Market share prediction method and device

Abstract

The invention relates to the technical field of market share prediction, and provides a market share prediction method and device. The method comprises the following steps: acquiring a plurality of characteristics which influence the hour market occupancy after the charging cost in a target charging station is adjusted; calculating a historical contribution value of each feature based on the historical data of the plurality of features; screening out a plurality of target features with relatively large influence degrees from the plurality of features based on the historical contribution values; and inputting the current data of the plurality of target features into the market share prediction model to obtain the market share of the target hour in one day after the current price adjustment. On one hand, the LSTM network can be fully utilized to mine the rule of the historical data in the time sequence and the contribution of the target feature to the prediction result in the time sequence, the characteristic that the change of the market share needs to be accumulated by influence factors for a long time is better met, and the prediction accuracy is improved; manual experience prediction can be replaced by a mechanical means, and the prediction efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of market share prediction, and in particular to a market share prediction method and device. Background Art

[0002] With the rapid development of the electric vehicle industry, consumers' demand for electric vehicle charging is also growing. For charging service providers, the charging fees that consumers need to pay when charging at the charging stations they cover include the basic electricity price and the service fee of the service provider. In order to maximize profits, service providers can adjust the prices of their controllable service fees, thereby adjusting the charging fees. However, the adjustment of charging fees will affect consumers' charging decisions, thereby affecting the market share of service providers' charging services and further affecting the profits of service providers.

[0003] Since the revenue of service providers is closely related to the charging costs of consumers at charging stations and the market share of charging services, and there is also a mutual influence between charging costs and market share, service providers need to predict the market share after the price adjustment before adjusting service fees, so as to comprehensively evaluate the overall impact of changes in charging costs and market share on their own revenue, so as to formulate reasonable pricing strategies and maximize their own profits.

[0004] However, the current prediction of market share after price adjustment mainly relies on the experience of service operators, and its prediction accuracy and efficiency are relatively low. Summary of the invention

[0005] The embodiments of the present application provide a market share prediction method and device to solve the technical problem that the current prediction of market share after price adjustment mainly relies on the experience of service operators, and the prediction accuracy and efficiency are low.

[0006] In a first aspect, an embodiment of the present application provides a market share prediction method, comprising: Obtain multiple features that affect the hourly market share of charging fees in the target charging station after the price adjustment; Based on the historical data of the multiple features, calculate the historical contribution value of each feature in the multiple features; Based on the historical contribution values, a plurality of target features with greater influence are selected from the plurality of features; Inputting the current data of the plurality of target features into a market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

[0007] In one embodiment, the historical contribution value of any feature among the multiple features is determined based on the following steps: Calculate the variance of the market share of the target hour on the historical adjusted day corresponding to the historical data of the multiple features to obtain a first variance; Based on candidate features among the multiple features, dividing the historical data of the multiple features into multiple data sets; the candidate feature is any feature among the multiple features; For each of the plurality of data sets, calculating the variance of the market share of the target hour on the corresponding historical adjusted day to obtain a plurality of second variances; Calculating a quantity ratio of the historical data in each data set to the historical data in the plurality of data sets to obtain a plurality of quantity ratios; A historical contribution value of the candidate feature among the multiple features is obtained based on the first variance, the multiple second variances, and the multiple quantity ratios.

[0008] In one embodiment, the selecting a plurality of target features with greater influence from the plurality of features based on the historical contribution values ​​includes: The historical contribution values ​​are sorted from large to small, and features corresponding to a plurality of historical contribution values ​​that are ranked high are determined as target features.

[0009] In one embodiment, the market share prediction model is obtained based on the following steps: Inputting historical data of the plurality of target features and the market share of the target hour one day after the historical price adjustment into the LSTM network, and obtaining a historical forecast value of the market share of the target hour one day after the historical price adjustment output by the LSTM network; Calculate the symmetric mean absolute percentage error between the historical predicted value and its corresponding historical actual value to obtain the prediction error of the LSTM network; Based on the prediction error, calculating the prediction accuracy of the LSTM network; If the prediction accuracy is less than the accuracy threshold, after updating the weight of the LSTM network using the stochastic gradient descent algorithm, return to the step of inputting the historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment into the LSTM network to obtain the historical prediction value of the market share of the target hour one day after the historical price adjustment output by the LSTM network, until the prediction accuracy is greater than or equal to the accuracy threshold, and determine the LSTM network at this time as the market share prediction model.

