battery in the battery changing cabinet.

When the number of power changes in the peak time is calculated, t epsilon (t ₁ ,t ₂ ]Wherein t is ₁ ，t ₂ Is the business peak period.

Optionally, the battery demand quantum model further includes a correction coefficient α of the charging depth, for adjusting the value of the average charging depth in different service scene settings, where the correction coefficient may be set in advance according to an empirical value, and gradually improved during the subsequent model iterative training, and preferably, the correction coefficient in this embodiment is set to 0.6.

Optionally, the battery demand quantum model further includes a power conversion peak time coefficient β, which is used for adjusting the value of the power conversion peak time in different service scene settings, where the time coefficient is preset according to an empirical value, and preferably, the power conversion peak time coefficient in this embodiment is set to 180 in minutes.

In one possible embodiment, the peak time is a single peak time, and the calculating the number of power changes in the peak time includes:

setting a plurality of time intervals according to preset intervals in a first preset time; calculating the electricity changing times in each time interval; and taking the power change frequency with the largest value as the power change frequency in the peak time. The number of power changes in the peak time can be confirmed by the following method:

wherein 0 is<k<j<＝24，t _a ＝[t _a1 ,t _a2 …t _an ]Representing a time series of days;

indicating the number of battery changes in each time series.

For example, the first preset time is set to be one day, the two hours are set to be a time interval, one day is set to be 12 time intervals, and the value of the number of power changes is set to be the highest in the whole day in the time interval from 6 points to 8 points, and is taken as the number of power changes in the peak time.

In a possible embodiment, the peak time is multiple peak times, and the setting of multiple time intervals according to the preset interval in the first preset time includes:

Further, the calculating the number of times of power conversion in the peak time further includes:

calculating the power conversion times in each combined time interval; and taking the power change frequency with the largest value as the power change frequency in the peak time. The number of power changes in the peak time can be confirmed by the following method:

for example, the first preset time is set to be one day, two hours are set to be one time interval, one day is set to be 12 time intervals, and the 12 time intervals are combined in pairs to obtain a plurality of combined time intervals. Assuming that the sum of the power change times is highest in the whole day in the time interval from 6 to 8 and from 20 to 22, the value is taken as the power change time in the peak time.

In one possible embodiment, constructing the battery growth quantum model includes:

acquiring the electricity taking quantity of new users of each website in a second preset time according to the second electricity changing information; calculating to obtain the average demand increase of each website according to the electricity taking quantity of the new user; and constructing a battery growth quantum model according to the average demand growth amount.

Specifically, the expression of the battery growth quantum model is:

Wherein n is _d And (5) taking electricity quantity for new users of each website in d days.

Through the scheme, the expression of the obtained prediction model is as follows:

N＝N _raw +N _est

wherein N is _raw A quantum model is required for the battery; n (N) _ext A quantum model is grown for the battery.

Optionally, in order to further improve the prediction precision of the prediction model, the present application further combines with the actual service scenario to correct the battery growth quantum model N, where the corrected prediction model is denoted as M, and the expression is: m=n×x= (N) _raw +N _ext )*x

Where x is a gain coefficient of the initial prediction model, and x= (1-k) is used to correct the initial prediction model when the initial prediction model is trained.

Referring to fig. 6-10, the gain factors are set based on actual traffic scenarios, and are illustrated below in conjunction with those shown in fig. 6-10.

Referring to fig. 6, four service scenarios are disclosed in this embodiment, and it should be understood that the service scenario in fig. 6 is only an example of the present application, and a new service scenario may be added according to an actual operation situation of a website in practical application.

Referring to fig. 7, fig. 7 is a variance distribution diagram of dots conforming to scene 1. In scenario 1, when k <1, i.e., x% <100, is 0.ltoreq.k, if the variance of the time interval distribution of the usage frequency of the day order is large, the artifact of excessive demand of the electric cabinet will be caused, the demand in the peak period is not necessarily large enough to complement the battery, in this case, N will be redundant, so the battery needs to be corrected.

