An orderly charging management system based on deep learning algorithm

Abstract

The present invention discloses an orderly charging management system based on a deep learning algorithm, which relates to the technical field of orderly charging management, and includes: a distributed energy module for supplying power to charging equipment; a data acquisition module for acquiring power grid data, distributed energy module data and status data of all connected devices to be charged; a feature extraction module for preprocessing the data acquired by the data acquisition module, and constructing features of the state to be charged according to the preprocessed data; an orderly charging management module for inputting the features of the state to be charged into a trained orderly charging management model, outputting a device charging priority table, and performing orderly charging management on all connected devices to be charged according to the device charging priority table. By using a deep learning algorithm to perform real-time analysis and prediction on the operating data of the charging equipment, the charging strategy can be automatically adjusted according to the state of the equipment and environmental changes, thereby realizing intelligent scheduling of the charging equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of orderly charging management, and specifically to an orderly charging management system based on a deep learning algorithm. Background Art

[0002] With the rapid development of renewable energy, such as the widespread use of solar energy, distributed energy systems have gradually become an important part of modern power systems. Distributed energy systems connect decentralized power generation units to the power grid to achieve distributed supply and storage of energy, thereby improving the flexibility and sustainability of the power system. However, due to the instability and intermittency of distributed energy output, how to efficiently manage these distributed energy sources has become an important technical challenge.

[0003] Traditional charging management systems mainly rely on fixed charging strategies and preset scheduling rules, which are difficult to cope with dynamic changes and complex situations in actual operation. This static management method often leads to low system efficiency, low energy utilization, and may even cause instability in the power grid. In recent years, with the development of deep learning and artificial intelligence technology, researchers have begun to try to apply intelligent algorithms to charging management systems, in order to achieve intelligent scheduling and optimized management of charging equipment through learning historical data and analyzing real-time data.

[0004] The existing orderly charging management system based on deep learning still has some shortcomings, such as how to accurately extract and analyze the operating characteristics of charging equipment, how to make real-time scheduling according to the status of different equipment and environmental changes, and how to optimize energy utilization efficiency while ensuring system safety. Summary of the invention

[0005] Based on the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide an orderly charging management system based on a deep learning algorithm to solve the above-mentioned technical problems.

[0006] To achieve the above object, the present invention provides the following technical solution: an orderly charging management system based on a deep learning algorithm, comprising:

[0007] A distributed energy module is used to supply power to the charging device, wherein the distributed energy module includes a photovoltaic power generation unit and an energy storage unit; the photovoltaic power generation unit is used to convert sunlight into electrical energy to directly supply power to the charging device or store it in the energy storage unit;

[0008] A data acquisition module is used to acquire power grid data, distributed energy module data and status data of all connected devices to be charged;

[0009] A feature extraction module, used to preprocess the power grid data, distributed energy module data and status data of all connected devices to be charged, and construct charging state features according to the preprocessed power grid data, distributed energy module data and status data of all connected devices to be charged, wherein the charging state features include dynamic charging demand features, battery health features, expected charging time features and intelligent charging scheduling features;

[0010] The orderly charging management module is used to input the characteristics of the state to be charged into the trained orderly charging management model, output a device charging priority table, and perform orderly charging management on all connected devices to be charged according to the device charging priority table.

[0011] The present invention is further configured such that the grid data includes the grid load level; the distributed energy module data includes photovoltaic power generation and energy storage capacity; and the status data of the device to be charged includes the number of charge and discharge cycles, maximum battery capacity, charging power, current remaining power, connection time and battery temperature.

[0012] The present invention is further configured to construct a charging state feature based on preprocessed grid data, distributed energy module data and state data of all connected devices to be charged, including:

[0013] Calculating a dynamic charging demand characteristic according to the current remaining power and the connection time;

[0014] Calculating a battery health characteristic based on the number of charge and discharge cycles and the battery temperature;

[0015] Calculating an expected charging time characteristic according to the maximum capacity of the battery and the charging power;

[0016] The intelligent charging scheduling characteristics are calculated according to the grid load level, the photovoltaic power generation and the energy storage capacity.

[0017] The present invention is further configured such that the calculation logic of the dynamic charging demand characteristic is: Among them, DCDI i is the dynamic charging demand characteristics of the i-th device to be charged, R i is the current remaining power of the i-th device to be charged, T i is the connection time of the i-th device to be charged, α is the correction coefficient, and its value range is [0,1]; σ() is the Sigmoid function, which is used to limit the result between 0 and 1.

[0018] The present invention is further configured such that the calculation logic of the battery health feature is: Among them, BHI iis the battery health characteristic of the i-th device to be charged, C i is the number of charge and discharge cycles of the i-th device to be charged, D i is the battery temperature of the i-th device to be charged, β and γ are correction coefficients used to correct the influence of the number of charge and discharge cycles and battery temperature, and the value range is [0,1].