[0010] In one embodiment, after calculating the prediction accuracy of the LSTM network based on the prediction error, the method further includes: If the prediction accuracy is less than the accuracy threshold, after updating the weight of the LSTM network using the stochastic gradient descent algorithm, return to the step of dividing the historical data of the multiple features into multiple data sets based on the candidate features among the multiple features until the prediction accuracy is greater than or equal to the accuracy threshold, and determine the LSTM network at this time as the market share prediction model.

[0011] In one embodiment, the number of layers of the LSTM network is 3, and the number of hidden units in each layer is 256.

[0012] In a second aspect, an embodiment of the present application provides a market share prediction device, comprising: The hourly market share impact feature acquisition module is used to: acquire multiple features that affect the hourly market share after the charging fee price adjustment in the target charging station; A feature historical contribution value calculation module, used to: calculate the historical contribution value of each feature among the multiple features based on the historical data of the multiple features; An hourly market share target feature screening module is used to: screen out a plurality of target features with greater influence from the plurality of features based on the historical contribution values; An hourly market share prediction module is used to: input the current data of the multiple target features into a market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

[0013] In a third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory storing a computer program, wherein when the processor executes the program, the steps of the market share prediction method described in the first aspect are implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the market share prediction method described in the first aspect.

[0015] In a fifth aspect, an embodiment of the present application provides a non-transitory computer-readable storage medium, including a computer program, which, when executed by a processor, implements the steps of the market share prediction method described in the first aspect.

[0016] The market share prediction method and device provided in the present application obtain multiple features that affect the hourly market share after the charging fee price adjustment in the target charging station, calculate the historical contribution value of each feature among the multiple features based on the historical data of the multiple features, screen out multiple target features with greater influence from the multiple features based on the historical contribution value, input the current data of the multiple target features into the market share prediction model, and obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model. This application first calculates the historical contribution value of each feature, and based on the historical contribution value, screens the target features that have a greater impact on the hourly market share after the price adjustment, and then uses the market share prediction model based on these target features to predict the hourly market share after the price adjustment. Since the market share prediction model is obtained by training the long short-term memory LSTM network using the historical data of multiple target features and the market share of the target hour one day after the historical price adjustment, this application uses the historical contribution value to perform a preliminary screening of the target features, and then predicts the hourly market share based on the market share prediction model with a high degree of adaptability to the target features. On the one hand, it can make full use of the LSTM network to mine the laws of historical data in time series and the contribution of target features to the prediction results in time series, which is more in line with the characteristics that changes in market share usually require a long period of accumulation of influencing factors, thereby improving the prediction accuracy. On the other hand, it can replace manual experience prediction with mechanized means to improve prediction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present application or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is one of the flow charts of the market share prediction method provided in the embodiment of the present application; Figure 2 This is the second flow chart of the market share prediction method provided in the embodiment of the present application; Figure 3 This is the third flow chart of the market share prediction method provided in the embodiment of the present application; Figure 4 is a schematic diagram of the structure of a market share prediction device provided in an embodiment of the present application; Figure 5 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] It should be noted that in the description of the embodiments of the present application, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or device including the elements. The orientation or position relationship indicated by the terms "upper", "lower" and the like is based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present application. Unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or it can be a connection between the two elements. For those skilled in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0021] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0022] Figure 1 This is one of the flow charts of the market share prediction method provided in the embodiment of the present application. Figure 1 , the embodiment of the present application provides a market share prediction method, which may include: 101. Obtain multiple features that affect the hourly market share of charging fees in the target charging station after the price adjustment; 102. Based on the historical data of multiple features, calculate the historical contribution value of each feature among the multiple features; 103. Select multiple target features with greater influence from multiple features based on historical contribution values; 104. Input the current data of multiple target features into the market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model.

[0023] The market share prediction model is trained on the basis of the long short-term memory (LSTM) network using historical data of multiple target features and the market share of the target hour one day after the historical price adjustment.

