A charging station planning method, device and storage medium

Abstract

The present invention discloses a planning method, device and storage medium for a charging station, the method comprising: querying multiple charging stations in a specified geographical area; identifying the type of target equipment according to the current signal generated by the first charging pile; writing the type into the operation data of the charging station; interlacing the environmental data of the charging station and the operation data of the charging station into hierarchical first target data and second target data; dividing the charging station into multiple first sets according to the first target data; in the first set, dividing the charging station into multiple second sets according to the second target data; in the first set, selecting one of the second sets as the target set; in the same first set, selecting some charging stations from the target set as target power stations according to the geographical area, and selecting some first charging piles from the target power station as second charging piles; generating planning information for each first set. Electric bicycles and mobile terminals are charged in an orderly manner, and the resource utilization rate of the charging station is improved.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a planning method, equipment and storage medium for a charging station. Background Art

[0002] With the widespread popularity of electric bicycles, the number of electric bicycles in urban and rural areas has increased dramatically. For this reason, charging stations have been configured in many locations in urban and rural areas, such as public parking lots, large shopping malls, roadsides with parking spaces, communities, office buildings, etc. The charging stations are equipped with multiple charging piles to provide convenient charging services for electric bicycles.

[0003] Users often carry mobile terminals with them when going out, such as mobile phones, tablet computers, laptop computers, sports cameras, etc. When they are out for a long time, the power of the mobile terminals is consumed a lot. When the power is exhausted or nearly exhausted, some users go to the nearest charging station and connect the mobile terminals to the charging piles for charging.

[0004] Some charging piles are equipped with special charging chips and charging programs to provide efficient charging services for electric bicycles. If users connect their mobile terminals to these charging piles for charging, the charging efficiency will be reduced due to compatibility issues and may even cause damage to the mobile terminals.

[0005] For charging stations, electric bicycles and mobile terminals competing for charging piles will reduce the utilization rate of charging stations. Summary of the invention

[0006] In view of this, the present invention provides a charging station planning method, device and storage medium for planning the charging service of the charging station and improving the resource utilization of the charging station.

[0007] A first aspect of the present invention provides a charging station planning method, comprising:

[0008] Query multiple charging stations in a specified geographical area; each of the charging stations is provided with multiple first charging piles, and the first charging piles are used to charge electric bicycles;

[0009] When the first charging pile charges the target device, identifying the type of the target device according to the current signal generated by the first charging pile; the type includes an electric bicycle or a mobile terminal;

[0010] writing the type into the operating data of the charging station;

[0011] Interlacingly dividing the environmental data of the charging station and the operating data of the charging station into hierarchical first target data and second target data;

[0012] dividing the charging stations into a plurality of first sets according to the first target data;

[0013] In the same first set, the charging stations are divided into a plurality of second sets according to the second target data;

[0014] In the same first set, selecting one of the second sets as a target set;

[0015] In the same first set, according to the geographical area, some of the charging stations are selected from the target set as target power stations, and some of the first charging piles are selected from the target power stations as second charging piles;

[0016] Generate planning information for each of the first sets; the planning information includes converting the second charging pile to charge both the electric bicycle and the mobile terminal, and when it is identified that the first charging pile is charging the mobile terminal, executing a guidance operation to go to the second charging pile for charging.

[0017] A second aspect of the present invention provides a charging station planning device, comprising:

[0018] A charging station query module, used to query a plurality of charging stations in a specified geographical area; each of the charging stations is provided with a plurality of first charging piles, and the first charging piles are used to charge electric bicycles;

[0019] A device type identification module, used to identify the type of the target device according to the current signal generated by the first charging pile when the first charging pile charges the target device; the type includes an electric bicycle or a mobile terminal;

[0020] an operation data generating module, used for writing the type into the operation data of the charging station;

[0021] A data partitioning module, used for alternately partitioning the environmental data of the charging station and the operating data of the charging station into hierarchical first target data and second target data;

[0022] A first set division module, configured to divide the charging stations into a plurality of first sets according to the first target data;

[0023] A second set division module, configured to divide the charging stations into a plurality of second sets in the same first set according to the second target data;

[0024] A target set selection module, configured to select one of the second sets in the same first set as a target set;

[0025] a charging screening module, configured to select, in the same first set, part of the charging stations from the target set as target power stations according to the geographical area, and screen part of the first charging piles from the target power stations as second charging piles;

[0026] A planning information generation module is used to generate planning information for each of the first sets; the planning information includes converting the second charging pile to charge both the electric bicycle and the mobile terminal, and when it is identified that the first charging pile is charging the mobile terminal, executing a guidance operation to go to the second charging pile for charging.

[0027] A third aspect of the present invention provides an electronic device, the electronic device comprising:

[0028] at least one processor; and

[0029] a memory communicatively connected to the at least one processor; wherein,

[0030] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the charging station planning method as described in the first aspect above.

[0031] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the charging station planning method as described in the first aspect above.

[0032] A fifth aspect of the present invention provides a computer program product, the computer program product comprising a computer program, and when the computer program is executed by a processor, the charging station planning method as described in the first aspect above is implemented.

[0033] In this embodiment, multiple charging stations are queried in a specified geographical area; multiple first charging piles are set in each charging station, and the first charging pile is used to charge electric bicycles; when the first charging pile charges the target device, the type of the target device is identified based on the current signal generated by the first charging pile; the type includes an electric bicycle or a mobile terminal; the type is written into the operation data of the charging station; the environmental data of the charging station and the operation data of the charging station are interlaced and divided into hierarchical first target data and second target data; the charging stations are divided into multiple first sets based on the first target data; in the same first set, the charging stations are divided into multiple second sets based on the second target data; in the same first set, one of the second sets is selected as the target set; in the same first set, some charging stations are selected from the target set as target power stations based on the geographical area, and some first charging piles are selected from the target power stations as second charging piles; planning information is generated for each first set; the planning information includes converting the second charging pile to charge the electric bicycle and the mobile terminal together, and when it is identified that the first charging pile is charging the mobile terminal, a guidance operation is performed to go to the second charging pile for charging. This embodiment classifies charging stations into hierarchical categories and rationally plans charging piles. While providing charging services for electric bicycles, it also provides safe charging services for mobile terminals, improves the charging efficiency of mobile terminals, and reduces damage to mobile terminals. For charging stations, electric bicycles and mobile terminals are charged in an orderly mixed manner, which increases the frequency of charging and can improve the resource utilization of charging stations.