Referring to fig. 8, fig. 8 is a statistical schematic diagram of the order growth of the mesh point conforming to scene 2. In scenario 2, when k<0, i.e. when x percent is more than or equal to 100, the mesh point is in an increasing stage, and the average increase of the extra required battery number N exists _ext In case of unexpected situations, the situation that the power conversion requirement cannot be met occurs, and the N needs to be expanded according to the growth rate.

Referring to fig. 9, fig. 9 is a schematic diagram illustrating dot distribution conforming to scene 3. The distribution of the network points in the area in the scene 3 is dense, other network points exist in preset distances around each network point, the influence of adjacent network points is considered, and the local network where the network points are located is expected to be capable of distributing a part of electricity conversion pressure. Therefore, to ensure overall asset efficiency, a correction to N is required.

Referring to fig. 10, fig. 10 is a schematic diagram of dot heat distribution conforming to scene 4. Each dot in the scene 4 is in a high-heat area, which means that the dot is in an area with large demand potential, is not suitable for massive calling, and also needs to be expanded for N.

According to the scheme, the number N of the batteries required by the network points is corrected through the gain coefficient, and finally the corrected training model M is obtained, and the model can accurately estimate the number of the batteries required by each network point.

According to the battery demand estimating method, the first power conversion information and the first charging depth information of the target mesh point are extracted based on the mesh point information of the target mesh point, and then the first power conversion information and the first charging depth information are calculated by adopting a trained prediction model to obtain a predicted value, and the predicted value is used as the battery demand of the target mesh point. According to the method and the device, data processing is carried out depending on big data, portrait description of the target network points is formed, the number of battery assets required by the target network points is accurately quantized, efficient configuration of the power conversion assets is achieved, and the asset utilization rate is improved.

It should be understood that, although the steps in the flowcharts related to the embodiments described above are sequentially shown as indicated by arrows, these steps are not necessarily sequentially performed in the order indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the flowcharts described in the above embodiments may include a plurality of steps or a plurality of stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of the steps or stages is not necessarily performed sequentially, but may be performed alternately or alternately with at least some of the other steps or stages.

Based on the same inventive concept, the embodiment of the application also provides a battery demand estimating device for realizing the battery demand estimating method. The implementation of the solution provided by the device is similar to the implementation described in the method above.

Referring to fig. 11, in one possible embodiment, the present application provides a battery demand amount calculation device, including: the device comprises an acquisition module, an extraction module and a calculation module, wherein:

and the acquisition module is used for acquiring the dot information of the target dot.

Specifically, the acquisition module relies on the IOT platform and the platform cloud gateway to acquire website information of target websites, wherein the website information comprises the number of electric cabinets of each website, battery data, application terminal embedded point data, population heat data acquired based on the map platform and POI data (Point of Information, interest point data). And carrying out aggregation processing on the data by depending on a big data platform to form the portrait description of the target network point. The portrait information can accurately describe the information such as the historical operation condition of the target network point, the order period distribution information, the surrounding adjacent network point condition, the surrounding area heat data, the number of new users, the number of network point in-cabinet batteries and the like. After the processing of the website portrait data is finished, the website portrait data is stored in a database of a big data platform, and daily level/hour level updating is achieved.

And the extraction module is used for extracting first power conversion information and first charging depth information of the target mesh point according to the mesh point information of the target mesh point.

Specifically, the first electricity changing behavior of the user in the target network point is represented in a period of time by the first electricity changing information extracted by the extraction module, and the electricity changing times of the user in unit time, the successful electricity changing times in unit time and the distribution of the service peak time of the target network point can be obtained by analyzing the electricity changing behavior.

The first charging depth information extracted by the extraction module characterizes the charging depth (depth of charge) of the battery in the target network point, and it is understood that the charging depth is the ratio of the electric quantity received by the battery from the external circuit in the charging process to the electric quantity in the full charging state of the battery, and the charging period and the charging electric quantity of the battery can be obtained through the charging depth, so that the available battery quantity is obtained.

And the calculation module is used for calculating by adopting the trained prediction model according to the first power conversion information and the first charging depth information to obtain a predicted value, and taking the predicted value as the battery demand of the target network point.

