Charging pile scheduling method and system based on dynamic data fusion
Through the charging pile scheduling method and system with dynamic data fusion, the problem of inefficient use of charging piles is solved, and the charging efficiency and user experience is improved. By real-time matching and optimization of charging pile resources are met, personalized needs are met.
Patent Information
- Application Number
- CN202411104956.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-13
AI Technical Summary
The existing charging pile scheduling methods lack comprehensive considerations for real-time operation data and user appointment information, resulting in low efficiency in charging piles, users' waiting time is too long, and cannot meet personalized needs, which increases charging costs.
Using charging pile scheduling methods and systems based on dynamic data fusion, the real-time operation status of the charging station end and the reservation charging timing of the user end are obtained in real time, and the charging piles that meet the conditions are selected, and the comprehensive performance of the charging pile is optimized and matched to generate the most Excellent charging paths to improve charging efficiency and user experience.
It improves the scheduling efficiency of charging piles, reduces user waiting time, improves user charging experience, ensures the efficiency and safety of the charging process, and meets personalized needs.
Smart Images

Figure CN118644051B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of smart energy management technology, specifically to the field of charging pile scheduling technology, and in particular to a charging pile scheduling method and system based on dynamic data fusion. Background Art
[0002] With the rapid expansion of the electric vehicle market, charging demand has shown explosive growth, which has put forward higher requirements for the scheduling and management of charging piles. However, the existing charging pile scheduling methods lack comprehensive consideration of real-time operation data and user reservation information, resulting in low efficiency of charging piles and poor user charging experience. In the existing scheduling methods, the real-time operation status of the charging station and the scheduled charging sequence of the user are often ignored or handled improperly. This leads to a shortage of charging piles during peak hours and low utilization during off-peak hours, resulting in a waste of resources. At the same time, due to the lack of sufficient analysis of user reservation information, such as timing analysis, the allocation of charging piles often cannot meet the personalized needs of users, increasing the waiting time and charging costs of users. Summary of the invention
[0003] The present application provides a charging pile scheduling method and system based on dynamic data fusion, aiming to solve the technical problem that low charging efficiency causes users to wait too long, resulting in poor user experience.
[0004] In view of the above problems, the present application provides a charging pile scheduling method and system based on dynamic data fusion.
[0005] In a first aspect disclosed in the present application, a charging pile scheduling method based on dynamic data fusion is provided, the method comprising: dynamically acquiring the real-time operating status of the charging station end, the real-time operating status comprising multiple groups of real-time usage information of multiple charging piles in the charging station; dynamically acquiring the scheduled charging sequence of the user end, the scheduled charging sequence comprising the first scheduled charging information of the first user; when the real-time operating status meets the first predetermined state constraint, extracting the first real-time usage information from the multiple groups of real-time usage information, and matching the first charging pile corresponding to the first real-time usage information among the multiple charging piles; when the first real-time usage information meets the second predetermined state constraint, adding the first charging pile to the candidate charging pile set; using the comprehensive performance of the charging pile as an optimization evaluation index, and using the candidate charging pile set as an optimization space; performing optimization analysis in the optimization space based on the optimization evaluation index to obtain a target charging pile; and sending the optimal charging path generated by combining the first position of the first user and the target position of the target charging pile to the first user, the optimal charging path having an optimal execution time identifier.
[0006] Another aspect disclosed in the present application provides a charging pile scheduling system based on dynamic data fusion, the system comprising: an operation status acquisition module, the operation status acquisition module is used to dynamically acquire the real-time operation status of the charging station end, the real-time operation status includes multiple groups of real-time usage information of multiple charging piles in the charging station; a charging sequence acquisition module, the charging sequence acquisition module is used to dynamically acquire the scheduled charging sequence of the user end, the scheduled charging sequence includes the first scheduled charging information of the first user; an information matching module, the information matching module is used to extract the first real-time usage information from the multiple groups of real-time usage information when the real-time operation status meets the first predetermined state constraint, and match the first real-time usage information corresponding to the multiple charging piles. A first charging pile; a charging pile adding module, the charging pile adding module is used to add the first charging pile to the candidate charging pile set when the first real-time usage information meets the second predetermined state constraint; an optimization setting module, the optimization setting module is used to use the comprehensive performance of the charging pile as an optimization evaluation index and the candidate charging pile set as an optimization space; an optimization analysis module, the optimization analysis module is used to perform optimization analysis in the optimization space based on the optimization evaluation index to obtain a target charging pile; an optimal charging path generation module, the optimal charging path generation module is used to send an optimal charging path generated by combining the first position of the first user and the target position of the target charging pile to the first user, the optimal charging path having an optimal execution time identifier.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0008] The above-mentioned charging pile scheduling method based on dynamic data fusion first collects the usage of each charging pile in the charging station and the user's charging reservation information in real time. When the real-time status of the charging station meets specific conditions, the charging piles that meet the requirements will be screened out and included in the candidate charging pile list. Subsequently, by comprehensively considering the performance of the charging piles, the best target charging pile is selected from the candidate list. After that, an optimal path to the target charging pile is planned for the user, and the user's estimated arrival time is sent. This method aims to improve the scheduling efficiency of charging piles, reduce user waiting time, and thus enhance the user's charging experience.
[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] 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.
[0011] Figure 1 Schematic diagram of a flow chart of a charging pile scheduling method based on dynamic data fusion in one embodiment;
[0012] Figure 2 The figure is a diagram of the architecture of a charging pile scheduling system based on dynamic data fusion in one embodiment.
