Charging pile income estimation and evaluation method, system and device and medium

Abstract

The invention provides a charging pile income estimation and evaluation method, system and device and a medium, and belongs to the technical field of charging facility management. The method comprises the following steps: collecting historical data of a user charging behavior through a charging pile terminal, and constructing a user feature vector; based on a clustering algorithm, dividing the users into different user groups by using the user feature vectors, and determining a user set of each type and a clustering center; introducing a time period convenience factor, and establishing a time period selection probability model considering user price sensitivity by using an improved Logit model to calculate a time period selection probability; constructing a dynamic income prediction function containing an electricity price and user behavior nonlinear coupling item, and calculating a single-day total income based on a user feature vector, each type of user set and a time period selection probability; and based on a gradient descent algorithm, an optimal electricity price strategy for maximizing the total income is solved according to the dynamic income prediction function. According to the invention, through real-time data updating and model iteration, the income prediction precision is significantly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of charging facility management, and more specifically relates to a charging pile revenue estimation and evaluation method, system, equipment and medium. Background Art

[0002] With the global energy transition and growing environmental awareness, the new energy vehicle industry is experiencing unprecedented growth opportunities. As a result, the construction of charging pile networks, as supporting infrastructure, has also rapidly expanded. However, in the operation and management of charging piles, scientifically and accurately estimating their revenue has become a major challenge facing the industry.

[0003] Traditional charging pile revenue forecasting methods primarily rely on historical data statistics and simple linear regression models, which have significant limitations. First, they fail to fully account for the diverse user groups. For example, users of different ages, occupational backgrounds, and travel habits have varying demands for and willingness to pay for charging services. This one-size-fits-all forecasting approach can easily lead to significant deviations in revenue estimates, impacting the accuracy of operational decisions.

[0004] Secondly, static electricity pricing models ignore the dynamic response of users to charging time. In practice, users flexibly adjust charging times based on factors such as electricity price fluctuations and their travel schedules. Traditional methods struggle to capture these dynamic changes, making revenue forecasts out of sync with reality.

[0005] Furthermore, existing technologies lack quantitative modeling of user behavior parameters (such as charging frequency, duration preferences, etc.), and are unable to deeply explore the patterns and trends behind user behavior, limiting the depth and breadth of revenue forecasts. Summary of the Invention

[0006] In response to the above problems, the purpose of the present invention is to provide a charging pile revenue estimation and evaluation method, system, device and medium, which significantly improves the revenue prediction accuracy through real-time data update and model iteration, and provides data support for charging pile layout planning, pricing strategy adjustment and user service optimization.

[0007] To achieve the above-mentioned purpose, the present invention is implemented through the following technical solutions: In a first aspect, an embodiment of the present application provides a charging pile revenue estimation and evaluation method, comprising: Collect historical data of user charging behavior through charging pile terminals and construct user feature vectors; Based on the K-means clustering algorithm, users are divided into different user groups using user feature vectors, and the user set and cluster center of each type are determined; The time period convenience factor is introduced, and the improved Logit model is used to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability; A dynamic revenue prediction function is constructed that includes nonlinear coupling terms between electricity prices and user behavior, and the total daily revenue is calculated based on user feature vectors, user sets of each type, and time period selection probabilities. Based on the gradient descent algorithm, the optimal electricity price strategy that maximizes total revenue is solved according to the dynamic revenue prediction function.

[0008] In an optional embodiment, collecting historical data of user charging behavior through the charging pile terminal to construct a user feature vector includes: Based on the Internet of Things, historical data on user charging behavior is collected through charging pile terminals; The number of users, average daily charging times, average charging time, and price sensitivity coefficient are obtained from the historical data of user charging behavior to construct the user feature vector U k =[N k ,C k ,T k ,β k ]; Among them, U k is the feature vector of the k-th user, N k is the total number of users in the kth category, C k is the average daily charging times for the k-th user, T k is the average single charging time for the k-th user, β k is the price sensitivity coefficient of the k-th category of users.

[0009] In an optional embodiment, the K-means clustering algorithm is used to divide users into different user groups using user feature vectors, and to determine the user set and cluster center of each type, including: The K-means clustering algorithm is used to divide users into K categories with the following formula as the optimization objective to reduce the impact of group differences on revenue forecasting:

[0010] in, is the k-th user set, is the feature vector cluster center of the k-th user.