[0019] The present invention is further configured such that the calculation logic of the expected charging time characteristic is: Among them, ECT i is the expected charging time characteristic of the i-th device to be charged, C max is the maximum battery capacity of the i-th device to be charged, S i is the charging power of the i-th device to be charged.

[0020] The present invention is further configured such that the calculation logic of the intelligent charging scheduling feature is: Where, ICSF is the intelligent charging scheduling feature, L(t) is the grid load level at time t, P pv (t) is the photovoltaic power generation at time t, tanh() is the hyperbolic tangent function, E storage (t) is the energy storage capacity at time t, g(E storage (t)) is the influence function of the energy storage capacity at time t, which is a nonlinear piecewise function based on the energy storage capacity. It is used to quantify the impact of the energy storage unit on charging scheduling when the capacity is within the set threshold range. g(E storage The calculation logic of (t)) is: Among them, E min When the grid is working, the minimum threshold of energy storage capacity is set. When the grid is not working, E min is 0.

[0021] The present invention is further configured such that the construction logic of the orderly charging management model is:

[0022] Performing a threshold judgment according to the intelligent charging scheduling characteristics, and suspending the charging of the device to be charged when the intelligent charging scheduling characteristics are greater than or equal to a preset intelligent charging scheduling threshold;

[0023] When the intelligent charging scheduling feature is less than a preset intelligent charging scheduling threshold, the dynamic charging demand feature, the battery health feature and the expected charging duration feature are taken as inputs and normalized;

[0024] According to the normalized dynamic charging demand feature, the battery health feature and the expected charging time feature, a nonlinear feature combination is constructed, and the interaction relationship between the features is mapped by a function;

[0025] According to the calculated nonlinear feature combination, the optimal weight of the features of the device to be charged is determined through an iterative optimization algorithm to maximize the overall charging efficiency;

[0026] The device charging priority is calculated according to the nonlinear feature combination and the corresponding optimal weight, a device charging priority table is generated according to the descending order of the device charging priority, and orderly charging management is performed on the charging devices according to the device charging priority table.

[0027] The present invention is further configured such that the calculation logic of the nonlinear feature combination is: Φ i is the nonlinear characteristic combination of the i-th device to be charged;

[0028] The optimal weights of the features of the device to be charged are determined through an iterative optimization algorithm. The calculation logic is: Among them, ω i is the nonlinear characteristic combination Φ of the i-th device to be charged i The optimal weight of L is the target loss function, ∑ j Φ j represents the sum of the nonlinear characteristic combinations of j devices to be charged;

[0029] The device charging priority is calculated based on the nonlinear feature combination and the corresponding optimal weight, and the calculation logic is: Among them, Priority i is the device charging priority of the i-th device to be charged, ∑ j ω j ·Φ j It is the sum of the product of the nonlinear feature combination and the optimal weight of the j devices to be charged.

[0030] The present invention is further configured such that the calculation logic of the target loss function L is: Among them, P max is the maximum power supply of the power grid, η is the target stable value of the power grid load, and its value range is (0,1].

[0031] The present invention provides an orderly charging management system based on a deep learning algorithm, comprising: a distributed energy module, used to power a charging device, wherein the distributed energy module comprises a photovoltaic power generation unit and an energy storage unit; the photovoltaic power generation unit is used to convert sunlight into electrical energy, directly powering the charging device, or storing it in the energy storage unit; a data acquisition module, used to acquire power grid data, distributed energy module data and status data of all connected devices to be charged; a feature extraction module, used to pre-process the power grid data, distributed energy module data and status data of all connected devices to be charged, and construct a to-be-charged state feature according to the pre-processed power grid data, distributed energy module data and status data of all connected devices to be charged, wherein the to-be-charged state feature comprises a dynamic charging demand feature, a battery health feature, an expected charging time feature and an intelligent charging scheduling feature; an orderly charging management module, used to input the to-be-charged state feature into a trained orderly charging management model, output a device charging priority table, and perform orderly charging management on all connected devices to be charged according to the device charging priority table, and the beneficial effects produced include:

[0032] 1. Intelligent charging management: The present invention uses deep learning algorithms to analyze and predict the operating data of charging equipment in real time, and can automatically adjust the charging strategy according to the status of the equipment and environmental changes, thereby realizing intelligent scheduling of charging equipment. Compared with the traditional static management method, this system can more effectively respond to dynamic changes in actual operation, significantly improving the flexibility and response speed of the system;