[0013] Explanation of the reference numerals: operating status acquisition module 1, charging timing acquisition module 2, information matching module 3, charging pile adding module 4, optimization setting module 5, optimization analysis module 6, optimal charging path generation module 7. DETAILED DESCRIPTION
[0014] The embodiments of the present application provide a charging pile scheduling method and system based on dynamic data fusion to solve the technical problem that low charging efficiency causes users to wait too long, resulting in a poor user experience.
[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.
[0017] Embodiment 1
[0018] like Figure 1 As shown, the present application provides a charging pile scheduling method based on dynamic data fusion, the charging pile scheduling method is applied to a charging pile scheduling system, the charging pile scheduling system is communicatively connected with a multi-data source end, the multi-data source end includes a charging station end and a user end, and the charging pile scheduling includes:
[0019] Dynamically obtain the real-time operating status of the charging station, wherein the real-time operating status includes multiple groups of real-time usage information of multiple charging piles in the charging station;
[0020] The charging pile scheduling method based on dynamic data fusion is applied to the charging pile scheduling system. This scheduling system communicates with multiple data sources, including the charging station side and the user side. By collecting and analyzing information from these data sources in real time, charging piles can be scheduled more intelligently to meet the charging needs of users and improve the overall charging efficiency.
[0021] In an embodiment of the present application, the system terminal collects and updates the usage information of each charging pile in the charging station in real time from the charging station end. This information includes whether the charging pile is idle, the current charging power, the charging time, the expected full time, etc. By obtaining this usage information in real time, it is possible to understand the real-time status of the charging pile and provide data support for subsequent intelligent scheduling. This step is a key link in achieving efficient charging pile scheduling, which helps to ensure that charging pile resources are reasonably allocated and utilized.
[0022] Dynamically acquiring a scheduled charging sequence of the user terminal, wherein the scheduled charging sequence includes first scheduled charging information of a first user;
[0023] In one embodiment, the system terminal collects and organizes the scheduled charging information of multiple users from the user terminal in real time, and sorts them according to the time to obtain the scheduled charging sequence. Among them, the first one in the scheduled charging sequence is the scheduled charging information of the user with the earliest scheduled use time, that is, the first scheduled charging information of the first user. By sorting the scheduled charging information, it can not only ensure that the first user can start charging on time, but also optimize the scheduling strategy of the entire charging station according to their reservation information, and improve the use efficiency of the charging piles.
[0024] When the real-time operating state meets the first predetermined state constraint, extracting the first real-time usage information from the multiple sets of real-time usage information, and matching the first charging pile corresponding to the first real-time usage information from the multiple charging piles;
[0025] In one embodiment, in order to effectively utilize idle charging pile resources and improve charging service efficiency, it is necessary to compare with the first predetermined state constraint. The first predetermined state constraint refers to the constraint used to limit whether there are at least one or more charging piles in an idle state, that is, no user is using them. When the real-time operating state of the charging pile meets the first predetermined state constraint, that is, there are idle charging piles, the system terminal will further process this information. First, the first real-time usage information related to the idle charging pile is extracted from multiple groups of real-time usage information. Subsequently, based on this information, the first charging pile that matches the first real-time usage information is screened out from multiple charging piles. This step is to ensure that when there are idle charging piles, a charging pile that suits the current needs can be quickly found to prepare for the next scheduling or appointment. Through such a processing method, the charging pile resources can be utilized more efficiently and the user's charging experience can be improved.
[0026] When the first real-time usage information meets a second predetermined state constraint, adding the first charging pile to a set of candidate charging piles;
[0027] In one embodiment, in order to further screen out the most suitable charging pile, it is necessary to determine whether the first real-time usage information satisfies the second predetermined state constraint. The second predetermined state constraint refers to limiting the usage of the current charging pile, that is, whether it is in an idle state. When the first real-time usage information of the first charging pile meets the second predetermined state constraint, that is, there is no ongoing charging task and no other appointment occupancy. The system terminal adds the charging piles that meet this condition to the list of candidate charging piles as candidates for subsequent selection of target charging piles. Such a screening process ensures that the candidate charging piles are truly available and waiting to be allocated, providing basic data for subsequent optimization scheduling.
[0028] Taking the comprehensive performance of the charging pile as the optimization evaluation index, and taking the set of candidate charging piles as the optimization space;
[0029] In one embodiment, during the charging pile scheduling process, in order to select the best charging pile for users to use, the system terminal will comprehensively consider the comprehensive performance of the charging pile, and use the comprehensive performance of the charging pile as an optimization evaluation indicator. This includes multiple aspects such as the charging speed, efficiency, and stability of the charging pile, aiming to find the charging pile with the best current charging performance for users. The candidate charging pile set is a set of idle charging piles selected by the system terminal from a large number of charging piles, and these charging piles may be the objects that users will eventually use. Therefore, the system terminal will use this candidate charging pile set as the basis for optimization, that is, the optimization space, and combine the comprehensive performance of the charging pile for optimization analysis, so as to select the best charging pile for users and ensure that users get the best charging experience.