[0024] In step 101, the charging fee includes a basic electricity price and a service fee, wherein the basic electricity price is set by the power supply unit, and each day is divided into peak hours, peak hours, parity hours and valley hours of electricity consumption according to hours, and different basic electricity prices are set for each time period. On the one hand, different regions have different divisions of different time periods, and the divisions of different time periods will also change with factors such as seasonal changes. On the other hand, there are differences in the basic electricity prices for different time periods in the same region and for the same time periods in different regions. Therefore, the data of dynamic features in multiple features in this application, such as charging volume, charging costs and other data, are adapted to use aggregated data within hours. Since the dynamic feature data on the charging order is usually across hours, these data can be split by minutes and aggregated into different hours to obtain aggregated data within hours. The use of these hourly aggregated data can maximize the realization of detailed quantitative predictions while ensuring the comparability of data during changes.

[0025] Specifically, assuming that you want to adjust the charging fee from 20:00 to 21:00 at the target charging station, the multiple features that affect the market share from 20:00 to 21:00 after the price adjustment may include the market share from 20:00 to 21:00 every day within 14 days before the price adjustment, the average charging fee from 20:00 to 21:00 14 days before the price adjustment, the average basic electricity price from 20:00 to 21:00 14 days before the price adjustment, the average service fee from 20:00 to 21:00 14 days before the price adjustment, whether it is a holiday before and after the price adjustment, the electricity consumption period from 20:00 to 21:00 before and after the price adjustment, the time characteristics from 20:00 to 21:00 before and after the price adjustment (for example, which month it belongs to, which week of a month it belongs to, which day of the week it belongs to, which hour of a day it belongs to, etc.), the latitude and longitude coordinates of the target charging station, the number of charging piles in the target charging station, the number of charging guns in the target charging station, and the charging fee, charging volume and market share of other charging stations from 20:00 to 21:00 within a specific range around the target charging station before the price adjustment, etc.

[0026] In step 103, based on the historical contribution values, multiple target features that have a greater impact on the market share from 20 points to 21 points after the price adjustment can be screened out from the multiple features in step 101.

[0027] In step 104, the current data of these target features are input into the market share prediction model, and the market share prediction value from 20:00 to 21:00 one day after the current price adjustment can be obtained.

[0028] By analogy, we can get the market share forecast values ​​for other target hours one day after the current price adjustment. The market share prediction method provided in this embodiment obtains multiple features that affect the hourly market share after the charging fee price adjustment in the target charging station, calculates the historical contribution value of each feature among the multiple features based on the historical data of the multiple features, screens out multiple target features with greater influence from the multiple features based on the historical contribution values, inputs the current data of the multiple target features into the market share prediction model, and obtains the market share of the target hour one day after the current price adjustment output by the market share prediction model. This embodiment first calculates the historical contribution value of each feature, and based on the historical contribution value, screens the target features that have a greater impact on the hourly market share after the price adjustment, and then uses the market share prediction model to predict the hourly market share after the price adjustment based on these target features. Since the market share prediction model is obtained by training the long short-term memory LSTM network using the historical data of multiple target features and the market share of the target hour one day after the historical price adjustment, this embodiment uses the historical contribution value to perform a preliminary screening of the target features, and then predicts the hourly market share based on the market share prediction model with a high degree of adaptability to the target features. On the one hand, it can make full use of the LSTM network to mine the laws of historical data in the time series and the contribution of the target features to the prediction results in the time series, which is more in line with the characteristics that the change of market share usually requires a long period of accumulation of influencing factors, thereby improving the prediction accuracy. On the other hand, it can replace manual experience prediction with mechanized means to improve prediction efficiency.

[0029] Figure 2 This is the second flow chart of the market share prediction method provided in the embodiment of the present application. Figure 2 In one embodiment, the historical contribution value of any feature among multiple features can be determined based on the following steps: 201. Calculate the variance of the market share of the target hour of the day after historical adjustment corresponding to the historical data of the multiple features to obtain a first variance; 202. Based on candidate features among the multiple features, divide the historical data of the multiple features into multiple data sets; A candidate feature is any one of multiple features; 203. For each of the multiple data sets, calculate the variance of the market share of the target hour of the corresponding historical adjusted day to obtain multiple second variances; 204. Calculate the quantity ratio of the historical data in each data set to the historical data in multiple data sets to obtain multiple quantity ratios; 205. Obtain a historical contribution value of the candidate feature among the multiple features based on the first variance, the multiple second variances, and the multiple quantity ratios.