[0034] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 This is a flow chart of a charging station planning method provided in Embodiment 1 of the present invention.

[0037] Figure 2 It is a structural diagram of a classification model provided in Example 1 of the present invention.

[0038] Figure 3 It is a structural diagram of a charging station planning device provided in Embodiment 2 of the present invention.

[0039] Figure 4 It is a structural schematic diagram of an electronic device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can cover sequential implementations other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] Embodiment 1

[0043] See also Figure 1 , shows a flow chart of a charging station planning method provided by the first embodiment of the present invention, the method can be executed by a charging station planning device, the charging station planning device can be implemented in the form of hardware and / or software, and the charging station planning device can be configured in an electronic device. Figure 1 As shown, the method includes:

[0044] Step 101 : Query multiple charging stations in a specified geographical area.

[0045] In this embodiment, a geographical area where both users and charging stations are densely populated may be screened out from urban and rural areas, and information of multiple charging stations located in the geographical area may be queried in a system such as a GIS (Geographic Information System).

[0046] Among them, each charging station is provided with a plurality of first charging piles, and the number of first charging piles in different charging stations may be different, and each first charging pile can be used to charge the electric bicycle.

[0047] Step 102: When the first charging pile charges the target device, identify the type of the target device according to the current signal generated by the first charging pile.

[0048] In actual application, the user starts the client and scans the QR code configured for the first charging pile in the client. The client converts the QR code into information of the first charging pile (such as the ID of the first charging pile, etc.), and requests the first charging pile to provide charging services from the server based on the information of the first charging pile. When the first charging pile detects no abnormalities in its self-detection, the user is prompted to connect the target device to the first charging pile. At this time, the first charging pile charges the target device.

[0049] When the first charging pile is charging the target device, the current signal generated by the first charging pile can be recorded, and the characteristics of the current signal can be analyzed using machine learning, deep learning and other methods to identify the type of the target device.

[0050] For charging piles, the types of target devices include electric bicycles or mobile terminals.

[0051] In one embodiment of the present invention, step 102 may include the following steps:

[0052] Step 1021: Load the classification model trained for the electric bicycle and the mobile terminal.

[0053] In actual applications, there are many models of electric bicycles, and the charging protocols of different models of electric bicycles are somewhat different. There are many models of mobile terminals, and the charging protocols of different models of mobile terminals are quite different. If statistical features (such as peaks, troughs, etc.) are used in conjunction with rules to identify the type of target device, there may be certain errors.

[0054] In this embodiment, a binary classification task can be constructed for electric bicycles and mobile terminals, and a binary classification model can be constructed and trained for the classification task based on deep learning, so that the classification model can identify whether the target device is an electric bicycle or a mobile terminal based on the current signal generated by the first charging pile for the target device, thereby improving the accuracy of identifying the target device.

[0055] like Figure 2 As shown, the classification model includes a main structure Main, a first branch structure Branch_1, a second branch structure Branch_2 and a head structure Head. The modular structure not only has a clear division of labor, but also can effectively improve the efficiency of building and training the classification model under the established framework.

[0056] Step 1022: input the current signal generated by the first charging pile into the trunk structure, and extract the first trunk current characteristics in time and space.

[0057] In this embodiment, the current signal generated by the first charging pile is input into the main structure Main, and the main structure Main extracts features from the current signal in time and space, which are recorded as the first main current features, providing common basic features for the first branch structure Branch_1 and the second branch structure Branch_2.

[0058] In one design, Figure 2 As shown in the figure, the first backbone structure includes a bidirectional long short-term memory network (Bidirectional Long Short Term Memory, BiLSTM), a residual network ResNet and an atrous spatial pyramid pooling layer (Atrous Spatial Pyramid Pooling, ASPP).

[0059] In this design, the current signal generated by the first charging pile is input into the bidirectional long short-term memory network BiLSTM, and the first basic current feature is extracted in time series. BiLSTM can capture the bidirectional dependency in the current signal by combining the outputs of the two forward and backward LSTMs (Long Short-Term Memory, long short-term memory network). That is, BiLSTM runs two independent LSTMs at each time step, one LSTM from the beginning to the end of the current signal (forward), and the other LSTM from the end to the beginning of the current signal (backward). The outputs of these two LSTMs are combined together, which can simultaneously consider the contextual information of the current signal, and is suitable for binary classification tasks in a global context.

[0060] The first basic current feature is input into the residual network ResNet, and the second basic current feature is extracted spatially. Generally, the residual network ResNet can use a shallow network, such as ResNet-18, ResNet-34, etc., which can not only ensure the quality of the feature but also improve the processing speed.

[0061] The second basic current feature is input into the atrous space pyramid pooling layer ASPP to perform a pooling operation and obtain the first trunk current feature. The atrous space pyramid pooling layer ASPP samples the input second basic current feature at different sampling rates, that is, extracts features from the second basic current feature at different scales, and fuses these features to obtain the first trunk current feature.

[0062] Step 1023: input the first trunk current characteristic into the first branch structure, and extract the first branch current characteristic in time sequence.

[0063] In this embodiment, the main structure Main outputs two first main current features, one of which is input into the first branch structure Branch_1, and the first branch structure Branch_1 continues to extract features from the first main current features in time sequence, recorded as first branch current features, to strengthen the weight of time sequence.

[0064] In one design, Figure 2 As shown, the first branch structure Branch_1 includes a first long short-term memory network LSTM_1, a second long short-term memory network LSTM_2, and a multi-head attention mechanism layer (Multi Head Attention, MHA).

[0065] In this design, the first main current feature is input into the first long short-term memory network LSTM_1, and the first time-series current feature is extracted in time series. The first time-series current feature is input into the second long short-term memory network LSTM_2, and the second time-series current feature is extracted in time series. The first long short-term memory network LSTM_1 and the second long short-term memory network LSTM_2 can remember longer time interval information when processing sequences (i.e., the first main current feature and the first time-series current feature), thereby dealing with long-term dependency problems.

[0066] The second time series current feature is input into the multi-head attention mechanism layer MHA to extract the first branch current feature. For different models of mobile terminals or different models of electric bicycles, the charging current signals may have complex correlations between different frames. The multi-head attention mechanism layer MHA introduces multiple attention heads to simultaneously learn the correlations between different frames. Each attention head can focus on capturing one correlation, learning to process different semantic information, integrating the semantic information and obtaining a more comprehensive representation, so that the classification model can understand the information of the current signal more comprehensively.