[0030] Performing an optimization analysis in the optimization space based on the optimization evaluation index to obtain a target charging pile;
[0031] In one embodiment, based on the optimization evaluation index of the comprehensive performance of the charging pile, the system terminal will conduct in-depth analysis and comparison in the optimization space of the candidate charging pile set. Specifically, the system terminal will extract each charging pile in the optimization space, and use the optimization evaluation index to compare and analyze the characteristic information of the extracted charging pile to obtain the corresponding performance index. In each extraction except the first extraction, the performance index of the current charging pile will be compared with the performance index of the previously saved target charging pile to determine whether to update the target charging pile. This process is to find the charging pile with the best performance in the candidate charging pile set. By comprehensively considering the performance of the charging pile in multiple aspects such as charging speed, efficiency, and stability, the system terminal can accurately find the charging pile that is currently most suitable for the user, that is, the target charging pile. In this way, not only can the available charging piles be found quickly, but also the efficiency and safety of the charging process can be ensured, and a better charging experience can be obtained.
[0032] Furthermore, the present application provides a method for obtaining a target charging pile by performing an optimization analysis in the optimization space based on the optimization evaluation index, including:
[0033] Step 1: extracting the first candidate charging pile in the optimization space;
[0034] Step 2: Based on the optimization evaluation index, traverse and analyze the first multi-dimensional feature information of the first candidate charging pile to obtain a first comprehensive performance index;
[0035] Preferably, the system terminal first randomly selects a charging pile from the optimization space as the first candidate charging pile. Subsequently, the system terminal analyzes and evaluates multiple key feature information of the first candidate charging pile one by one, including the first predicted failure probability, the first rated power, etc., based on the evaluation index of the comprehensive performance of the charging pile. In each traversal, the system terminal performs weighted calculation on key feature information such as the first predicted failure probability and the first rated power based on the optimization evaluation index to obtain the first comprehensive performance index. This index can intuitively reflect the overall performance of the first candidate charging pile. At the same time, the system terminal uses the first candidate charging pile as the target charging pile as a benchmark for subsequent optimization. In this way, the system terminal can more accurately evaluate the performance of each candidate charging pile and provide strong support for users to select the optimal charging pile.
[0036] Step 3: extracting a second candidate charging pile in the optimization space, where the second candidate charging pile is any candidate charging pile in the optimization space except the first candidate charging pile;
[0037] Step 4: performing a traversal analysis on the second multi-dimensional feature information of the second candidate charging pile based on the optimization evaluation index to obtain a second comprehensive performance index;
[0038] Preferably, in the optimization process, in addition to the first candidate charging pile, there are other candidate charging piles waiting to be evaluated in the optimization space. The system terminal will continue to extract one of these candidate charging piles as the second candidate charging pile. This second candidate charging pile is any charging pile in the optimization space except the first candidate charging pile that has been analyzed. Subsequently, the system terminal uses the same method as mentioned above to obtain the first comprehensive performance index to perform a detailed analysis of the multi-dimensional feature information of the second candidate charging pile, and obtains the second comprehensive performance index of the second candidate charging pile. This index is a quantitative evaluation of the performance of the second candidate charging pile, which is used for subsequent comparison with other candidate charging piles. In this way, the system terminal can comprehensively evaluate each candidate charging pile in the optimization space, thereby selecting the charging pile with the best performance for users to use.
[0039] Step 5: Determine whether the second comprehensive performance index is greater than the first comprehensive performance index;
[0040] Step 6: If it is greater than, use the second candidate charging pile as the target charging pile; if it is less than or equal to, use the first candidate charging pile as the target charging pile;
[0041] Step 7: Repeat steps 3 to 6 until a predetermined repetition number threshold is reached to obtain the target charging pile at that time.
[0042] Preferably, after obtaining the first comprehensive performance index and the second comprehensive performance index, the system terminal compares the second comprehensive performance index of the second candidate charging pile with the first comprehensive performance index of the first candidate charging pile. If the second comprehensive performance index is higher, it means that the performance of the second candidate charging pile is superior, so the system terminal determines the second candidate charging pile as the target charging pile. On the contrary, if the second comprehensive performance index is not higher than the first comprehensive performance index, then the performance of the first candidate charging pile is better, and the system terminal will continue to regard it as the target charging pile. In order to ensure that the optimal target charging pile is found, the system terminal will repeat the above steps three to six, that is, continue to extract the next candidate charging pile in the optimization space for analysis, and compare its comprehensive performance index. This process will continue until the predetermined repetition threshold is reached, and the final result is the target charging pile with the best performance after multiple rounds of comparison. In this way, the system terminal can comprehensively evaluate the performance of all candidate charging piles and select the target charging pile that is most suitable for users.
[0043] Furthermore, the present application provides step 2, the method further comprising:
[0044] The first multi-dimensional feature information includes a first basic parameter and a first historical fault sequence of the first candidate charging pile, wherein the first basic parameter includes a first rated power of the first candidate charging pile;
[0045] Analyzing the first historical fault time series to obtain a first predicted fault probability;
[0046] The first comprehensive performance index is obtained by weighting the first rated power and the first predicted failure probability based on the optimization evaluation index.
[0047] Optionally, in the process of obtaining the first comprehensive performance index, the system terminal first traverses the first multi-dimensional feature information of the first candidate charging pile. The multi-dimensional feature information of the first candidate charging pile includes the first basic parameter and the first historical fault sequence. The first basic parameter is the data of the first candidate charging pile itself, including the rated power of the first candidate charging pile, and this value directly reflects the basic performance of the charging pile. The first historical fault sequence is the time series of the first candidate charging pile having a fault, including a timestamp, fault information, etc. Subsequently, the system terminal extracts the first historical fault sequence to obtain the first historical fault information at the first historical moment. Then, the predetermined graph features and the probability prediction model are used to generate the first predicted fault probability in combination with the first historical fault information. Afterwards, the system terminal sets weights for the first rated power and the first predicted fault probability respectively according to the requirements of the optimization evaluation index and the actual needs. Among them, the sum of the set weights is 1, so that weighted calculation can be performed later. Then, the first rated power and the first predicted fault probability of the first candidate charging pile are multiplied by the corresponding weights to obtain the weighted rated power and fault probability. Finally, the weighted rated power and weighted failure probability are added together to obtain the first comprehensive performance index of the first candidate charging pile. This index not only takes into account the charging capacity of the charging pile, but also its operating stability, thereby ensuring that the target charging pile finally selected by the system terminal can both charge quickly and operate stably.