Specifically, the calculation module includes a model training unit for training a prediction model, where the prediction model predicts the battery demand of the target network point at the time of t+1 according to the power conversion information and the charging depth information of the target network point at the time of T. The prediction model is obtained based on the training of the dot information of a plurality of dots, wherein the step of training the prediction model comprises the following steps:

Step one, acquiring dot information of a plurality of dots in a preset area.

And step two, extracting training data according to the dot information of the plurality of dots.

And thirdly, constructing an initial prediction model.

And step four, based on the training data, iteratively training an initial prediction model to obtain a trained prediction model.

Specifically, the training data comprises second power conversion information and second charging depth information of each website, the training data is divided into a training set and an evaluation set according to a preset proportion, and the training set is input into an initial prediction model for calculation to obtain a predicted value output by the initial prediction model; and comparing the predicted value output by the initial predicted model with the battery demand corresponding to the evaluation set, and if the difference value of the predicted value and the battery demand is larger than a preset threshold value, correcting the initial predicted model until the difference value of the predicted value output by the initial predicted model and the battery demand corresponding to the evaluation set is smaller than the preset threshold value.

In order to improve prediction accuracy, in this embodiment, when the prediction model is trained, in addition to taking the conventional electric field scene at the peak of service, the additional electric field scene when a new user appears is considered, and based on this, the prediction model of this embodiment includes a battery demand quantum model and a battery growth quantum model. The battery demand quantum model is the battery demand quantity of peak time obtained according to the second power conversion information and the second charging depth information, and the battery growth quantum model is the average power taking quantity of new users in a past period of time. In one possible embodiment, the initial predictive model is the sum of a battery demand quantum model and a battery growth quantum model. The modifying of the initial predictive model includes modifying a sum of a battery demand quantum model and a battery growth quantum model. Specifically, the correction of the sum of the battery demand quantum model and the battery growth quantum model includes: and comparing the predicted value output by the initial predicted model with the battery demand corresponding to the evaluation set, and correcting the sum of the battery demand quantum model and the battery growth quantum model until the difference value is smaller than a preset threshold value if the difference value is larger than the preset threshold value. It should be understood that the training data is obtained based on the actual operation condition of each website, so that the evaluation set has the actual battery demand of each website, and the prediction accuracy of the initial prediction model can be obtained by comparing the prediction value output by the initial prediction model with the actual value.

In one possible embodiment, the model training unit constructs the battery demand quantum model including:

Specifically, the expression of the battery demand quantum model is:

indicating the number of battery changes in each time series.

in one possible embodiment, the model training unit constructs the battery growth quantum model including:

Specifically, the expression of the battery growth quantum model is:

N＝N _raw +N _ext

Optionally, in order to further improve the prediction precision of the prediction model, the present application further combines with the actual service scenario to correct the battery growth quantum model N, where the corrected prediction model is denoted as M, and the expression is:

M＝N*x＝(N _raw +N _ext ) X is the gain coefficient of the initial prediction model, and x= (1-k) is used for correcting the initial prediction model when the initial prediction model is trained, and finally a corrected training model M is obtained, wherein the model can accurately estimate the number of batteries required by each network point.

According to the battery demand estimating device, the first power conversion information and the first charging depth information of the target mesh point are extracted based on the mesh point information of the target mesh point, the trained prediction model is adopted to calculate the first power conversion information and the first charging depth information, a predicted value is obtained, and the predicted value is used as the battery demand of the target mesh point. According to the method and the device, data processing is carried out depending on big data, portrait description of the target network points is formed, the number of battery assets required by the target network points is accurately quantized, efficient configuration of the power conversion assets is achieved, and the asset utilization rate is improved.

Each module in the battery demand estimating apparatus may be implemented in whole or in part by software, hardware, or a combination thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.

Referring to fig. 12, in a possible embodiment, a scheduling method is provided in an embodiment of the present application, and the following description will take an application of the method to the server in fig. 1 as an example.

Referring to fig. 12, fig. 12 is a flow chart of a scheduling method according to the present embodiment, which specifically includes:

step S1202, calculating the battery demand of the mesh point to be scheduled, where the battery demand of the mesh point to be scheduled is calculated by using the battery demand estimation method provided in the above embodiment.