[0011] In an optional embodiment, the time period convenience factor is introduced and a time period selection probability model that takes into account user price sensitivity is established using an improved Logit model to calculate the time period selection probability, including: Introducing time period convenience factor ,Using the improved Logit model, based on the divided user groups and price sensitivity coefficients, a time period selection probability model considering user price sensitivity is established; The time period selection probability model includes:

[0012] in, The probability of selecting time period i for the k-th user, is the electricity price in period i, is the convenience factor of period i, The time period preference coefficient of the k-th user;

[0013] In the above formula, is the idle rate of charging piles in period i, is the normalized charging times of period i, is the charging pile idle rate coefficient determined based on historical data, is the charging times coefficient determined based on historical data.

[0014] In an optional embodiment, the dynamic revenue prediction function includes:

[0015] in, is the total daily income, is the average charging power at different times of the day, is the charging selection probability of the kth user in time period i.

[0016] In an optional embodiment, the method of solving the optimal electricity price strategy that maximizes total revenue based on the dynamic revenue prediction function based on the gradient descent algorithm includes: Based on the gradient descent algorithm, the dynamic profit prediction function is optimized to calculate the electricity price in each period. Find the partial derivative, and iteratively calculate the partial derivative expression until the partial derivative approaches zero, and solve the optimal electricity price strategy that maximizes the total revenue ; The partial derivative expression includes: .

[0017] In an optional embodiment, the historical data of the user's charging behavior includes: user type, charging period, duration, power, and frequency.

[0018] In a second aspect, the embodiment of the present application further provides a charging pile revenue estimation and evaluation system, including: The data collection module is used to collect historical data of user charging behavior through the charging pile terminal and construct user feature vectors; The user group segmentation module is used to divide users into different user groups based on the K-means clustering algorithm and user feature vectors, and to determine the user set and cluster center of each category; The selection probability model building module is used to introduce the time period convenience factor and use the improved Logit model to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability; Dynamic revenue prediction module, which is used to construct a dynamic revenue prediction function that includes nonlinear coupling terms between electricity prices and user behavior, and calculate the total daily revenue based on user feature vectors, each type of user set, and time period selection probability; The strategy optimization module is used to solve the optimal electricity price strategy that maximizes total revenue based on the gradient descent algorithm and the dynamic revenue prediction function.

[0019] In a third aspect, an embodiment of the present application further provides an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the charging pile revenue estimation and evaluation method as described in any one of the above items are implemented.

[0020] In a fourth aspect, an embodiment of the present application further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the charging pile revenue estimation and evaluation method as described in any one of the above items are implemented.

[0021] It can be seen from the above technical solutions that the present invention has the following advantages: The charging pile revenue estimation and evaluation method provided in this application collects user charging behavior data (including charging time period, duration, power demand, usage frequency, etc.) and combines it with the charging pile device status, electricity pricing strategy, and external environmental factors to construct a multi-dimensional user behavior characteristic model. Machine learning algorithms are used to analyze the correlation between user behavior and revenue, dynamically predict the short-term and long-term revenue of charging piles, and generate revenue optimization recommendations. This method significantly improves revenue prediction accuracy through real-time data updates and model iteration, providing data support for charging pile layout planning, pricing strategy adjustments, and user service optimization. This method features intelligent prediction and strong adaptability, and can be widely applied in the field of charging pile operation and management.

[0022] This application constructs feature vectors by collecting historical data on user charging behavior, and uses the K-means clustering algorithm to divide user groups, determine the user sets and cluster centers of each type, and can accurately segment users, reduce the impact of group differences on revenue forecasts, and provide a solid foundation for subsequent analysis.

[0023] This application introduces a time period convenience factor and uses an improved Logit model to establish a time period selection probability model that takes into account user price sensitivity. This can accurately calculate the time period selection probability, better reflect the actual situation of users choosing to charge at different time periods, and make the revenue forecast more realistic.

[0024] This application constructs a dynamic revenue prediction function that includes nonlinear coupling terms of electricity prices and user behavior. It calculates the total daily revenue based on user feature vectors, each type of user set, and time period selection probability. It can dynamically reflect the impact of changes in electricity prices and user behavior on revenue, and improve the accuracy of revenue prediction.

[0025] This application is based on the gradient descent algorithm and solves the optimal electricity price strategy that maximizes total revenue according to the dynamic revenue prediction function, providing a scientific and reasonable pricing basis for charging pile operators, which helps to increase revenue.