[0048] Furthermore, the present application provides that, analyzing the first historical fault time sequence to obtain a first predicted fault probability, the method further includes:
[0049] Extracting first historical fault information at a first historical moment in the first historical fault time sequence;
[0050] Analyze the first historical fault information to obtain a first historical fault severity level, obtain a first autocorrelation function in combination with the first historical moment, and draw a first autocorrelation graph of the first autocorrelation function;
[0051] Optionally, the system terminal extracts the first historical fault information of the first historical moment from the first historical fault sequence of the first candidate charging pile. The first historical moment is the time when the first fault of the first candidate charging pile occurs. The first historical fault information is key details about the type of fault, the time of occurrence, the duration of the fault, etc. With this information, the system terminal can further analyze the nature and severity of the fault. Based on this information, the system terminal makes a qualitative assessment of the first historical fault, comprehensively considers factors such as the impact range, duration, and difficulty of repair of the fault, and combines professional knowledge to determine the severity level of the first historical fault, which can be a graded score used to quantify the severity of the fault. Subsequently, a suitable time delay range is determined using the extracted first historical fault sequence and the corresponding historical moment information. This range is dynamically adjusted according to the characteristics of the data. Afterwards, for each selected time delay, the correlation between the first historical fault sequence and its own version after the time delay is calculated. This involves calculating the dot product of the two time series and dividing it by the product of their standard deviations to obtain a standardized autocorrelation value. Then, the calculated autocorrelation values are recorded in the order of time delay to form an autocorrelation function. The autocorrelation function measures the similarity of the fault sequence with itself under different time delays, helping the system terminal understand the timing characteristics and potential patterns of the fault. Finally, the system terminal draws the first autocorrelation graph based on the calculated first autocorrelation function. The autocorrelation graph uses the time delay as the horizontal axis and the autocorrelation coefficient as the vertical axis to intuitively show the change of the autocorrelation function with the time delay. By observing the autocorrelation graph, the periodicity, trend and potential laws of the fault sequence can be identified, providing an important basis for subsequent fault prediction. In general, this process is to extract and analyze historical fault information, combine autocorrelation functions and autocorrelation graphs, and conduct an in-depth exploration of the fault situation of the first candidate charging pile. This not only helps to understand the fault characteristics and laws of the charging pile, but also provides an important basis for subsequent fault prediction and performance evaluation.
[0052] Reading a predetermined graph feature, and acquiring first graph feature information of the first autocorrelation graph based on the predetermined graph feature;
[0053] The first predicted fault probability is obtained by analyzing the first graph feature information through a probability prediction model, wherein the probability prediction model is an intelligent model for predicting the fault probability of a charging pile trained based on the principle of a neural network;
[0054] The predetermined graph features include significance level, highest periodic peak, unit period and autocorrelation coefficient decay speed.
[0055] Optionally, the system terminal reads predetermined graph features, which are pre-set based on historical data and expert knowledge, and are used to describe and quantify specific attributes of the autocorrelation graph, including the significance level, the highest periodic peak, the unit period, and the decay rate of the autocorrelation coefficient. Subsequently, the system terminal uses these predetermined graph features to collect the first graph feature information from the first autocorrelation graph. These graph feature information can reflect the key characteristics of the first autocorrelation graph, such as the significance, periodic fluctuation, cycle duration, and the decay rate of the autocorrelation coefficient with time delay. After that, the system terminal collects the historical fault data of the charging pile, including fault data and fault probability. The collected data is then cleaned to remove outliers in the data. Then, according to the complexity of the problem and the scale of the data, the structural parameters such as the number of layers of the network and the number of neurons in each layer are determined. The preprocessed data and the corresponding fault probability are then divided into a training set, a validation set, and a test set. Further, the system terminal uses the training set to train the neural network, and adjusts the network parameters through the back propagation algorithm to minimize the error between the predicted value and the true value. During the training process, the validation set is used to monitor the performance of the model to prevent overfitting. After the training is completed, the trained model is evaluated using the test set to calculate indicators such as prediction accuracy and recall rate. The structure of the model is adjusted according to the evaluation results and iterative optimization is performed. When the evaluation results meet expectations, the probability prediction model is output. This model is an intelligent model used to predict the failure probability of charging piles. By inputting graph feature information, the probability prediction model can analyze and output the corresponding predicted failure probability. In summary, this process is to obtain the predicted failure probability of the charging pile by reading and collecting predetermined graph feature information, and then using a trained probability prediction model. This method combines graph feature analysis and machine learning technology to help improve the accuracy and reliability of fault prediction.
[0056] An optimal charging path generated by combining the first position of the first user and the target position of the target charging pile is sent to the first user, and the optimal charging path has an identifier of an optimal execution time.