[0030] In step 201, it is assumed that multiple features include A, B, and C, each feature includes multiple historical data, and there is a corresponding relationship between the historical data of these features. For example, the historical data of A includes a1, a2, and a3, the historical data of B includes b1, b2, and b3, and the historical data of C includes c1, c2, and c3, wherein a1, b1, and c1 correspond to form sample 1, a2, b2, and c2 correspond to form sample 2, and a3, b3, and c3 correspond to form sample 3, wherein the market share of the historical adjusted target hour of one day corresponding to sample 1 is d1, the market share of the historical adjusted target hour of one day corresponding to sample 2 is d2, and the market share of the historical adjusted target hour of one day corresponding to sample 3 is d3, then the variances of these d1, d2, and d3 are calculated to obtain the first variance.

[0031] In step 202, A is used as a candidate feature. Assuming that A's historical data a2 is used as a dividing point, sample 1 and sample 2 are divided into data set 1, and sample 3 is divided into data set 2, wherein a1 and a2 in data set 1 are less than or equal to a2, and a3 in data set 2 is greater than a2.

[0032] In step 203, the variances of d1 and d2 corresponding to the samples in data set 1 and the variance of d3 corresponding to the samples in data set 2 are calculated to obtain two second variances.

[0033] In step 204, the number of historical data in data set 1 is 6, the number of historical data in data set 2 is 3, and the number of historical data in both data sets is 9. The ratio of the number of historical data in data set 1 to the number of historical data in both data sets is 2 / 3, and the ratio of the number of historical data in data set 2 to the number of historical data in both data sets is 1 / 3.

[0034] In step 205, the historical contribution value of A in multiple features can be calculated based on the following formula: : ; in, is the first variance, is the number of data sets, that is, the number of second variances and the number of ratios, It is The number ratio corresponding to the data set is It is The second variance corresponding to the data set.

[0035] Then, B and C are taken as candidate features respectively, and the historical data of multiple features are divided based on the above steps to obtain the historical contribution values ​​of B and C in multiple features.

[0036] Furthermore, the historical contribution values ​​are sorted from large to small, and features corresponding to multiple historical contribution values ​​with the highest sorting are determined as target features.

[0037] This embodiment divides historical data of multiple features based on specific features to form multiple data sets, and then calculates the second variance and quantity ratio based on the data sets, multiplies the second variance corresponding to each data set by the data ratio and then adds them up, and finally calculates the difference between the summed value and the first variance when the data set is not divided, so as to measure the fluctuation of the data variance before and after the data set is divided, and uses the fluctuation to characterize the historical contribution of the specific features of the divided data set, so that the target features that have a greater impact on the hourly market share after the price adjustment can be accurately screened out based on the ranking of historical contributions.

[0038] Figure 3 This is the third flow chart of the market share prediction method provided in the embodiment of the present application. Figure 3 In one embodiment, the market share prediction model can be obtained based on the following steps: 301. Input historical data of multiple target features and the market share of the target hour one day after the historical price adjustment into the LSTM network, and obtain the historical prediction value of the market share of the target hour one day after the historical price adjustment output by the LSTM network; 302. Calculate the symmetric mean absolute percentage error between the historical predicted value and its corresponding historical actual value to obtain the prediction error of the LSTM network; 303. Based on the prediction error, calculate the prediction accuracy of the LSTM network; 304. If the prediction accuracy is less than the accuracy threshold, the weight of the LSTM network is updated using the stochastic gradient descent algorithm, and then the process returns to step 301; 305. If the prediction accuracy is greater than or equal to the accuracy threshold, the LSTM network at this time is determined as a market share prediction model.