[0026] The historical data on user charging behavior collected in this application covers various information such as user type, charging period, duration, power, frequency, etc. Comprehensive analysis using these data can deeply explore the patterns of user charging behavior, provide all-round and multi-angle decision support for charging pile operations, and improve the level of operation management. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for the description. 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 paying any creative work.

[0028] Figure 1 A flow chart of the charging pile revenue estimation and evaluation method provided in this application.

[0029] Figure 2 This is a schematic diagram of the structure of the charging pile revenue estimation and evaluation system provided in this application.

[0030] Figure 3 This is a schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0031] The specific steps of the charging pile revenue estimation and evaluation method will be described in detail below, and various embodiments of the present disclosure will be described more fully. The present disclosure can have various embodiments, and adjustments and changes can be made therein. However, it should be understood that there is no intention to limit the various embodiments of the present disclosure to the specific embodiments disclosed herein, but rather that the present disclosure should be understood to cover all adjustments, equivalents, and / or alternatives that fall within the spirit and scope of the various embodiments of the present disclosure.

[0032] Hereinafter, the terms "include" or "may include" as used in various embodiments of the present disclosure indicate the presence of disclosed functions, operations, or elements, and do not limit the addition of one or more functions, operations, or elements. In addition, as used in various embodiments of the present disclosure, the terms "include," "have," and their cognates are intended only to indicate specific features, numbers, steps, operations, elements, components, or combinations of the foregoing, and should not be understood as excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing, or the possibility of adding one or more features, numbers, steps, operations, elements, components, or combinations of the foregoing.

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0034] See also Figure 1 The figure is a flowchart of a method for estimating and evaluating the benefits of a charging pile in a specific embodiment, the method comprising: S1: Collect historical data of user charging behavior through charging pile terminals and construct user feature vectors.

[0035] In a specific embodiment, this step obtains historical data from the charging pile terminal, including user type, charging period, charging duration, and power, and constructs a user feature vector.

[0036] Specifically, first, historical data on user charging behavior is collected through charging pile terminals based on the Internet of Things; among them, historical data on user charging behavior includes: user type, charging period, duration, power, and frequency.

[0037] Then, the number of users, average daily charging times, average charging time, and price sensitivity coefficient are obtained from the historical data of user charging behavior to construct the user feature vector U k =[N k ,C k ,T k ,β k ]; Among them, U k is the feature vector of the k-th user, N k is the total number of users in the kth category, C k is the average daily charging times for the k-th user, T k is the average charging time of the k-th user, β k is the price sensitivity coefficient of the k-th category of users.

[0038] For example, historical data on user charging behavior is obtained through the charging pile terminal, and the following key behavior parameters are extracted: User basic attributes: user type (such as private car, taxi, etc.).

[0039] Time dimension features: charging period (accurate to hours, recorded as period i), single charging time (recorded as t i , unit: hours), and further calculate the average single charging time T for the k-th user k .

[0040] Frequency and scale characteristics: Charging frequency within the statistical period (average number of charging times per day, recorded as C k ), user scale (the total number of users of a certain type, recorded as N k ).

[0041] Price sensitivity characteristics: Quantify the price sensitivity coefficient (denoted as β) by the user's response to the historical electricity price adjustment k ).

[0042] Integrate the above parameters into the user feature vector U k =[N k ,C k ,T k ,β k ].

[0043] It should be noted that the quantification process of the price sensitivity coefficient includes: First, select the historical electricity price adjustment records for the past n months (for example, 6 months) (including the complete period before and after the price change). Then, classify users according to their electricity usage characteristics (such as online ride-hailing drivers, private car owners, low-frequency users, etc.).

[0044] Then, for each type of user, we first calculate the elasticity coefficient Ed of a single price adjustment. If the same user group experiences multiple price adjustments, we aggregate the elasticity coefficients Ed across multiple rounds to calculate the average elasticity coefficient. As the price sensitivity coefficient of this type of user.

[0045] For example, a charging station raises the electricity price from 1.5 yuan / kWh to 1.65 yuan / kWh (+10%) during peak hours (18:00-20:00), and the charging volume during the same period drops from 1,000 kWh to 920 kWh (-8%).

[0046] Ed = −8% / 10% = −0.8 If the same user group experiences multiple price adjustments, the average elasticity is calculated:

[0047] For example, if this user group experiences three price adjustments, with elasticity coefficients of -0.8, -1.2, and -1.0 respectively, then:

[0048] That is, the price sensitivity coefficient of this type of user is -1.0.