[0057] In one embodiment, the system terminal combines the current location of the first user and the location of the target charging pile, and selects a path based on the road traffic network to plan the shortest path to the target charging pile for the user, that is, the optimal charging path. This optimal charging path also has an identification of the optimal execution time. The identification of this optimal execution time is obtained by the system terminal based on the analysis of historical road congestion information, which can plan the best time for the user to depart for the target charging pile to avoid the waiting time process, so as to improve the user experience. Subsequently, the system terminal sends this optimal charging path to the first user so that the user can go to the charging pile for charging quickly and efficiently. In this way, users can solve charging problems more conveniently, save time, and improve the charging experience.
[0058] Furthermore, the present application provides that the multi-data source end also includes a road traffic network end, and sends the optimal charging path generated by combining the first position of the first user and the target position of the target charging pile to the first user, and the method also includes:
[0059] Acquire road network information of the road traffic network terminal;
[0060] According to the road network information, the shortest driving path from the first position to the target position is obtained with the shortest driving distance as a constraint, and the shortest driving path is used as the optimal charging path.
[0061] Preferably, the multi-data source terminal includes a road traffic network terminal, and the system terminal obtains detailed road network information from the road traffic network terminal, including the layout, length, width, traffic conditions, speed limit, etc. of the road. This information is the basis for the subsequent calculation of the shortest driving path. Subsequently, the system terminal takes the first location where the first user is currently located as the starting point and the location of the target charging pile as the end point. After that, the path planning algorithm is used to start calculating the shortest driving path from the starting point to the end point. During the calculation process, the algorithm will evaluate the driving distance of different paths based on factors such as the road length and traffic conditions in the road network information, and generate multiple candidate paths. Then, the system terminal selects the path with the shortest driving distance from the candidate paths as the optimal solution. Finally, the system terminal sends this optimal charging path to the user so that the user can go to the charging pile for charging at the shortest distance, thereby improving the user's charging experience.
[0062] Furthermore, the present application provides an indication that the optimal charging path has an optimal execution time, and the method further includes:
[0063] Extracting a first scheduled charging time from the first scheduled charging information, and acquiring a first time tolerance of the first scheduled charging time;
[0064] Acquire historical road congestion information of the road traffic network terminal;
[0065] Optionally, the optimal charging path has an identification of the optimal execution time. The first time tolerance refers to the time fluctuation range that the user can accept to reach the charging pile before and after the scheduled charging time. The system terminal extracts the first scheduled charging time from the first scheduled charging information, which is the charging start time scheduled by the user. In order to ensure that the charging process is not affected by small changes in time, the system terminal also obtains the time tolerance related to this scheduled charging time, that is, the time fluctuation range that the user can accept. For example, when the user schedules the charging time at 5 pm, the system terminal will consider the time taken by the user to scan the code to confirm and install the charging head, and set a suitable time tolerance for the user. Within this time tolerance range, that is, the user arrives at the target charging pile between 4:58 and 5 pm, which can ensure that the user can start charging without waiting, thereby optimizing the user's charging experience. Subsequently, in order to ensure that the user can reach the charging pile on time, the system terminal further obtains historical road congestion information from the road traffic network. This information can help understand the traffic conditions on the road, especially the congestion before and after the user's scheduled charging time. By combining this information, a charging route can be planned for the user that meets time requirements and avoids congested roads as much as possible, thereby improving the user's charging efficiency and charging experience.
[0066] Combining the historical road congestion information to obtain a first arrival time under a first execution time;
[0067] When the first arrival time is within the first time tolerance, the first execution time is used as the optimal execution time.
[0068] Optionally, the first execution time refers to the departure time of the user to the target charging pile. The system terminal combines historical road congestion information and the planned driving route to simulate the expected driving time after the user departs at the first execution time. The simulation process takes into account factors such as changes in traffic flow in different time periods and possible traffic events. The predicted driving time is added to the first execution time, and the system calculates the time when the user is expected to arrive at the charging pile, that is, the first arrival time. Subsequently, the system terminal compares the calculated first arrival time with the first time tolerance of the user's scheduled charging time. If the first arrival time falls within the first time tolerance range, it means that the user departs according to the first execution time, can arrive at the charging pile at the right time and start charging without waiting. Therefore, when the first arrival time is within the first time tolerance, the system will determine this departure time, that is, the first execution time, as the optimal execution time. In this way, the user can set off according to this optimal execution time to ensure that the charging process is completed on time and smoothly, thereby improving the charging experience.
[0069] Furthermore, the present application provides a charging experience for the first user, and the method further includes:
[0070] Obtaining a first experience distribution weight of the first user;
[0071] An experience evaluation is performed on the first charging pile scheduling decision based on the first experience allocation weight to obtain a first experience index, wherein the first charging pile scheduling decision refers to the first user driving to the target charging pile for charging based on the optimal charging path at the optimal execution time, and the first experience index represents the charging experience of the first user.
[0072] Preferably, the system terminal obtains the first user's first experience allocation weight from the first user's first database. This weight represents the user's emphasis on different charging experience factors, such as charging speed, charging pile location, waiting time, etc. The experience allocation weight of each user is different because everyone's needs and preferences are different. Subsequently, the system terminal analyzes the user's driving route, expected arrival time, and possible waiting time under the first charging pile scheduling decision. This first charging pile scheduling decision refers to the process in which the user goes to the target charging pile for charging based on the optimal charging path at the optimal execution time. After that, the system terminal scores each link in the scheduling decision according to the first user's experience allocation weight. For example, if the user has a high requirement for charging speed and the charging pile can provide fast charging service, the score for charging speed will be higher. Similarly, if the user has specific requirements for the location of the charging pile and the location of the selected charging pile meets the user's expectations, the score for location will also increase accordingly. Then, the scores of each link are weighted according to the corresponding weights to obtain a weighted total score. This weighted total score is the first experience index, which comprehensively reflects the user's overall experience during the charging process. The higher the index, the better the user's charging experience; otherwise, it means the experience needs to be improved. This index is an important basis for the system to evaluate the quality of charging services and optimize scheduling decisions.