[0067] Step 1024: input the first trunk current feature into the second branch structure, and extract the second branch current feature in space.

[0068] In this embodiment, the main structure Main outputs two first main current features, and another first main current feature is input into the second branch structure Branch_2. The second branch structure Branch_2 continues to extract features from the first main current features in space, recorded as second branch current features, to strengthen the spatial weight.

[0069] In one design, Figure 2As shown, the second branch structure Branch_2 includes multiple first convolutional layers Conv_1, second convolutional layers Conv_2, and compression and excitation networks (Squeeze and Excitation Networks, SE).

[0070] In this design, the first trunk current feature is sequentially input into a plurality of first convolutional layers Conv_1 to perform convolution operations to extract shallow features in space, which are recorded as the first spatial current feature.

[0071] The first spatial current feature is input into the compression and excitation network SE for processing to obtain the second spatial current feature. The compression and excitation network SE can explicitly model the interdependence between convolutional feature channels to improve the representation ability of the classification model, enabling the classification model to perform feature recalibration. Through this mechanism, it can learn to use global information to selectively emphasize informative features and suppress less useful features.

[0072] The first spatial current feature and the second spatial current feature are concatenated into a third spatial current feature to increase the information content of the feature.

[0073] The third spatial current feature is input into the second convolutional layer Conv_2 to perform a convolution operation to extract deep features in space, which is recorded as the second branch current feature.

[0074] Step 1025: concatenate the first branch current characteristic and the second branch current characteristic into a second main current characteristic.

[0075] In this embodiment, the first branch current feature and the second branch current feature may be concatenated into a second main current feature, so that the second main current feature is rich in temporal and spatial information.

[0076] Step 1026: Input the second trunk current feature into the header structure, and calculate a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal.

[0077] In this embodiment, the second trunk current feature is input into the head structure Head to perform a binary classification operation, and the head structure Head calculates a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal according to the second trunk current feature.

[0078] In one design, the head structure Head includes a third convolutional layer Conv_3, multiple fourth convolutional layers Conv_4, a fifth convolutional layer Conv_5, multiple first fully connected layers FC_1, a second fully connected layer FC_2 and a pooling layer Pooling.

[0079] In this design, the second trunk current feature is input into the third convolutional layer Conv_3 to perform a convolution operation and extract shallow features to obtain the first head current features of the two paths.

[0080] For one of the first-head current features, the first-head current feature is sequentially input into multiple fourth convolutional layers Conv_4 to perform convolution operations, and shallow-level features are further extracted to obtain the second-head current feature. Furthermore, the second-head current feature is input into the pooling layer Pooling to perform average pooling operations to obtain the third-head current feature.

[0081] For another first-head current feature, the first-head current feature is sequentially input into a plurality of first fully connected layers FC_1 and mapped into a fourth-head current feature.

[0082] The multiple fourth convolutional layers Conv_4 and the multiple first fully connected layers FC_1 are complementary to each other. The multiple fourth convolutional layers Conv_4 share the convolution kernel, are more sensitive to the characteristics of the real current signal, and can be robustly classified objects. The classification probability of the multiple first fully connected layers FC_1 is related to the characteristics of the real current signal, which can provide more accurate classification. The combination of the two can improve the efficiency of executing classification tasks.

[0083] The third-end current feature and the fourth-end current feature are concatenated into the fifth-end current feature to increase the information content of the fifth current feature.

[0084] The fifth-head current feature is input into the fifth convolutional layer Conv_5 to perform a convolution operation and extract deep features to obtain the sixth-head current feature.

[0085] The sixth head current feature is input into the second fully connected layer FC_2 and mapped into a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal.

[0086] Step 1027: If the first probability is greater than the second probability, determine that the type of the target device is an electric bicycle.

[0087] Step 1028: If the first probability is less than the second probability, determine that the type of the target device is a mobile terminal.

[0088] In this embodiment, a first probability that the target device belongs to an electric bicycle is compared with a second probability that the target device belongs to a mobile terminal.

[0089] If the first probability that the target device belongs to an electric bicycle is greater than the second probability that the target device belongs to a mobile terminal, it can be determined that the type of the target device is an electric bicycle.

[0090] If the first probability that the target device belongs to an electric bicycle is less than the second probability that the target device belongs to a mobile terminal, it can be determined that the type of the target device is a mobile terminal.

[0091] Step 103: Write the type into the operation data of the charging station.

[0092] In the same charging station, the operating information of each charging pile can be recorded each time it provides charging services, such as the charging pile ID, charging start time, charging end time, etc., and the type of target device (i.e., automatic bicycle or mobile terminal) can be written into the corresponding operating information to obtain the overall charging data of the charging station.

[0093] Step 104: alternately divide the environmental data of the charging station and the operation data of the charging station into hierarchical first target data and second target data.

[0094] For the same charging station, the environmental data of the charging station and the operation data of the charging station may be alternately divided into hierarchical first target data and second target data.

[0095] The so-called interleaving may mean that the first target data includes both the environmental data and the operating data of the charging station, and the second target data includes both the environmental data and the operating data of the charging station.

[0096] The so-called hierarchical means that the first target data is used to preliminarily classify the charging stations, and the second target data is used to further classify the charging stations based on the preliminary classification.

[0097] In one embodiment of the present invention, step 104 may include the following steps:

[0098] Step 1041: Query the environmental data of the charging station.

[0099] Building charging stations at different locations makes the environments in which the charging stations are located different, and the experience of providing charging services to users in different environments is also different.

[0100] In this embodiment, the environmental data of the charging station can be read from a system such as GIS, where the environmental data of the charging station includes the geographic location (such as longitude and latitude) and obstruction information. The obstruction information refers to information about objects that obstruct the charging station, such as awnings, trees, buildings, etc. When the charging station is open to the sky and there are no obstructions, the obstruction information is empty.

[0101] Step 1042: Determine a first time period.

[0102] In this embodiment, a plurality of first time periods (such as 1 week) may be divided on the time axis, and a plurality of second time periods (such as 1 hour) may be divided in the first time period, that is, a plurality of second time periods are included in the first time period.

[0103] Step 1043: Use the operation data of the charging station to count the mobile charging frequency, mobile charging duration ratio, and idle ratio.