[0049] This step converts unstructured charging data into computable quantitative features, providing a basis for user group segmentation and behavior modeling, and addressing the defect of traditional methods in "lack of quantification of user behavior parameters."

[0050] S2: Based on the K-means clustering algorithm, users are divided into different user groups using user feature vectors, and the user set and cluster center of each type are determined.

[0051] In a specific implementation, based on the K-means clustering algorithm, user feature vectors are used to divide users into different groups according to multi-dimensional features such as charging frequency, duration, and user scale, and the cluster center of each group is determined.

[0052] Specifically, the K-means clustering algorithm is used to divide users into K categories with the following formula as the optimization objective to reduce the impact of group differences on revenue prediction:

[0053] in, is the k-th user set, is the feature vector cluster center of the k-th user.

[0054] This step can eliminate the interference of user group differences on revenue forecasts. For example, it can distinguish between high-frequency, short-duration taxi users and low-frequency, long-duration private car users, provide support for refined modeling, and solve the problem of traditional methods "not considering user group differences."

[0055] S3: Introduce the time period convenience factor and use the improved Logit model to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability.

[0056] In a specific implementation, this step introduces a time period convenience factor, utilizes an improved Logit model, and establishes a time period selection probability model that takes into account user price sensitivity based on the divided user groups and price sensitivity coefficients to calculate the time period selection probability.

[0057] For example, introduce the time period convenience factor ,Using the improved Logit model, based on the divided user groups and price sensitivity coefficients, a time period selection probability model considering user price sensitivity is established; The time period selection probability model includes:

[0058] in, The probability of selecting time period i for the k-th user, is the electricity price in period i, is the convenience factor of period i, The time period preference coefficient of the k-th user.

[0059] in,

[0060] In the above formula, is the idle rate of charging piles in period i, is the normalized charging times of period i, The idle rate coefficient of the charging pile is determined manually based on historical data. It is a charging frequency coefficient determined manually based on historical data.

[0061] This step can quantify the dynamic response of users to electricity prices at different time periods. For example, price-sensitive users are more likely to charge during off-peak hours, thus resolving the defect of the traditional static electricity pricing model that "it cannot reflect the dynamic nature of user time selection."

[0062] S4: Construct a dynamic revenue prediction function that includes nonlinear coupling terms between electricity prices and user behavior, and calculate the total daily revenue based on user feature vectors, each type of user set, and time period selection probability.

[0063] In a specific implementation, a dynamic revenue prediction function including a nonlinear coupling term between electricity price and user behavior is constructed, and the total daily revenue is calculated based on the number of users, charging frequency, average single charging time, power and time period selection probability.

[0064] The dynamic profit prediction function includes:

[0065] in, is the total daily income, is the average charging power at different times of the day, is the charging selection probability of the kth user in time period i.

[0066] This step dynamically links user behavior with electricity pricing strategies, such as predicting the revenue contribution of different groups under peak and valley electricity prices, providing data-driven revenue estimates for charging pile operations, and effectively improving prediction accuracy.

[0067] S5: Based on the gradient descent algorithm, the optimal electricity price strategy that maximizes the total revenue is solved according to the dynamic revenue prediction function.

[0068] In a specific implementation, based on the gradient descent algorithm, the dynamic revenue prediction function is used as the optimization target to calculate the electricity price for each period. Find the partial derivative, and iteratively calculate the partial derivative expression until the partial derivative approaches zero, and solve the optimal electricity price strategy that maximizes the total revenue ; The partial derivative expression includes: .

[0069] This step optimizes the electricity pricing strategy. Specifically, the gradient descent method is used to derive the dynamic revenue prediction function, which includes a nonlinear coupling term between electricity price and user behavior. By iteratively calculating the partial derivative expression until it approaches zero, combined with a user time-slot selection probability model that considers price sensitivity and time-slot convenience factors, the optimal electricity pricing strategy that maximizes total revenue is solved, achieving dynamic optimization of electricity prices to increase charging pile revenue.