[0073] Furthermore, the present application provides a charging experience for the first user, and the method further includes:
[0074] The first experience allocation weight is stored in a first database of the first user, where the first database is charging record data of the first user constructed based on a predetermined access control policy.
[0075] Optionally, the system terminal determines the first experience distribution weight according to the personal preferences and historical behavior data of the first user, and these weights reflect the user's emphasis on different charging experience factors. In order to ensure the security and privacy of user data, these weights are stored in the charging record database of the first user constructed based on the predetermined access control policy, that is, the first database. This first database is constructed in accordance with strict data encryption standards to ensure the security of user data. Through data encryption, even if the database is accessed without authorization, attackers cannot easily obtain sensitive information such as the user's experience distribution weight. At the same time, the predetermined access control policy also guarantees the integrity and availability of user data. Only authorized personnel or systems can access and modify the data in the first database, which ensures the accuracy and reliability of user data. Through the predetermined access control policy, the system terminal not only protects the user's data security, but also improves the user's charging experience. Users can safely share their preferences and experience feelings, and the system terminal can optimize the charging pile scheduling decision based on these feedbacks to provide users with more personalized, efficient and convenient charging services.
[0076] Furthermore, the application provides that the multi-data source end also includes a power grid end, and the method further includes:
[0077] Dynamically acquiring the real-time grid charge at the grid end;
[0078] When the real-time grid charge reaches a predetermined charge threshold, a predetermined charging pile locking strategy is called to lock the charging pile.
[0079] Preferably, the system terminal dynamically obtains the real-time charge data of the power grid by monitoring the charge situation at the power grid end in real time. When the real-time charge of the power grid reaches a predetermined charge threshold, it means that the load of the power grid is already high, and some measures need to be taken to avoid power grid overload. At this time, the system terminal will perform corresponding operations according to the preset charging pile locking strategy. The predetermined charging pile locking strategy means that when the power grid charge reaches a certain threshold, the system terminal will lock some charging piles so that they cannot be booked and used by users. The purpose of this is to reduce the load on the power grid and ensure the stable operation of the power grid. By implementing this strategy, the system terminal can effectively control the use of charging piles, avoid the situation where too many charging piles work at the same time during the peak period of power grid charge, thereby reducing the risk of power grid overload. At the same time, this can also remind users to pay attention to the load of the power grid, arrange charging time reasonably, and improve the utilization efficiency of the power grid.
[0080] In summary, the embodiments of the present application have at least the following technical effects:
[0081] The embodiment of the present application obtains real-time operation status and scheduled charging sequence information from the charging station end and the user end. Subsequently, a candidate charging pile set is selected according to the real-time data and the predetermined state constraints. The candidate charging pile set is optimized and analyzed, and the comprehensive performance of the charging pile is used as an evaluation index to obtain the target charging pile. At the same time, by analyzing the historical fault information of the charging pile, combining the predetermined graph features and the probability prediction model, the failure probability of the charging pile is predicted. After that, the optimal charging path is generated in combination with the information of the road traffic network end, with the shortest driving distance as the constraint. Then, weights are assigned according to the user's experience, the experience index of the charging pile scheduling decision is evaluated, and the relevant information is stored in the user's charging record database. Finally, when the real-time charge on the grid end reaches the predetermined threshold, the predetermined charging pile locking strategy is called to lock the charging pile. The method comprehensively considers the information of multiple data sources, realizes the intelligent scheduling of charging piles and the effective management of grid loads, thereby improving the charging efficiency and user experience. These technical effects jointly solve the technical problem that the user's waiting time is too long due to low charging efficiency, resulting in poor user experience, and achieve the effect of improving the use efficiency of charging piles, reducing user waiting time, and improving the user's charging experience.
[0082] Embodiment 2
[0083] Based on the same inventive concept as the charging pile scheduling method based on dynamic data fusion in the above-mentioned embodiment, Figure 2 As shown, the present application provides a charging pile scheduling system based on dynamic data fusion, the system comprising:
[0084] Operation status acquisition module 1: the operation status acquisition module 1 is used to dynamically acquire the real-time operation status of the charging station, and the real-time operation status includes multiple groups of real-time usage information of multiple charging piles in the charging station;
[0085] Charging sequence acquisition module 2: the charging sequence acquisition module 2 is used to dynamically acquire the scheduled charging sequence of the user end, and the scheduled charging sequence includes the first scheduled charging information of the first user;
[0086] Information matching module 3: the information matching module 3 is used to extract the first real-time usage information from the multiple sets of real-time usage information when the real-time operating state meets the first predetermined state constraint, and match the first charging pile corresponding to the first real-time usage information from the multiple charging piles;
[0087] Charging pile adding module 4: the charging pile adding module 4 is used to add the first charging pile to the candidate charging pile set when the first real-time usage information meets the second predetermined state constraint;
[0088] Optimization setting module 5: the optimization setting module 5 is used to use the comprehensive performance of the charging pile as an optimization evaluation index and the candidate charging pile set as an optimization space;
[0089] Optimization analysis module 6: the optimization analysis module 6 is used to perform optimization analysis in the optimization space based on the optimization evaluation index to obtain a target charging pile;
[0090] Optimal charging path generation module 7: The optimal charging path generation module 7 is used to send the optimal charging path generated by combining the first position of the first user and the target position of the target charging pile to the first user, and the optimal charging path has an optimal execution time identifier.