[0104] In this embodiment, the operation data of the charging station can be statistically analyzed in the dimension of the first time period (including the second time period) to obtain the mobile charging frequency, mobile charging time ratio, and idle ratio.

[0105] The mobile charging frequency is the ratio between the total frequency of charging of the mobile terminal and the total frequency of charging of all target devices in the first time period.

[0106] The mobile charging duration ratio is the ratio between the total charging duration of the mobile terminal and the total charging duration of all target devices in the first time period.

[0107] The idle ratio is the ratio between the number of idle charging piles in the second time period and the total number of charging piles in the second time period within the same first time period. The idle charging pile in the second time period refers to a first charging pile that is idle (i.e., not charging any target device) within the second time period.

[0108] Step 1044: Set the geographic location and mobile charging frequency as the first target data.

[0109] Geographic location and mobile charging frequency can provide a macro analysis for classifying charging stations. Therefore, geographic location and mobile charging frequency can be set as the first target data.

[0110] Step 1045: Set the shielding information, the mobile charging time ratio and the idle time ratio as the second target data.

[0111] The occlusion information, the mobile charging time ratio and the idle ratio can provide a detailed analysis for classifying the charging stations. Therefore, the occlusion information, the mobile charging time ratio and the idle ratio can be set as the second target data.

[0112] Step 105: Divide the charging stations into a plurality of first sets according to the first target data.

[0113] In this embodiment, the charging stations may be classified according to more macroscopic first target data and divided into a plurality of first sets.

[0114] In one embodiment of the present invention, step 105 may include the following steps:

[0115] Step 1051: Initialize multiple first sets.

[0116] In this embodiment, multiple first sets may be initialized in a multi-dimensional space, each of which has a first center point. The first center point belongs to a multi-dimensional vector and may be constructed by random methods or the like.

[0117] Step 1052: for each charging station, use the geographical location as the first vector, calculate the first Euclidean distance between the first vector and each first center point, and divide the charging station into a first set with the smallest first Euclidean distance.

[0118] In this embodiment, the geographical location of the charging station can be converted into a first vector, the first Euclidean distance between the first vector and the first center point in each first set is calculated, and the first Euclidean distance between the first vector and the first center point in each first set is compared, so that the charging stations are divided into the first set with the smallest first Euclidean distance, thereby achieving the goal of clustering charging stations with similar distribution into the same first set according to their geographical locations.

[0119] Step 1053: In the first set, assign the average value between the first vectors to the first center point.

[0120] In each first set, an average value is calculated for the first vectors corresponding to all charging stations in the first set, and the average value is assigned to the first center point of the first set to update the first center point.

[0121] Step 1054: In the first set, the average value of the actual distance from the charging station to the first center point is calculated to obtain the average travel distance, the difference between the mobile charging frequency and the preset benchmark charging frequency is summed to obtain the charging deviation value, and the average travel distance and the charging deviation value are linearly merged into the original loss value.

[0122] In each first set, the actual distance from the charging station to the first center point by walking or riding a bicycle (including electric bicycles) can be queried in a GIS or other system, and the average value of the actual distance can be calculated as the average travel distance. The average travel distance can be used to measure the user's travel distance to the charging station in the first set as a whole.

[0123] Ideally, scheduling (i.e., guiding the user to the designated second charging station to charge the mobile terminal) should minimize the user's travel time and improve the user experience.

[0124] In addition, the difference between the mobile charging frequency and the preset benchmark charging frequency is calculated. The benchmark charging frequency is greater than 0, so that the difference may be positive or negative. Ideally, the charging stations in the first set are in a complementary state, that is, some charging stations are relatively busy and some charging stations are relatively idle, which facilitates scheduling.

[0125] The difference is summed up to obtain a charging deviation value, which can measure the degree of complementarity of the charging stations in the first set as a whole.

[0126] At this time, the average travel distance and the charging deviation value can be linearly fused (ie, weighted summation) to obtain the original loss value of the first set.

[0127] Step 1055: Add up the original loss values ​​to obtain the candidate loss value.

[0128] In this embodiment, the original loss values ​​of the charging stations are added together to obtain all candidate loss values ​​of the first set.

[0129] Step 1056: Calculate the sum of the overlapping areas between any two service areas to obtain the overlap amount.

[0130] Taking reducing the candidate loss value (i.e. reducing the average travel distance and lowering the charging deviation value) as the optimization goal may lead to overfitting, resulting in a certain overlap between the first sets, which will increase the management cost.

[0131] Therefore, a service area may be fitted for the first set in a system such as GIS, wherein the service area is an area fitted by the geographical locations of the charging stations in the first set.

[0132] Compare the service areas pairwise, calculate the overlapping areas between the service areas pairwise, and sum the overlapping areas between the service areas pairwise to get the overlap amount, which can be used as a penalty item to add constraints to the optimization objective.

[0133] Step 1057: Linearly fuse the candidate loss value and the overlap amount into a target loss value.

[0134] In this embodiment, the candidate loss value and the overlap amount are linearly fused (ie, weighted summed) into a target loss value.

[0135] Step 1058, determine whether the target loss value meets the preset first classification condition; if so, execute step 1059; if not, return to execute step 1052.

[0136] Step 1059: output multiple first sets.

[0137] In each round of iteration, it can be determined whether the target loss value meets the preset first classification condition, for example, whether the target loss value is less than a certain threshold, whether the change range of the target loss value is less than a certain threshold, and so on.

[0138] If the target loss value satisfies the first classification condition, it can be determined that the first set converges. At this time, multiple first sets (including charging stations) can be output.

[0139] If the target loss value does not meet the first classification condition, it can be determined that the first set has not converged, and the next round of iteration is entered to execute steps 1052 to 1058.

[0140] Step 106: In the same first set, the charging stations are divided into a plurality of second sets according to the second target data.

[0141] In each first set, the charging stations may be further classified according to more detailed second target data and divided into multiple second sets.

[0142] In one embodiment of the present invention, step 106 may include the following steps:

[0143] Step 1061: Initialize multiple second sets in the same first set.

[0144] In this embodiment, multiple first sets can be initialized in each first set respectively, each second set has a second center point, and the second center point belongs to a multidimensional vector and can be constructed by random method or the like.

[0145] Step 1062: for each charging station, use the occlusion information, mobile charging time ratio and idle ratio as the second vector, calculate the second Euclidean distance between the second vector and each second center point, and divide the charging station into a second set with the smallest second Euclidean distance.