[0070] In this embodiment, full-dimensional data on user charging behavior is collected through the Internet of Things, and a feature vector containing key indicators such as price sensitivity coefficient and charging frequency is constructed. The K-means clustering algorithm is then used to segment user groups. This quantitative modeling breaks through the limitations of the traditional method's homogeneity assumption and can accurately identify differentiated user groups such as those with price sensitivity and time preference. The cluster center parameters objectively reflect the group behavior pattern, enabling the revenue prediction model to perceive user heterogeneity, significantly improving prediction accuracy and decision-making targeting. For example, when formulating targeted discount strategies for high-frequency, low-price-sensitive users, changes in their demand elasticity can be accurately predicted to avoid revenue losses caused by strategic errors.

[0071] In this example, a time-of-day convenience factor and an improved Logit model were innovatively introduced to construct a dynamic choice probability model that couples price sensitivity with spatiotemporal preferences. This model uses nonlinear functions to quantitatively analyze user charging decisions under the influence of multiple factors, such as electricity price fluctuations and geographic convenience, transcending the assumptions of static electricity pricing models. For example, during peak electricity price periods, the model can predict the probability distribution of users delaying charging due to insufficient convenience, making revenue forecasts sensitive to both time and space. This dynamic response mechanism provides a scientific basis for optimizing time-of-use electricity pricing strategies, effectively balancing grid load and user satisfaction.

[0072] In this embodiment, the established dynamic revenue prediction function innovatively incorporates factors such as electricity pricing strategy, user behavior parameters, and device power into a unified analysis framework, specifically introducing a quadratic coupling term between electricity price and user behavior. This model can systematically quantify the combined impact of electricity price adjustments on user charging frequency, time period selection, and single charge volume, overcoming the drawback of traditional linear models that ignore behavioral feedback. The optimal electricity price solution model based on the gradient descent algorithm achieves a closed-loop decision-making process from revenue prediction to strategy optimization, supports real-time data updates and rolling optimization, and enables pricing strategies to have adaptive evolutionary capabilities, significantly improving decision-making efficiency and revenue levels.

[0073] In this embodiment, a four-dimensional feature system encompassing user behavior, device status, and the external environment is constructed to achieve collaborative management of "people-charging piles-network-policy." By quantitatively analyzing complex relationships such as the correlation between user charging frequency and device utilization, and the impact of ambient temperature on charging efficiency, this system provides comprehensive decision support for charging pile network planning, equipment operation and maintenance scheduling, and value-added service design. This systematic approach transcends the limitations of optimizing a single factor, enabling operators to achieve global optimization within a multi-dimensional decision space. In conjunction with real-time IoT data streams, the system possesses continuous evolutionary capabilities, driving iterative upgrades of charging pile networks toward intelligent and adaptive capabilities.

[0074] like Figure 2 As shown, the following is an embodiment of the charging pile revenue estimation and evaluation system provided by the embodiment of the present disclosure. This system and the charging pile revenue estimation and evaluation method of the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiment of the charging pile revenue estimation and evaluation system, please refer to the embodiment of the above-mentioned charging pile revenue estimation and evaluation method.

[0075] A charging pile revenue estimation and evaluation system includes: a data acquisition module, a user group classification module, a selection probability model construction module, a dynamic revenue prediction module and a strategy optimization module.

[0076] The data acquisition module is used to collect historical data of user charging behavior through the charging pile terminal and construct user feature vectors.

[0077] The user group segmentation module is used to divide users into different user groups based on the K-means clustering algorithm and user feature vectors, and to determine the user set and cluster center of each category.

[0078] The selection probability model construction module is used to introduce the time period convenience factor and use the improved Logit model to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability.

[0079] The dynamic revenue prediction module is used to construct a dynamic revenue prediction function that includes nonlinear coupling terms between electricity prices and user behavior, and calculate the total daily revenue based on user feature vectors, each type of user set, and time period selection probability.

[0080] The strategy optimization module is used to solve the optimal electricity price strategy that maximizes total revenue based on the gradient descent algorithm and the dynamic revenue prediction function.

[0081] The charging pile revenue estimation and evaluation system provided in this embodiment collects historical data on user charging behavior to construct feature vectors and uses K-means clustering to accurately divide user groups. It introduces a time period convenience factor to accurately calculate the time period selection probability using an improved Logit model, constructs a dynamic revenue prediction function that includes a nonlinear coupling term between electricity price and user behavior to achieve accurate revenue prediction, and then solves the optimal electricity price strategy based on a gradient descent algorithm. It also fully utilizes various aspects of user charging behavior data to provide scientific decision-making support for charging pile operations from multiple dimensions such as user segmentation, time period selection, revenue prediction, electricity price strategy, and data utilization, significantly improving operational management and revenue.