[0091] Furthermore, the optimization analysis module 6 is also used to perform the following method:
[0092] Step 1: extracting the first candidate charging pile in the optimization space;
[0093] Step 2: Based on the optimization evaluation index, traverse and analyze the first multi-dimensional feature information of the first candidate charging pile to obtain a first comprehensive performance index;
[0094] Step 3: extracting a second candidate charging pile in the optimization space, where the second candidate charging pile is any candidate charging pile in the optimization space except the first candidate charging pile;
[0095] Step 4: performing a traversal analysis on the second multi-dimensional feature information of the second candidate charging pile based on the optimization evaluation index to obtain a second comprehensive performance index;
[0096] Step 5: Determine whether the second comprehensive performance index is greater than the first comprehensive performance index;
[0097] Step 6: If it is greater than, use the second candidate charging pile as the target charging pile; if it is less than or equal to, use the first candidate charging pile as the target charging pile;
[0098] Step 7: Repeat steps 3 to 6 until a predetermined repetition number threshold is reached to obtain the target charging pile at that time.
[0099] Furthermore, the optimization analysis module 6 is also used to perform the following method:
[0100] The first multi-dimensional feature information includes a first basic parameter and a first historical fault sequence of the first candidate charging pile, wherein the first basic parameter includes a first rated power of the first candidate charging pile;
[0101] Analyzing the first historical fault time series to obtain a first predicted fault probability;
[0102] The first comprehensive performance index is obtained by weighting the first rated power and the first predicted failure probability based on the optimization evaluation index.
[0103] Furthermore, the optimization analysis module 6 is also used to perform the following method:
[0104] Extracting first historical fault information at a first historical moment in the first historical fault time sequence;
[0105] Analyze the first historical fault information to obtain a first historical fault severity level, obtain a first autocorrelation function in combination with the first historical moment, and draw a first autocorrelation graph of the first autocorrelation function;
[0106] Reading a predetermined graph feature, and acquiring first graph feature information of the first autocorrelation graph based on the predetermined graph feature;
[0107] The first predicted fault probability is obtained by analyzing the first graph feature information through a probability prediction model, wherein the probability prediction model is an intelligent model for predicting the fault probability of a charging pile trained based on the principle of a neural network;
[0108] The predetermined graph features include significance level, highest periodic peak, unit period and autocorrelation coefficient decay speed.
[0109] Furthermore, the optimal charging path generation module 7 is also used to execute the following method:
[0110] Acquire road network information of the road traffic network terminal;
[0111] According to the road network information, the shortest driving path from the first position to the target position is obtained with the shortest driving distance as a constraint, and the shortest driving path is used as the optimal charging path.
[0112] Furthermore, the optimal charging path generation module 7 is also used to execute the following method:
[0113] Extracting a first scheduled charging time from the first scheduled charging information, and acquiring a first time tolerance of the first scheduled charging time;
[0114] Acquire historical road congestion information of the road traffic network terminal;
[0115] Combining the historical road congestion information to obtain a first arrival time under a first execution time;
[0116] When the first arrival time is within the first time tolerance, the first execution time is used as the optimal execution time.
[0117] Furthermore, the optimal charging path generation module 7 is also used to execute the following method:
[0118] Obtaining a first experience distribution weight of the first user;
[0119] An experience evaluation is performed on the first charging pile scheduling decision based on the first experience allocation weight to obtain a first experience index, wherein the first charging pile scheduling decision refers to the first user driving to the target charging pile for charging based on the optimal charging path at the optimal execution time, and the first experience index represents the charging experience of the first user.
[0120] Furthermore, the optimal charging path generation module 7 is also used to execute the following method:
[0121] The first experience sense allocation weight is stored in a first database of the first user, where the first database is the charging record data of the first user constructed based on a predetermined access control policy.
[0122] Furthermore, the optimal charging path generation module 7 is also used to execute the following method:
[0123] Dynamically acquiring the real-time grid charge at the grid end;
[0124] When the real-time grid charge reaches a predetermined charge threshold, a predetermined charging pile locking strategy is called to lock the charging pile.