Step S1204, acquiring current inventory information of the to-be-scheduled network points, wherein the inventory information comprises the number of the battery in the cabinet and the number of the electric cabinets.

Step S1206, determining the number of the transferred batteries and the number of the electric cabinets, and the number of the transferred batteries and the number of the electric cabinets according to the current inventory information and the battery demand.

Specifically, the maximum battery capacity of the network point to be scheduled can be determined according to the number of the electric cabinets, and it is understood that, in order to ensure that the network point to be scheduled can reliably schedule the batteries when special conditions occur, in this embodiment, a part of the electric cabinets are reserved as standby, and only a preset number of electric cabinets are selected for use, where the preset number can be set according to the actual operation condition of the network point to be scheduled, and the better the operation condition is, the higher the numerical value of the preset number can be set.

For convenience of understanding, the following specific description will be given for the above steps, and it should be understood that the following example is only one implementation manner of this embodiment, and in practical application, the scheduling policy may be adjusted appropriately according to the mesh point information of the mesh point.

The battery demand of the network points to be scheduled is recorded as M1, the current number of in-cabinet batteries of the network points to be scheduled is recorded as M2, and the current number of electric cabinets of the network points to be scheduled is recorded as M.

If M2 is more than or equal to M1:

under the condition that M1 is larger than the maximum battery accommodating amount of the M-1 electric cabinets, the corresponding dispatching action is used for dispatching out M2-M1 batteries;

and under the condition that M1 is smaller than the maximum battery accommodating amount of M-1 electric cabinets, the existence of electric cabinet redundancy is indicated, the corresponding dispatching action is used for dispatching out M2-M1 batteries, and one electric cabinet is called out.

If M2 is less than M1:

under the condition that M1 is smaller than the maximum battery accommodating amount of the network point to be scheduled, the corresponding scheduling action is to schedule in M1-M2 batteries;

and under the condition that M1 is larger than the maximum battery accommodating amount of the network point to be scheduled, the corresponding scheduling action is to call in M1-M2 batteries and call in an electric cabinet.

According to the scheduling method, the battery demand of the network points to be scheduled is obtained through calculation, and then the current number of in-cabinet batteries and the current number of electric cabinets are combined for automatic analysis, so that a corresponding scheduling strategy is provided for on-line operation and maintenance personnel, efficient configuration of the electricity exchanging assets is realized, and the asset utilization rate is improved. Meanwhile, a reasonable scheduling mechanism also meets the power conversion requirement of high-demand network points, and the user experience is improved. In addition, operation and maintenance personnel do not need redundant operation, so that the working difficulty is reduced, and the working efficiency is improved.

In a possible embodiment, the task statistics information obtained in the scheduling method, such as the battery demand of each website, the current number of batteries in the cabinet, the number of electric cabinets, the number of adjustable batteries out, the number of adjustable electric cabinets, the number of batteries to be adjusted in, the number of electric cabinets to be adjusted in, and the like, are all stored in a form, and an operation and maintenance person can pull the form in real time to obtain various task statistics information.

Referring to fig. 13, in one possible embodiment, a scheduling system is provided in an embodiment of the present application, including: the system comprises a task analysis module, an algorithm module and a task management module.

And the task analysis module is used for counting the dot information of each dot and generating task feedback.

Referring to fig. 14, the task analysis module invokes the website information of each website, for example, pulls multiple sets of task data, gathers the task data, generates a summary table, and finally displays a task report in a form of a tree diagram, so as to complete the portrait description of the target website, so as to determine whether the target website and the associated website need to schedule a battery or an electric cabinet. For example, if it is detected that the number of leasing services of a website in a preset time or the number of searching the website by a user is obviously increased, task feedback is generated and sent to the algorithm module, so that the task feedback is analyzed to determine whether the number of batteries of the website meets the use requirement in a future period of time.