[0033] 2. Optimized energy utilization: Through intelligent scheduling of distributed energy, the present invention can maximize the use of renewable energy and reduce energy waste. In addition, the system can also manage the charging and discharging process of energy storage equipment through optimization algorithms, thereby improving the battery life and energy efficiency, thereby improving the overall energy utilization rate;

[0034] 3. Improve system stability: The orderly charging management system provided by the present invention can effectively predict and prevent potential operating risks and ensure the stable operation of the system by comprehensively monitoring and extracting the status of charging equipment. Especially when the load of the power grid fluctuates greatly, the system can avoid the instability of the power grid caused by overload or underload of charging equipment through intelligent scheduling;

[0035] 4. Extend the life of equipment: The present invention avoids equipment loss caused by overcharging or discharging in traditional charging management systems by accurately calculating and controlling the charging and discharging process of charging equipment, thereby extending the service life of batteries and other charging equipment. In addition, through real-time evaluation of health characteristics, the system can also promptly detect and handle abnormal conditions of equipment, further improving the reliability and durability of the equipment.

[0036] 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

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing 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. In the drawings:

[0038] Figure 1 A schematic diagram of the structure of an orderly charging management system based on a deep learning algorithm is shown as an exemplary embodiment of the present invention;

[0039] Figure 2 A flowchart for constructing an orderly charging management model of an orderly charging management system based on a deep learning algorithm is shown as an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0041] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0042] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0043] An orderly charging management system based on deep learning algorithm, such as Figure 1 As shown, including:

[0044] A distributed energy module is used to supply power to the charging device, wherein the distributed energy module includes a photovoltaic power generation unit and an energy storage unit; the photovoltaic power generation unit is used to convert sunlight into electrical energy to directly supply power to the charging device or store it in the energy storage unit;

[0045] A data acquisition module is used to acquire power grid data, distributed energy module data and status data of all connected devices to be charged;

[0046] A feature extraction module, used to preprocess the power grid data, distributed energy module data and status data of all connected devices to be charged, and construct charging state features according to the preprocessed power grid data, distributed energy module data and status data of all connected devices to be charged, wherein the charging state features include dynamic charging demand features, battery health features, expected charging time features and intelligent charging scheduling features;

[0047] The orderly charging management module is used to input the characteristics of the state to be charged into the trained orderly charging management model, output a device charging priority table, and perform orderly charging management on all connected devices to be charged according to the device charging priority table.

[0048] Specifically, the photovoltaic power generation unit directly converts the light energy in sunlight into electrical energy through photovoltaic components (including solar panels). When the photovoltaic power generation is sufficient, the generated electricity can be directly used to power the connected charging equipment. This direct power supply method reduces the dependence on traditional grid electricity and uses clean energy to reduce carbon emissions. When the photovoltaic power generation exceeds the immediate electricity demand, the excess electricity will be stored in the energy storage unit. The purpose of storing electricity is to balance the time difference between power supply and electricity consumption, ensuring stable power supply and emergency power supply in the absence of sunlight or insufficient photovoltaic power generation.

[0049] The data acquisition module acquires the grid data, the distributed energy module data and the status data of all connected devices to be charged. The present invention is further configured that the grid data includes the grid load level; the distributed energy module data includes the photovoltaic power generation and the energy storage power; the status data of the device to be charged includes the number of charge and discharge cycles, the maximum capacity of the battery, the charging power, the current remaining power, the connection time and the battery temperature; specifically, the grid load level represents the ratio of the power demand carried by the grid at the current moment to the maximum power supply of the grid, and the value range is [0,1]. The grid load level is a dynamic value that changes in real time; the photovoltaic power generation refers to the power generated by the photovoltaic power generation unit at the current moment; the energy storage power represents the power currently stored in the energy storage unit; the number of charge and discharge cycles is an important indicator for measuring the service life of the battery, which represents the number of times the battery completes a complete charging and discharging process. As the number of cycles increases, the health status and capacity of the battery usually decrease gradually; the maximum capacity of the battery represents the maximum power that the battery can store in the current state; the charging power represents the actual power of the current device when charging; the current remaining power represents the current power of the device battery; the connection time represents the connection time of the device to be charged; the battery temperature represents the difference between the current temperature of the battery and the ambient temperature.