[0039] In step 301, the data of multiple target features in the past five years and their corresponding market share of the target hour one day after the price adjustment can be input into the LSTM network to obtain the historical forecast value of the market share of the target hour one day after the historical price adjustment output by the LSTM network. It should be noted that the data of multiple target features in the past five years need to be split by minutes and aggregated into different hours to form hourly aggregated data before input.

[0040] In step 302, the symmetric mean absolute percentage error between the historical predicted value and its corresponding historical actual value It can be obtained based on the following formula: ; in, is the current cumulative number of training times, yes In the training The historical prediction value of training times, It is the historical actual value corresponding to the historical predicted value.

[0041] In step 303, the prediction accuracy .

[0042] In step 304 to step 305, the LSTM network is trained with the goal of making the prediction accuracy greater than or equal to the accuracy threshold by using the stochastic gradient descent algorithm to update the weights of the LSTM network, and finally a market share prediction model is obtained.

[0043] It should be noted that the number of layers of the LSTM network and the number of hidden units in each layer can be set according to demand and are not limited here. In this embodiment, the number of layers of the LSTM network is 3, and the number of hidden units in each layer is 256.

[0044] This embodiment aims to achieve a model prediction accuracy that meets the requirements, and trains the LSTM network by using a stochastic gradient descent algorithm to update the weights of the LSTM network. Since the training data covers a large amount of data from the past five years, the use of a stochastic gradient descent algorithm to update the weights of the LSTM network can, on the one hand, effectively improve the training efficiency, and on the other hand, when the hourly market share and charging cost fluctuation range are small, it can introduce randomness to jump out of the local optimality of the model and reach the global optimality of the model.

[0045] In one embodiment, after calculating the prediction accuracy of the LSTM network based on the prediction error, the following steps may also be performed: If the prediction accuracy is less than the accuracy threshold, the weight of the LSTM network is updated using the stochastic gradient descent algorithm, and then the step of dividing the historical data of multiple features into multiple data sets based on the candidate features among the multiple features is returned until the prediction accuracy is greater than or equal to the accuracy threshold, and the LSTM network at this time is determined as the market share prediction model.

[0046] That is, if the corrected accuracy is less than the accuracy threshold, the weight of the LSTM network is updated using the stochastic gradient descent algorithm, and then the process returns to step 202 until the corrected accuracy is greater than or equal to the accuracy threshold, and the gradient boosting tree model at this time is determined as the market share prediction model.

[0047] This embodiment not only uses the stochastic gradient descent algorithm to adjust the weights of the LSTM network, but also re-divides the data set to re-screen the target features. The two adjustments work together to complete the training of the model more flexibly and efficiently, and obtain a market share prediction model with higher prediction accuracy.

[0048] The market share prediction device provided in the embodiment of the present application is described below. The market share prediction device described below and the market share prediction method described above can be referenced to each other.

[0049] Figure 4 Schematic diagram of the structure of the market share prediction device provided in the embodiment of the present application. Figure 4 , the embodiment of the present application provides a market share prediction device, which may include: The hourly market share influence feature acquisition module 401 is used to: acquire multiple features that affect the hourly market share after the charging fee price adjustment in the target charging station; The feature historical contribution value calculation module 402 is used to: calculate the historical contribution value of each feature among the multiple features based on the historical data of the multiple features; The hourly market share target feature screening module 403 is used to screen out multiple target features with greater influence from the multiple features based on the historical contribution values; The hourly market share prediction module 404 is used to: input the current data of the multiple target features into the market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

[0050] The market share prediction device provided in this embodiment obtains multiple features that affect the hourly market share after the charging fee price adjustment in the target charging station, calculates the historical contribution value of each feature among the multiple features based on the historical data of the multiple features, screens out multiple target features with greater influence from the multiple features based on the historical contribution values, inputs the current data of the multiple target features into the market share prediction model, and obtains the market share of the target hour one day after the current price adjustment output by the market share prediction model. This embodiment first calculates the historical contribution value of each feature, and based on the historical contribution value, screens the target features that have a greater impact on the hourly market share after the price adjustment, and then uses the market share prediction model to predict the hourly market share after the price adjustment based on these target features. Since the market share prediction model is obtained by training the long short-term memory LSTM network using the historical data of multiple target features and the market share of the target hour one day after the historical price adjustment, this embodiment uses the historical contribution value to perform a preliminary screening of the target features, and then predicts the hourly market share based on the market share prediction model with a high degree of adaptability to the target features. On the one hand, it can make full use of the LSTM network to mine the laws of historical data in the time series and the contribution of the target features to the prediction results in the time series, which is more in line with the characteristics that the change of market share usually requires a long period of accumulation of influencing factors, thereby improving the prediction accuracy. On the other hand, it can replace manual experience prediction with mechanized means to improve prediction efficiency.