The algorithm module is used for calculating the battery demand of the network point to be scheduled according to task feedback, and simultaneously acquiring the current inventory information of the network point to be scheduled; and generating a scheduling task according to the current inventory information and the battery demand, wherein the scheduling task comprises the steps of determining the number of the called batteries and the number of the electric cabinets, and the number of the called batteries and the number of the electric cabinets.

And the task management module is used for sending the scheduling task to related staff. After receiving the dispatching task, the staff can dispatch the battery or the electric cabinet according to the content of the dispatching task. Furthermore, the staff can log in the dispatching system through the task management module to check the work record or the next stage work task.

Referring to fig. 15, the task management module of this embodiment is configured to monitor a scheduled task, that is, create a monitored task after a scheduled task is triggered, determine whether a worker performs the scheduled task according to current inventory information of a to-be-scheduled website, and send a reminder to the worker or a previous worker if the task is detected not to be performed.

Further, the task management module of this embodiment is further configured to record information related to performance of each staff, for example, workload of the staff, execution condition of the scheduled task, etc., for subsequent calling or analysis.

According to the scheduling system, the battery demand of the network point to be scheduled is obtained through calculation, and then the current inventory information is combined for automatic analysis, so that a corresponding scheduling strategy is provided for on-line operation and maintenance personnel, efficient configuration of the electricity exchanging asset is realized, and the asset utilization rate is improved. Meanwhile, a reasonable scheduling mechanism also meets the power conversion requirement of high-demand network points, and the user experience is improved. In addition, operation and maintenance personnel do not need redundant operation, so that the working difficulty is reduced, and the working efficiency is improved.

In one possible embodiment, a computer device is provided, which may be a server, and the internal structure of which may be as shown in fig. 16. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used for storing data in a battery demand estimation method and a scheduling method. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program, when executed by the processor, implements a battery demand estimation method and a scheduling method.

It will be appreciated by those skilled in the art that the structure shown in fig. 16 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the computer device to which the present application is applied, and that a particular computer device may include more or fewer components than shown, or may combine certain components, or have a different arrangement of components.

In one possible embodiment, a computer device is provided that includes a memory and a processor, the memory having a computer program stored therein, the processor implementing a battery demand estimation method and a scheduling method when executing the computer program.

In one possible embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements a battery demand estimation method and a scheduling method.

Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magnetic random access Memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric Memory (Ferroelectric Random Access Memory, FRAM), phase change Memory (Phase Change Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in the form of a variety of forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), and the like. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The above embodiments are merely illustrative of the principles of the present invention and its effectiveness, and are not intended to limit the invention. Modifications and variations may be made to the above-described embodiments by those skilled in the art without departing from the spirit and scope of the invention. Accordingly, it is intended that all equivalent modifications and variations of the invention be covered by the claims, which are within the ordinary skill of the art, be within the spirit and scope of the present disclosure.

Claims

1. A battery demand estimation method, characterized in that the method comprises:

acquiring dot information of a target dot;

extracting first power conversion information and second charging depth information of the target mesh point according to mesh point information of the target mesh point;

2. The method of claim 1, wherein said trained predictive model comprises:

acquiring dot information of a plurality of dots in a preset area;

constructing an initial prediction model;

3. The method of claim 2, wherein,

4. The method of claim 3, wherein the initial predictive model includes a battery demand quantum model and a battery growth quantum model, wherein said modifying the initial predictive model includes: and correcting the sum of the battery demand quantum model and the battery growth quantum model.

5. The method of claim 4, wherein constructing the battery demand quantum model comprises:

wherein the peak time is less than or equal to the first preset time.

6. The method of claim 5, wherein said calculating the number of power changes during the peak time comprises:

calculating the number of power conversion times in each time interval;

7. The method of claim 6, wherein the setting a plurality of time intervals at predetermined intervals during the first predetermined time includes:

8. The method of any one of claims 4 to 6, wherein constructing the battery growth quantum model comprises:

9. A battery demand amount calculation apparatus, characterized by comprising:

10. A scheduling method, the method comprising:

calculating to obtain the battery demand of the network points to be scheduled by adopting the method of any one of claims 1 to 8;

11. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any one of claims 1 to 8 and the steps of the method of claim 10 when the computer program is executed.