[0082] Figure 3 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0083] The charging pile revenue estimation and evaluation method provided in the embodiment of the present application can be applied to electronic devices. Those skilled in the art will understand that the electronic device structure involved in the embodiment of the present invention does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In the embodiment of the present invention, the electronic device includes but is not limited to laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.

[0084] The electronic device may include a processor, an external memory interface, an internal memory, a universal serial bus (USB) interface, a charging management module, a power management module, a battery, a wireless communication module, an audio module, a speaker, a microphone, a sensor module, a button, a camera, a display, and a SIM card interface, etc.

[0085] A processor may include one or more processing units, such as a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU). Different processing units may be independent devices or integrated into one or more processors.

[0086] The processor can be the nerve center and command center of the electronic device. The controller can generate operation control signals based on the instruction opcode and timing signal to complete the control of instruction fetching and execution.

[0087] The processor may also include a memory for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. This memory can store instructions or data that the processor has just used or is reusing. If the processor needs to use the instruction or data again, it can directly call it from the memory. This avoids repeated accesses, reduces processor latency, and thus improves system efficiency.

[0088] The external memory interface can be used to connect an external memory card, such as a MicroSD card, to expand the storage capacity of an electronic device. The external memory card communicates with the processor through the external memory interface, enabling data storage. For example, files such as music and videos can be stored on the external memory card.

[0089] Internal memory can be used to store computer-executable program code, which includes instructions. The processor executes the instructions stored in the internal memory to perform various functional applications and data processing of the electronic device. The internal memory can include a program storage area and a data storage area. The internal memory can include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, universal flash storage (UFS), etc.

[0090] The wireless communication function of an electronic device can be implemented through an antenna, a wireless communication module, a modem processor, and a baseband processor.

[0091] Wireless communication modules can provide wireless communication solutions for electronic devices, including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared technology (IR), etc.

[0092] Electronic devices can implement audio functions through audio modules, speakers, receivers, microphones, headphone jacks, and application processors.

[0093] Electronic devices can achieve shooting functions through ISP, camera, video codec, GPU, display and application processor.

[0094] Electronic devices can achieve display functions through GPU, display screen and application processor.

[0095] A GPU is a microprocessor for image processing that connects the display screen to the application processor. The GPU performs mathematical and geometric calculations for graphics rendering. A processor may include one or more GPUs, which execute program instructions to generate or modify display information.

[0096] The display screen is used to display images, videos, etc. The display screen includes a display panel.

[0097] The above-mentioned electronic device realizes the charging pile revenue estimation and evaluation method of the present application by constructing a user charging behavior feature vector and adopting the K-means clustering algorithm to realize refined user grouping, introducing the time period convenience factor and the improved Logit model to establish a dynamic selection probability model, constructing a dynamic revenue prediction function including the nonlinear coupling term of electricity price and user behavior and combining the gradient descent algorithm to solve the optimal electricity price strategy under multi-dimensional constraints, and at the same time integrating multi-dimensional data such as user behavior, equipment status, external environment and power grid constraints to establish a full-factor collaborative management framework, achieving multiple beneficial effects of accurately identifying differentiated user groups, quantifying the interaction between electricity price fluctuations and user convenience preferences, forming a prediction-decision-making-optimization closed-loop system, and promoting the evolution and upgrading of operation management to a global system.

[0098] The storage medium provided in this application stores a program product that can implement a charging pile revenue estimation and evaluation method.

[0099] The evaluation methods for estimating the benefits of charging piles include: Collect historical data of user charging behavior through charging pile terminals and construct user feature vectors; Based on the K-means clustering algorithm, users are divided into different user groups using user feature vectors, and the user set and cluster center of each type are determined; The time period convenience factor is introduced, and the improved Logit model is used to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability; A dynamic revenue prediction function is constructed that includes nonlinear coupling terms between electricity prices and user behavior, and the total daily revenue is calculated based on user feature vectors, user sets of each type, and time period selection probabilities. Based on the gradient descent algorithm, the optimal electricity price strategy that maximizes total revenue is solved according to the dynamic revenue prediction function.

[0100] In some possible embodiments, the charging pile revenue estimation and evaluation method disclosed herein can be implemented in the form of a program product, which includes program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps described in the above "Exemplary Method" section of this specification according to various exemplary embodiments of the present disclosure.

[0101] The storage medium of the present disclosure can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable 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 thereof.