[0051] In one embodiment, the feature history contribution value calculation module 402 is specifically used to: Calculate the variance of the market share of the target hour on the historical adjusted day corresponding to the historical data of the multiple features to obtain a first variance; Based on candidate features among the multiple features, dividing the historical data of the multiple features into multiple data sets; the candidate feature is any feature among the multiple features; For each of the plurality of data sets, calculating the variance of the market share of the target hour on the corresponding historical adjusted day to obtain a plurality of second variances; Calculating a quantity ratio of the historical data in each data set to the historical data in the plurality of data sets to obtain a plurality of quantity ratios; A historical contribution value of the candidate feature among the multiple features is obtained based on the first variance, the multiple second variances, and the multiple quantity ratios.

[0052] In one embodiment, the hourly market share target feature screening module 403 is specifically used to: The historical contribution values ​​are sorted from large to small, and features corresponding to a plurality of historical contribution values ​​that are ranked high are determined as target features.

[0053] In one embodiment, a market share prediction model building module (not shown in the figure) is further included, which is used to: Inputting historical data of the plurality of target features and the market share of the target hour one day after the historical price adjustment into the LSTM network, and obtaining a historical forecast value of the market share of the target hour one day after the historical price adjustment output by the LSTM network; Calculate the symmetric mean absolute percentage error between the historical predicted value and its corresponding historical actual value to obtain the prediction error of the LSTM network; Based on the prediction error, calculating the prediction accuracy of the LSTM network; If the prediction accuracy is less than the accuracy threshold, after updating the weight of the LSTM network using the stochastic gradient descent algorithm, return to the step of inputting the historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment into the LSTM network to obtain the historical prediction value of the market share of the target hour one day after the historical price adjustment output by the LSTM network, until the prediction accuracy is greater than or equal to the accuracy threshold, and determine the LSTM network at this time as the market share prediction model.

[0054] In one embodiment, the market share prediction model building module is further used to: If the prediction accuracy is less than the accuracy threshold, after updating the weight of the LSTM network using the stochastic gradient descent algorithm, return to the step of dividing the historical data of the multiple features into multiple data sets based on the candidate features among the multiple features until the prediction accuracy is greater than or equal to the accuracy threshold, and determine the LSTM network at this time as the market share prediction model.

[0055] In one embodiment, the number of layers of the LSTM network is 3, and the number of hidden units in each layer is 256.

[0056] FIG5 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 5 As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call a computer program in the memory 530 to execute the steps of the market share prediction method, for example including: Obtain multiple features that affect the hourly market share of charging fees in the target charging station after the price adjustment; Based on the historical data of the multiple features, calculate the historical contribution value of each feature in the multiple features; Based on the historical contribution values, a plurality of target features with greater influence are selected from the plurality of features; Inputting the current data of the plurality of target features into a market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

[0057] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0058] On the other hand, an embodiment of the present application further provides a computer program product, the computer program product comprising a computer program, the computer program may be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute the steps of the market share prediction method provided in the above embodiments, for example, including: Obtain multiple features that affect the hourly market share of charging fees in the target charging station after the price adjustment; Based on the historical data of the multiple features, calculate the historical contribution value of each feature in the multiple features; Based on the historical contribution values, a plurality of target features with greater influence are selected from the plurality of features; Inputting the current data of the plurality of target features into a market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

[0059] On the other hand, an embodiment of the present application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is used to cause a processor to execute the steps of the market share prediction method provided in the above embodiments, for example, including: Obtain multiple features that affect the hourly market share of charging fees in the target charging station after the price adjustment; Based on the historical data of the multiple features, calculate the historical contribution value of each feature in the multiple features; Based on the historical contribution values, a plurality of target features with greater influence are selected from the plurality of features; Inputting the current data of the plurality of target features into a market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

[0060] The non-transitory computer-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)), etc.