[0146] In the same first set, the occlusion information, mobile charging time ratio and idle ratio of each charging station are converted into a second vector. For the same charging station, the second Euclidean distance between the second vector of the charging station and each second center point is calculated. For the same charging station, the second Euclidean distance between the second vector of the charging station and each second center point is compared, and the charging station is divided into the second set with the smallest second Euclidean distance.

[0147] Step 1063: In the second set, assign the average value between the second vectors to the second center point.

[0148] In each second set, an average value is calculated for the second vectors corresponding to all charging stations in the second set, and the average value is assigned to the second center point of the second set to update the second center point.

[0149] Step 1064, determine whether the second center point meets the preset second classification condition; if so, execute step 1065; if not, return to execute step 1062.

[0150] Step 1065: output multiple second sets.

[0151] In each round of iteration, it can be determined whether the second center point satisfies a preset second classification condition, for example, whether the variation range of the second center point is less than a certain threshold, and so on.

[0152] If the second center point satisfies the second classification condition, it can be determined that the second set converges. In this case, multiple second sets (including charging stations) can be output.

[0153] If the second center point does not satisfy the second classification condition, it can be determined that the second set has not converged, and the next round of iteration is entered to execute steps 1062 to 1064.

[0154] Step 107: In the same first set, select one of the second sets as the target set.

[0155] In this embodiment, in the same first set, the second sets in the first set can be compared, and one of the second sets suitable for transforming the charging station is selected and recorded as the target set.

[0156] In a specific implementation, in the same first set, the second center points in each second set are inversely quantized and mapped respectively into a first eigenvalue representing the occlusion information, a second eigenvalue representing the proportion of mobile charging time, and a third eigenvalue representing the idle proportion.

[0157] For each second set, the first eigenvalue, the second eigenvalue and the third eigenvalue are linearly fused (ie, weighted summed) into an index value, and the index value can comprehensively represent the idle status of the charging station.

[0158] In the same first set, the index values ​​of the second sets are compared, and the second set with the largest index value (ie, the most idle) is selected as the target set.

[0159] Step 108: In the same first set, select some charging stations from the target set as target power stations according to the geographical area, and select some first charging piles from the target power stations as second charging piles.

[0160] In this embodiment, each first set can be traversed, and in the target set of the same first set, some charging stations can be selected as target power stations according to information about their geographical areas, and some first charging piles can be selected from the target power stations as second charging piles, and the target power stations and second charging piles can be dynamically set to further improve the resource utilization of the charging stations.

[0161] In a specific implementation, the charging areas covered by the charging stations in the same first set may be queried in the geographical area.

[0162] All public service buildings (such as shopping malls, markets, etc.) in the charging area are mapped to service values ​​by table lookup or other means; the service value is positively correlated with the total flow of people in all public service buildings, that is, the higher the total flow of people in all public service buildings, the greater the service value, and conversely, the lower the total flow of people in all public service buildings, the smaller the service value.

[0163] Under the constraint of discretized target power stations, that is, the distance between target power stations is increased as much as possible to provide users with more convenient charging services. Some charging stations are selected as target power stations from the target set according to the service value using piecewise functions or other methods. The number of target power stations is positively correlated with the service value, that is, the larger the service value, the more target power stations there are to meet the user's demand for charging mobile terminals. Conversely, the smaller the service value, the fewer target power stations there are to reduce the cost of transforming charging stations while meeting the user's demand for charging mobile terminals.

[0164] A piecewise function or the like is used to select some of the first charging piles from the target power station as second charging piles according to the service value; wherein the number of the second charging piles is positively correlated with the service value, that is, the greater the service value, the greater the number of the second charging piles to meet the user's demand for charging the mobile terminal; conversely, the smaller the service value, the smaller the number of the second charging piles to reduce the cost of transforming the first charging piles while meeting the user's demand for charging the mobile terminal.

[0165] Step 109: Generate planning information for each first set.

[0166] In this embodiment, each first set can be traversed, and planning information can be generated for each first set for reference by operation and maintenance personnel of the charging station.

[0167] On the one hand, the planning information includes converting the second charging pile to charge both electric bicycles and mobile terminals. The hybrid charging method can improve the resource utilization of the second charging pile. The conversion method may include replacing the charging chip and / or updating the charging program (such as the driver, charging protocol, etc.) to charge the electric bicycle and the mobile terminal at the same time in a compatible manner, thereby improving the compatibility of the second charging pile.

[0168] On the other hand, the planning information includes, when it is identified that the first charging pile is charging the mobile terminal, executing a guidance operation to go to the second charging pile for charging. The guidance operation may include mandatory guidance, for example, stopping the first charging pile from charging the mobile terminal, determining that the user is in breach of contract for using the first charging pile to charge the mobile terminal, etc. It may also include encouraging guidance, for example, displaying the path from the current first charging pile to the corresponding (i.e., the same first set) second charging pile on the user's client, canceling the user's breach of contract for using the first charging pile to charge the mobile terminal, giving discounts to users who go to the second charging pile to charge the mobile terminal, etc.

[0169] In this embodiment, multiple charging stations are queried in a specified geographical area; multiple first charging piles are set in each charging station, and the first charging pile is used to charge electric bicycles; when the first charging pile charges the target device, the type of the target device is identified based on the current signal generated by the first charging pile; the type includes an electric bicycle or a mobile terminal; the type is written into the operation data of the charging station; the environmental data of the charging station and the operation data of the charging station are interlaced and divided into hierarchical first target data and second target data; the charging stations are divided into multiple first sets based on the first target data; in the same first set, the charging stations are divided into multiple second sets based on the second target data; in the same first set, one of the second sets is selected as the target set; in the same first set, some charging stations are selected from the target set as target power stations based on the geographical area, and some first charging piles are selected from the target power stations as second charging piles; planning information is generated for each first set; the planning information includes converting the second charging pile to charge the electric bicycle and the mobile terminal together, and when it is identified that the first charging pile is charging the mobile terminal, a guidance operation is performed to go to the second charging pile for charging. This embodiment classifies charging stations into hierarchical categories and rationally plans charging piles. While providing charging services for electric bicycles, it also provides safe charging services for mobile terminals, improves the charging efficiency of mobile terminals, and reduces damage to mobile terminals. For charging stations, electric bicycles and mobile terminals are charged in an orderly mixed manner, which increases the frequency of charging and can improve the resource utilization of charging stations.