[0102] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A charging pile revenue estimation and evaluation method, characterized in that: include: Collect historical data of user charging behavior through charging pile terminals and construct user feature vectors; Based on the K-means clustering algorithm, users are divided into different user groups using user feature vectors, and the user set and cluster center of each type are determined; The time period convenience factor is introduced, and the improved Logit model is used to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability; A dynamic revenue prediction function is constructed that includes nonlinear coupling terms between electricity prices and user behavior, and the total daily revenue is calculated based on user feature vectors, user sets of each type, and time period selection probabilities. Based on the gradient descent algorithm, the optimal electricity price strategy that maximizes total revenue is solved according to the dynamic revenue prediction function.

2. The charging pile revenue estimation and evaluation method according to claim 1, characterized in that: The method of collecting historical data of user charging behavior through the charging pile terminal and constructing a user feature vector includes: Based on the Internet of Things, historical data on user charging behavior is collected through charging pile terminals; The number of users, average daily charging times, average charging time, and price sensitivity coefficient are obtained from the historical data of user charging behavior to construct the user feature vector U k =[N k ,C k ,T k ,β k ]; Among them, U k is the feature vector of the k-th user, N k is the total number of users in the kth category, C k is the average daily charging times for the k-th user, T k is the average single charging time for the k-th user, β k is the price sensitivity coefficient of the k-th category of users.

3. The charging pile revenue estimation and evaluation method according to claim 2, characterized in that: The K-means clustering algorithm is based on dividing users into different user groups using user feature vectors, and determining the user set and cluster center of each type, including: The K-means clustering algorithm is used to divide users into K categories with the following formula as the optimization objective to reduce the impact of group differences on revenue forecasting: in, is the k-th user set, is the feature vector cluster center of the k-th user.

4. The charging pile revenue estimation and evaluation method according to claim 3, characterized in that: The time period convenience factor is introduced, and the improved Logit model is used to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability, including: Introducing time period convenience factor ,Using the improved Logit model, based on the divided user groups and price sensitivity coefficients, a time period selection probability model considering user price sensitivity is established; The time period selection probability model includes: in, The probability of selecting time period i for the k-th user, is the electricity price in period i, is the convenience factor of period i, The time period preference coefficient of the k-th user; In the above formula, is the idle rate of charging piles in period i, is the normalized charging times of period i, is the charging pile idle rate coefficient determined based on historical data, is the charging times coefficient determined based on historical data.

5. The charging pile revenue estimation and evaluation method according to claim 4, characterized in that: The dynamic profit prediction function includes: in, is the total daily income, is the average charging power at different times of the day, is the charging selection probability of the kth user in time period i.

6. The charging pile revenue estimation and evaluation method according to claim 5, characterized in that: The method is based on the gradient descent algorithm and solves the optimal electricity price strategy that maximizes the total revenue according to the dynamic revenue prediction function, including: Based on the gradient descent algorithm, the dynamic profit prediction function is optimized to calculate the electricity price in each period. Find the partial derivative, and iteratively calculate the partial derivative expression until the partial derivative approaches zero, and solve the optimal electricity price strategy that maximizes the total revenue ; The partial derivative expression includes: 。 7. The charging pile revenue estimation and evaluation method according to claim 2, characterized in that: The historical data of the user's charging behavior includes: user type, charging period, duration, power, and frequency.

8. A charging pile revenue estimation and evaluation system, characterized in that: The system adopts the charging pile revenue estimation and evaluation method according to any one of claims 1 to 7; The system comprises: The data collection module is used to collect historical data of user charging behavior through the charging pile terminal and construct user feature vectors; The user group segmentation module is used to divide users into different user groups based on the K-means clustering algorithm and user feature vectors, and to determine the user set and cluster center of each category; The selection probability model building module is used to introduce the time period convenience factor and use the improved Logit model to establish a time period selection probability model that takes into account user price sensitivity to calculate the time period selection probability; Dynamic revenue prediction module, which is used to construct a dynamic revenue prediction function that includes nonlinear coupling terms between electricity prices and user behavior, and calculate the total daily revenue based on user feature vectors, each type of user set, and time period selection probability; The strategy optimization module is used to solve the optimal electricity price strategy that maximizes total revenue based on the gradient descent algorithm and the dynamic revenue prediction function.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the charging pile revenue estimation and evaluation method according to any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the charging pile revenue estimation and evaluation method according to any one of claims 1 to 7 are implemented.

Images