[0061] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0062] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A market share prediction method, characterized in that: include: Obtain multiple features that affect the hourly market share of charging fees in the target charging station after the price adjustment; Based on the historical data of the multiple features, calculate the historical contribution value of each feature in the multiple features; Based on the historical contribution values, a plurality of target features with greater influence are selected from the plurality of features; Inputting the current data of the plurality of target features into a market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

2. The market share prediction method according to claim 1, characterized in that: The historical contribution value of any feature among the multiple features is determined based on the following steps: Calculate the variance of the market share of the target hour on the historical adjusted day corresponding to the historical data of the multiple features to obtain a first variance; Based on candidate features among the multiple features, dividing the historical data of the multiple features into multiple data sets; the candidate feature is any feature among the multiple features; For each of the plurality of data sets, calculating the variance of the market share of the target hour on the corresponding historical adjusted day to obtain a plurality of second variances; Calculating a quantity ratio of the historical data in each data set to the historical data in the plurality of data sets to obtain a plurality of quantity ratios; A historical contribution value of the candidate feature among the multiple features is obtained based on the first variance, the multiple second variances, and the multiple quantity ratios.

3. The market share prediction method according to claim 1, characterized in that: The step of selecting a plurality of target features with greater influence from the plurality of features based on the historical contribution values ​​includes: The historical contribution values ​​are sorted from large to small, and features corresponding to a plurality of historical contribution values ​​that are ranked high are determined as target features.

4. The market share prediction method according to claim 1, characterized in that: The market share prediction model is obtained based on the following steps: Inputting historical data of the plurality of target features and the market share of the target hour one day after the historical price adjustment into the LSTM network, and obtaining a historical forecast value of the market share of the target hour one day after the historical price adjustment output by the LSTM network; Calculate the symmetric mean absolute percentage error between the historical predicted value and its corresponding historical actual value to obtain the prediction error of the LSTM network; Based on the prediction error, calculating the prediction accuracy of the LSTM network; If the prediction accuracy is less than the accuracy threshold, after updating the weight of the LSTM network using the stochastic gradient descent algorithm, return to the step of inputting the historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment into the LSTM network to obtain the historical prediction value of the market share of the target hour one day after the historical price adjustment output by the LSTM network, until the prediction accuracy is greater than or equal to the accuracy threshold, and determine the LSTM network at this time as the market share prediction model.

5. The market share prediction method according to claim 4, characterized in that: After calculating the prediction accuracy of the LSTM network based on the prediction error, the method further includes: If the prediction accuracy is less than the accuracy threshold, after updating the weight of the LSTM network using the stochastic gradient descent algorithm, return to the step of dividing the historical data of the multiple features into multiple data sets based on the candidate features among the multiple features until the prediction accuracy is greater than or equal to the accuracy threshold, and determine the LSTM network at this time as the market share prediction model.

6. The market share prediction method according to claim 1, characterized in that: The number of layers of the LSTM network is 3, and the number of hidden units in each layer is 256.

7. A market share prediction device, characterized in that: include: The hourly market share impact feature acquisition module is used to: acquire multiple features that affect the hourly market share after the charging fee price adjustment in the target charging station; A feature historical contribution value calculation module, used to: calculate the historical contribution value of each feature among the multiple features based on the historical data of the multiple features; An hourly market share target feature screening module is used to: screen out a plurality of target features with greater influence from the plurality of features based on the historical contribution values; An hourly market share prediction module is used to: input the current data of the multiple target features into a market share prediction model to obtain the market share of the target hour one day after the current price adjustment output by the market share prediction model; The market share prediction model is obtained by training on the basis of a long short-term memory (LSTM) network using historical data of the multiple target features and the market share of the target hour one day after the historical price adjustment.

8. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the steps of the market share prediction method according to any one of claims 1 to 6 are implemented.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the market share prediction method according to any one of claims 1 to 6 are implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the market share prediction method according to any one of claims 1 to 6 are implemented.

Images