[0170] In addition, by improving the resource utilization of the charging station, the operational efficiency of the charging station can be effectively improved.

[0171] Embodiment 2

[0172] See also Figure 3 , shows a schematic diagram of the structure of a charging station planning device provided in Embodiment 2 of the present invention. Figure 3 As shown, the device comprises:

[0173] The charging station query module 301 is used to query a plurality of charging stations in a specified geographical area; each of the charging stations is provided with a plurality of first charging piles, and the first charging piles are used to charge electric bicycles;

[0174] A device type identification module 302 is used to identify the type of the target device according to the current signal generated by the first charging pile when the first charging pile charges the target device; the type includes an electric bicycle or a mobile terminal;

[0175] An operation data generating module 303, used for writing the type into the operation data of the charging station;

[0176] A data partitioning module 304, configured to alternately partition the environmental data of the charging station and the operating data of the charging station into hierarchical first target data and second target data;

[0177] A first set division module 305, configured to divide the charging stations into a plurality of first sets according to the first target data;

[0178] A second set division module 306, configured to divide the charging stations into a plurality of second sets in the same first set according to the second target data;

[0179] A target set selection module 307, configured to select one of the second sets in the same first set as a target set;

[0180] A charging screening module 308, configured to select, in the same first set, some of the charging stations from the target set as target power stations according to the geographical area, and screen some of the first charging piles from the target power stations as second charging piles;

[0181] The planning information generation module 309 is used to generate planning information for each of the first sets; the planning information includes converting the second charging pile to charge both the electric bicycle and the mobile terminal, and when it is identified that the first charging pile is charging the mobile terminal, executing a guidance operation to go to the second charging pile for charging.

[0182] In one embodiment of the present invention, the device type identification module 302 includes:

[0183] A classification model loading module, used to load a classification model trained jointly for an electric bicycle and a mobile terminal; the classification model includes a trunk structure, a first branch structure, a second branch structure and a head structure;

[0184] A trunk processing module, used for inputting the current signal generated by the first charging pile into the trunk structure and extracting the first trunk current characteristics in time sequence and space;

[0185] A first branch processing module, used for inputting the first trunk current feature into the first branch structure and extracting the first branch current feature in time sequence;

[0186] A second branch processing module, used for inputting the first trunk current feature into the second branch structure and extracting the second branch current feature in space;

[0187] A trunk splicing module, used for splicing the first branch current characteristic and the second branch current characteristic into a second trunk current characteristic;

[0188] a header processing module, configured to input the second trunk current feature into the header structure, and respectively calculate a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal;

[0189] an electric bicycle determination module, configured to determine that the type of the target device is an electric bicycle if the first probability is greater than the second probability;

[0190] A mobile terminal determination module is configured to determine that the type of the target device is a mobile terminal if the first probability is less than the second probability.

[0191] In one embodiment of the present invention, the first backbone structure includes a bidirectional long short-term memory network, a residual network and a dilated spatial pyramid pooling layer; the backbone processing module is further used for:

[0192] Inputting the current signal generated by the first charging pile into the bidirectional long short-term memory network and extracting the first basic current feature in time sequence;

[0193] Inputting the first basic current feature into the residual network and extracting the second basic current feature in space;

[0194] Inputting the second basic current feature into the hole space pyramid pooling layer to perform a pooling operation to obtain a first trunk current feature;

[0195] The first branch structure includes a first long short-term memory network, a second long short-term memory network, and a multi-head attention mechanism layer; the first branch processing module is also used for:

[0196] Inputting the first trunk current feature into the first long short-term memory network and extracting the first time series current feature in time series;

[0197] Inputting the first time-series current feature into the second long short-term memory network and extracting the second time-series current feature in time sequence;

[0198] Inputting the second time series current feature into the multi-head attention mechanism layer to extract the first branch current feature;

[0199] The second branch structure includes a plurality of first convolutional layers, a second convolutional layer, a compression and excitation network; the second branch processing module is further used for:

[0200] Inputting the first trunk current feature into the first convolution layers in sequence to perform convolution operations, so as to extract the first spatial current feature in space;

[0201] Inputting the first space current characteristic into the compression and excitation network for processing to obtain a second space current characteristic;

[0202] splicing the first spatial current feature and the second spatial current feature into a third spatial current feature;

[0203] Inputting the third spatial current feature into the second convolution layer to perform a convolution operation to extract the second branch current feature in space;

[0204] The head structure includes a third convolutional layer, a plurality of fourth convolutional layers, a fifth convolutional layer, a plurality of first fully connected layers, a second fully connected layer and a pooling layer; the head processing module is further used for:

[0205] Inputting the second trunk current feature into the third convolution layer to perform a convolution operation to obtain a first head current feature;

[0206] Inputting the first head current feature into the plurality of fourth convolutional layers in sequence to perform convolution operations to obtain a second head current feature;

[0207] Inputting the second head current feature into the pooling layer to perform an average pooling operation to obtain a third head current feature;

[0208] Inputting the first head current feature into a plurality of the first fully connected layers in sequence to map into a fourth head current feature;

[0209] splicing the third head current characteristic and the fourth head current characteristic into a fifth head current characteristic;

[0210] Inputting the fifth head current feature into the fifth convolution layer to perform a convolution operation to obtain a sixth head current feature;

[0211] The sixth head current feature is input into the second fully connected layer and mapped into a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal.

[0212] In the first embodiment of the present invention, the data partitioning module 304 is further used for:

[0213] Querying environmental data of the charging station; the environmental data includes geographic location and shielding information; the shielding information refers to information about objects shielding the charging station;

[0214] Determine a first time period; the first time period has a plurality of second time periods;

[0215] Using the operation data of the charging station to count the mobile charging frequency, mobile charging duration ratio, and idle ratio;

[0216] Setting the geographical location and the mobile charging frequency as first target data;

[0217] The shielding information, the mobile charging duration ratio and the idle ratio are set as second target data;

[0218] The mobile charging frequency is the ratio of the total frequency of charging of the mobile terminal to the total frequency of charging of all the target devices within the first time period;

[0219] The mobile charging duration ratio is the ratio between the total charging duration of the mobile terminal and the total charging duration of all the target devices in the first time period;

[0220] The idle ratio is the ratio between the number of idle first time periods and the total number of idle second time periods within the same first time period. Idle second time period means that there are idle first charging piles within the second time period.