[0125] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0126] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0127] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A charging pile scheduling method based on dynamic data fusion, characterized in that: The charging pile scheduling method is applied to a charging pile scheduling system, the charging pile scheduling system is communicatively connected with multiple data source terminals, the multiple data source terminals include a charging station terminal and a user terminal, and the charging pile scheduling includes: Dynamically obtain the real-time operating status of the charging station, wherein the real-time operating status includes multiple groups of real-time usage information of multiple charging piles in the charging station; Dynamically acquiring a scheduled charging sequence of the user terminal, wherein the scheduled charging sequence includes first scheduled charging information of a first user; When the real-time operation state meets the first predetermined state constraint, extracting the first real-time usage information from the multiple sets of real-time usage information, and matching the first charging pile corresponding to the first real-time usage information among the multiple charging piles, wherein the first predetermined state constraint is used to limit whether at least one or more charging piles in the real-time operation state are in an idle state; When the first real-time usage information meets a second predetermined state constraint, adding the first charging pile to a candidate charging pile set, wherein the second predetermined state constraint is used to limit the usage of the current charging pile of the first real-time usage information; Taking the comprehensive performance of the charging pile as the optimization evaluation index, and taking the set of candidate charging piles as the optimization space; Performing an optimization analysis in the optimization space based on the optimization evaluation index to obtain a target charging pile; Sending an optimal charging path generated by combining the first position of the first user and the target position of the target charging pile to the first user, wherein the optimal charging path has an identifier of an optimal execution time; Based on the optimization evaluation index, an optimization analysis is performed in the optimization space to obtain a target charging pile, including: Step 1: extracting the first candidate charging pile in the optimization space; Step 2: Based on the optimization evaluation index, traverse and analyze the first multi-dimensional feature information of the first candidate charging pile to obtain a first comprehensive performance index; Step 3: extracting a second candidate charging pile in the optimization space, where the second candidate charging pile is any candidate charging pile in the optimization space except the first candidate charging pile; Step 4: performing a traversal analysis on the second multi-dimensional feature information of the second candidate charging pile based on the optimization evaluation index to obtain a second comprehensive performance index; Step 5: Determine whether the second comprehensive performance index is greater than the first comprehensive performance index; Step 6: If it is greater than, use the second candidate charging pile as the target charging pile; if it is less than or equal to, use the first candidate charging pile as the target charging pile; Step 7: Repeat steps 3 to 6 until a predetermined repetition number threshold is reached to obtain the target charging pile at that time; Step 2 includes: The first multi-dimensional feature information includes a first basic parameter and a first historical fault sequence of the first candidate charging pile, wherein the first basic parameter includes a first rated power of the first candidate charging pile; Analyzing the first historical fault time series to obtain a first predicted fault probability; The first comprehensive performance index is obtained by weighting the first rated power and the first predicted failure probability based on the optimization evaluation index.
2. The charging pile scheduling method according to claim 1, characterized in that: Analyzing the first historical fault time sequence to obtain a first predicted fault probability includes: Extracting first historical fault information at a first historical moment in the first historical fault time sequence; Analyze the first historical fault information to obtain a first historical fault severity level, obtain a first autocorrelation function in combination with the first historical moment, and draw a first autocorrelation graph of the first autocorrelation function; Reading a predetermined graph feature, and acquiring first graph feature information of the first autocorrelation graph based on the predetermined graph feature; The first predicted fault probability is obtained by analyzing the first graph feature information through a probability prediction model, wherein the probability prediction model is an intelligent model for predicting the fault probability of a charging pile trained based on the principle of a neural network; The predetermined graph features include significance level, highest periodic peak, unit period and autocorrelation coefficient decay speed.
3. The charging pile scheduling method according to claim 1, characterized in that: The multi-data source end also includes a road traffic network end, which sends an optimal charging path generated by combining the first position of the first user and the target position of the target charging pile to the first user, including: Acquire road network information of the road traffic network terminal; According to the road network information, the shortest driving path from the first position to the target position is obtained with the shortest driving distance as a constraint, and the shortest driving path is used as the optimal charging path.
4. The charging pile scheduling method according to claim 3, characterized in that: The optimal charging path has an identification of an optimal execution time, including: Extracting a first scheduled charging time from the first scheduled charging information, and acquiring a first time tolerance of the first scheduled charging time; Acquire historical road congestion information of the road traffic network terminal; Combining the historical road congestion information to obtain a first arrival time under a first execution time; When the first arrival time is within the first time tolerance, the first execution time is used as the optimal execution time.
5. The charging pile scheduling method according to claim 1, characterized in that: The method also includes: Obtaining a first experience distribution weight of the first user; An experience evaluation is performed on the first charging pile scheduling decision based on the first experience allocation weight to obtain a first experience index, wherein the first charging pile scheduling decision refers to the first user driving to the target charging pile for charging based on the optimal charging path at the optimal execution time, and the first experience index represents the charging experience of the first user.
6. The charging pile scheduling method according to claim 5, characterized in that: The first experience allocation weight is stored in a first database of the first user, where the first database is charging record data of the first user constructed based on a predetermined access control policy.
7. The charging pile scheduling method according to claim 1, characterized in that: The multiple data source ends also include a power grid end, and the method further includes: Dynamically acquiring the real-time grid charge at the grid end; When the real-time grid charge reaches a predetermined charge threshold, a predetermined charging pile locking strategy is called to lock the charging pile.
8. Charging pile scheduling system based on dynamic data fusion, characterized in that: The steps for implementing the charging pile scheduling method based on dynamic data fusion as described in any one of claims 1 to 7 include: Operation status acquisition module: dynamically acquires the real-time operation status of the charging station, wherein the real-time operation status includes multiple sets of real-time usage information of multiple charging piles in the charging station; Charging sequence acquisition module: dynamically acquires the scheduled charging sequence of the user terminal, wherein the scheduled charging sequence includes the first scheduled charging information of the first user; An information matching module: when the real-time operation state meets the first predetermined state constraint, extract the first real-time usage information from the multiple sets of real-time usage information, and match the first charging pile corresponding to the first real-time usage information from the multiple charging piles, wherein the first predetermined state constraint is used to limit whether at least one or more charging piles in the real-time operation state are in an idle state; Charging pile adding module: when the first real-time usage information meets the second predetermined state constraint, the first charging pile is added to the candidate charging pile set, and the second predetermined state constraint is used to limit the usage of the current charging pile of the first real-time usage information; Optimization setting module: taking the comprehensive performance of the charging pile as the optimization evaluation index and taking the candidate charging pile set as the optimization space; Optimization analysis module: performs optimization analysis in the optimization space based on the optimization evaluation index to obtain a target charging pile; The optimal charging path generation module sends the optimal charging path generated by combining the first position of the first user and the target position of the target charging pile to the first user, wherein the optimal charging path has an identifier of the optimal execution time.