[0221] In the first embodiment of the present invention, the first set partitioning module 305 is further configured to:

[0222] Initializing a plurality of first sets; each of the first sets has a first center point;

[0223] For each of the charging stations, using the geographical location as a first vector, calculating a first Euclidean distance between the first vector and each of the first center points, and dividing the charging stations into the first set with the smallest first Euclidean distance;

[0224] In the first set, assigning an average value between the first vectors to the first center point;

[0225] In the first set, an average value of the actual distance from the charging station to the first center point is calculated to obtain an average travel distance, a difference between the mobile charging frequency and a preset reference charging frequency is summed to obtain a charging deviation value, and the average travel distance and the charging deviation value are linearly merged into an original loss value;

[0226] Adding the original loss values ​​to obtain a candidate loss value;

[0227] Calculating the sum of the overlapping areas between any two service areas to obtain an overlapping amount; the service area is an area fitted by the geographical locations of the charging stations in the first set;

[0228] Linearly fuse the candidate loss value and the overlap amount into a target loss value;

[0229] Determine whether the target loss value satisfies a preset first classification condition; if so, output a plurality of the first sets; if not, return to execute the step of taking the geographic location as the first vector, calculating the first Euclidean distance between the first vector and each of the first center points for each of the charging stations, and dividing the charging stations into the first set with the smallest first Euclidean distance.

[0230] In the first embodiment of the present invention, the second set partitioning module 306 is further configured to:

[0231] In the same first set, multiple second sets are initialized; each of the second sets has a second center point;

[0232] For each of the charging stations, using the shielding information, the mobile charging duration ratio and the idle ratio as a second vector, calculating a second Euclidean distance between the second vector and each of the second center points, and dividing the charging stations into the second set with the smallest second Euclidean distance;

[0233] In the second set, assigning an average value between the second vectors to the second center point;

[0234] Determine whether the second center point meets the preset second classification condition; if so, output multiple second sets; if not, return to execute the second Euclidean distance between the second vector and each second center point for each charging station, and divide the charging station into the second set with the smallest second Euclidean distance, using the occlusion information, the mobile charging time ratio and the idle ratio as the second vector.

[0235] In the first embodiment of the present invention, the target set selection module 307 is further used to:

[0236] In the same first set, the second center points in each of the second sets are respectively mapped to a first eigenvalue representing the occlusion information, a second eigenvalue representing the mobile charging duration ratio, and a third eigenvalue representing the idle ratio;

[0237] For each of the second sets, linearly fuse the first eigenvalue, the second eigenvalue and the third eigenvalue into an index value;

[0238] In the same first set, the second set with the largest index value is selected as the target set.

[0239] In the first embodiment of the present invention, the charging screening module 308 is further used for:

[0240] Querying, in the geographic area, charging areas covered by the charging stations in the same first set;

[0241] Mapping all public service buildings in the charging area into service values; the service values ​​are positively correlated with the total flow of people in all the public service buildings;

[0242] Under the constraint of the discretized target power stations, some of the charging stations are selected from the target set as target power stations according to the service value; the number of the target power stations is positively correlated with the service value;

[0243] According to the service value, some of the first charging piles are selected from the target power station as second charging piles; and the number of the second charging piles is positively correlated with the service value.

[0244] The charging station planning device provided in the embodiment of the present invention can execute the charging station planning method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the charging station planning method.

[0245] Embodiment 3

[0246] See also Figure 4 , shows a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0247] like Figure 4As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0248] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0249] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the planning method of the charging station.

[0250] In some embodiments, the planning method of the charging station may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the planning method of the charging station described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the planning method of the charging station in any other appropriate manner (e.g., by means of firmware).

[0251] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0252] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0253] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0254] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0255] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0256] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0257] Embodiment 4

[0258] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the charging station planning method provided by any embodiment of the present invention is implemented.

[0259] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, through the Internet using an Internet service provider).

[0260] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0261] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for planning a charging station, characterized in that: include: Query multiple charging stations in a specified geographical area; each of the charging stations is provided with multiple first charging piles, and the first charging piles are used to charge electric bicycles; When the first charging pile charges the target device, identifying the type of the target device according to the current signal generated by the first charging pile; the type includes an electric bicycle or a mobile terminal; writing the type into the operating data of the charging station; Interlacingly dividing the environmental data of the charging station and the operating data of the charging station into hierarchical first target data and second target data; dividing the charging stations into a plurality of first sets according to the first target data; In the same first set, the charging stations are divided into a plurality of second sets according to the second target data; In the same first set, selecting one of the second sets as a target set; In the same first set, selecting some of the charging stations from the target set as target power stations according to the geographical area, and selecting some of the first charging piles from the target power stations as second charging piles; Generate planning information for each of the first sets; the planning information includes converting the second charging pile to charge the electric bicycle and the mobile terminal together, and when it is identified that the first charging pile is charging the mobile terminal, executing a guidance operation to go to the second charging pile for charging; The step of alternately dividing the environmental data of the charging station and the operating data of the charging station into hierarchical first target data and second target data includes: Querying environmental data of the charging station; the environmental data includes geographic location and shielding information; the shielding information refers to information about objects shielding the charging station; Determine a first time period; the first time period has a plurality of second time periods; Using the operation data of the charging station to count the mobile charging frequency, mobile charging duration ratio, and idle ratio; Setting the geographical location and the mobile charging frequency as first target data; The shielding information, the mobile charging duration ratio and the idle ratio are set as second target data; The mobile charging frequency is the ratio of the total frequency of charging of the mobile terminal to the total frequency of charging of all the target devices within the first time period; The mobile charging duration ratio is the ratio between the total charging duration of the mobile terminal and the total charging duration of all the target devices during the first time period; The idle ratio is the ratio between the number of idle first charging piles in the second time period and the total number of idle first charging piles in the second time period within the same first time period, and the idle second time period means that there are idle first charging piles in the second time period; The dividing the charging stations into a plurality of first sets according to the first target data comprises: Initializing a plurality of first sets; each of the first sets has a first center point; For each of the charging stations, using the geographical location as a first vector, calculating a first Euclidean distance between the first vector and each of the first center points, and dividing the charging stations into the first set with the smallest first Euclidean distance; In the first set, assigning an average value between the first vectors to the first center point; In the first set, an average value is calculated for the actual distance from the charging station to the first center point to obtain an average travel distance, a difference between the mobile charging frequency and a preset reference charging frequency is summed to obtain a charging deviation value, and the average travel distance and the charging deviation value are linearly merged into an original loss value; Adding the original loss values ​​to obtain a candidate loss value; Calculating the sum of the overlapping areas between any two service areas to obtain an overlapping amount; the service area is an area fitted by the geographical locations of the charging stations in the first set; Linearly fuse the candidate loss value and the overlap amount into a target loss value; Determine whether the target loss value satisfies a preset first classification condition; if so, output a plurality of the first sets; if not, return to execute the step of calculating, for each of the charging stations, a first Euclidean distance between the first vector and each of the first center points using the geographical location as a first vector, and classifying the charging station into the first set with the smallest first Euclidean distance; The step of dividing the charging stations into a plurality of second sets in the same first set according to the second target data comprises: In the same first set, multiple second sets are initialized; each of the second sets has a second center point; For each of the charging stations, using the shielding information, the mobile charging duration ratio and the idle ratio as a second vector, calculating a second Euclidean distance between the second vector and each of the second center points, and dividing the charging stations into the second set with the smallest second Euclidean distance; In the second set, assigning an average value between the second vectors to the second center point; Determine whether the second center point meets the preset second classification condition; if so, output a plurality of the second sets; if not, return to execute the step of calculating the second Euclidean distance between the second vector and each of the second center points for each of the charging stations, using the occlusion information, the mobile charging time ratio, and the idle ratio as the second vector, and classifying the charging station into the second set with the smallest second Euclidean distance; The selecting one of the second sets as a target set in the same first set includes: In the same first set, the second center points in each of the second sets are respectively mapped to a first eigenvalue representing the occlusion information, a second eigenvalue representing the mobile charging duration ratio, and a third eigenvalue representing the idle ratio; For each of the second sets, linearly fuse the first eigenvalue, the second eigenvalue and the third eigenvalue into an index value; In the same first set, the second set with the largest index value is selected as the target set.

2. The method according to claim 1, characterized in that The identifying the type of the target device according to the current signal generated by the first charging pile includes: Loading a classification model trained jointly for electric bicycles and mobile terminals; the classification model includes a trunk structure, a first branch structure, a second branch structure and a head structure; Inputting the current signal generated by the first charging pile into the trunk structure, and extracting the first trunk current characteristics in time sequence and space; Inputting the first trunk current feature into the first branch structure, and extracting the first branch current feature in time sequence; Inputting the first trunk current feature into the second branch structure to extract the second branch current feature in space; splicing the first branch current characteristic and the second branch current characteristic into a second main current characteristic; Inputting the second trunk current feature into the header structure, respectively calculating a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal; If the first probability is greater than the second probability, determining that the type of the target device is an electric bicycle; If the first probability is less than the second probability, it is determined that the type of the target device is a mobile terminal.

3. The method according to claim 2, characterized in that The backbone structure includes a bidirectional long short-term memory network, a residual network and a void space pyramid pooling layer; the current signal generated by the first charging pile is input into the backbone structure, and the first backbone current feature is extracted in time and space, including: Inputting the current signal generated by the first charging pile into the bidirectional long short-term memory network to extract the first basic current feature in time sequence; Inputting the first basic current feature into the residual network to extract the second basic current feature in space; Inputting the second basic current feature into the hole space pyramid pooling layer to perform a pooling operation to obtain a first trunk current feature; The first branch structure includes a first long short-term memory network, a second long short-term memory network, and a multi-head attention mechanism layer; the first trunk current feature is input into the first branch structure, and the first branch current feature is extracted in time sequence, including: Inputting the first trunk current feature into the first long short-term memory network to extract the first time series current feature in time series; Inputting the first time-series current feature into the second long short-term memory network to extract the second time-series current feature in time sequence; Inputting the second time series current feature into the multi-head attention mechanism layer to extract the first branch current feature; The second branch structure includes a plurality of first convolutional layers, a second convolutional layer, a compression and excitation network; the inputting the first trunk current feature into the second branch structure and extracting the second branch current feature spatially includes: Inputting the first trunk current feature into the plurality of first convolutional layers in sequence to perform a convolution operation, so as to extract a first spatial current feature in space; Inputting the first space current characteristic into the compression and excitation network for processing to obtain a second space current characteristic; splicing the first spatial current feature and the second spatial current feature into a third spatial current feature; Inputting the third spatial current feature into the second convolution layer to perform a convolution operation, so as to extract the second branch current feature in space; The header structure includes a third convolutional layer, a plurality of fourth convolutional layers, a fifth convolutional layer, a plurality of first fully connected layers, a second fully connected layer and a pooling layer; the inputting the second trunk current feature into the header structure, respectively calculating a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal, including: Inputting the second trunk current feature into the third convolution layer to perform a convolution operation to obtain a first head current feature; Inputting the first head current feature into the plurality of fourth convolutional layers in sequence to perform convolution operations, thereby obtaining a second head current feature; Inputting the second head current feature into the pooling layer to perform an average pooling operation to obtain a third head current feature; Inputting the first head current feature into a plurality of the first fully connected layers in sequence to map into a fourth head current feature; splicing the third head current characteristic and the fourth head current characteristic into a fifth head current characteristic; Inputting the fifth head current feature into the fifth convolutional layer to perform a convolution operation to obtain a sixth head current feature; The sixth head current feature is input into the second fully connected layer and mapped into a first probability that the target device belongs to an electric bicycle and a second probability that the target device belongs to a mobile terminal.

4. The method according to any one of claims 1 to 3, characterized in that In the same first set, selecting some of the charging stations from the target set as target power stations according to the geographical area, and selecting some of the first charging piles from the target power stations as second charging piles, comprises: Querying, in the geographic area, charging areas covered by the charging stations in the same first set; Mapping all public service buildings in the charging area into service values; the service values ​​are positively correlated with the total flow of people in all the public service buildings; Under the constraint of the discretized target power stations, some of the charging stations are selected from the target set as target power stations according to the service value; the number of the target power stations is positively correlated with the service value; According to the service value, some of the first charging piles are selected from the target power station as second charging piles; and the number of the second charging piles is positively correlated with the service value.

5. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the charging station planning method according to any one of claims 1 to 4.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the charging station planning method according to any one of claims 1 to 4 is implemented.

Images