[0050] The grid data, distributed energy module data and status data of all connected devices to be charged are preprocessed by the feature extraction module. Data preprocessing is the first step of the feature extraction module, and its purpose is to ensure that the data collected from the grid, distributed energy modules and devices to be charged are accurate, consistent and suitable for further analysis. Preprocessing usually includes the following steps: Data cleaning: remove noise data and outliers to ensure data accuracy, including eliminating spikes in grid load data or obviously erroneous records in device status data; Data normalization: standardize data of different dimensions so that the data can be compared at the same scale, including normalizing data of different units to the range of 0 to 1; Data interpolation: For missing data, appropriate methods are used to fill in to ensure data integrity, including linear interpolation or prediction based on historical data, which is not limited here;

[0051] The present invention is further configured to construct a charging state feature based on preprocessed grid data, distributed energy module data and state data of all connected devices to be charged, including:

[0052] The dynamic charging demand feature is calculated according to the current remaining power and the connection time; the present invention is further configured that the calculation logic of the dynamic charging demand feature is: Among them, DCDI i is the dynamic charging demand characteristics of the i-th device to be charged, R iis the current remaining power of the i-th device to be charged, T i is the connection time of the i-th device to be charged, α is the correction coefficient, and its value range is [0,1], σ() is the Sigmoid function, which is used to limit the result between 0 and 1; Specifically, the dynamic charging demand feature is used to quantify the urgency of the device's charging demand and compare it between different devices to reasonably allocate charging resources. In the above calculation logic, With R i As the remaining power decreases (i.e., the remaining power decreases), its contribution value will increase significantly, indicating that devices with lower power have higher charging requirements. This calculation uses the rapid growth characteristics of the exponential function, which makes the charging demand rise sharply when the power is close to a low value; by ln(1+T i ), taking the impact of connection time into consideration, the logarithmic function is used to deal with the nonlinear impact of connection time, ensuring that the growth rate of charging demand characteristics gradually slows down with the increase of connection time, reflecting the cumulative effect of waiting time, but avoiding the situation where the demand characteristics increase infinitely due to waiting too long; through the correction coefficient α, a balance is adjusted between the remaining power and the connection time, so that the system can autonomously adjust the impact of the two according to the actual application needs. When α increases, it tends to pay more attention to the impact of the connection time on the charging demand;

[0053] The battery health feature is calculated according to the number of charge and discharge cycles and the battery temperature; the present invention is further configured such that the calculation logic of the battery health feature is: Among them, BHI i is the battery health characteristic of the i-th device to be charged, C i is the number of charge and discharge cycles of the i-th device to be charged, D i is the battery temperature of the i-th device to be charged, β and γ are correction coefficients used to correct the influence of the number of charge and discharge cycles and battery temperature, and the value range is [0,1]; specifically, the battery health feature is used to evaluate the current health status of the device battery, aiming to reflect the degree of battery degradation caused by cyclic use and temperature influence; through the square root function in the formula As the number of charge and discharge cycles increases, the health characteristic value BHI i It will gradually decrease, reflecting the aging trend of the battery. The introduction of the square root function can smooth the impact of the number of cycles and avoid a sharp drop in the health characteristic value when the number of cycles increases. i , the impact of battery temperature on health characteristics is further quantified. When the battery temperature is high, the value of this item increases, resulting in the overall health characteristic value BHI iDecreases indicate that temperature has an adverse effect on battery health. By adjusting the correction factor β, the effects of charge and discharge cycles and temperature on battery health are balanced. Increasing β can enhance the effect of cycle times on health characteristics.

[0054] The expected charging time characteristic is calculated according to the maximum capacity of the battery and the charging power; the present invention is further configured such that the calculation logic of the expected charging time characteristic is: Among them, ECT i is the expected charging time characteristic of the i-th device to be charged, C max is the maximum battery capacity of the i-th device to be charged, S i is the charging power of the i-th device to be charged; the expected charging time feature is used to predict the time required for the device to be fully charged from the current state; through 1-R i , quantifies the proportion of the device's current power that is not fully charged. The larger this item is, the lower the current power of the device is, and the longer it takes to charge. Conversely, the higher the remaining power, the shorter the charging time. max Reflects the maximum chargeable capacity of the battery. For the same remaining power ratio 1-R i The larger the battery capacity, the longer the charging time. The calculation formula shows that for a large-capacity battery, even if its remaining power is high, the charging time may still be long; through S i , which can reflect the impact of charging speed on charging time. The greater the power, the faster the charging speed and the shorter the expected charging time;

[0055] The intelligent charging scheduling feature is calculated according to the grid load level, the photovoltaic power generation and the energy storage power; the present invention is further configured that the calculation logic of the intelligent charging scheduling feature is: Where, ICSF is the intelligent charging scheduling feature, L(t) is the grid load level at time t, P pv (t) is the photovoltaic power generation at time t, tanh() is the hyperbolic tangent function, E storage (t) is the energy storage capacity at time t, g(E storage (t)) is the influence function of the energy storage capacity at time t, which is a nonlinear piecewise function based on the energy storage capacity. It is used to quantify the impact of the energy storage unit on charging scheduling when the capacity is within the set threshold range. g(E storage The calculation logic of (t)) is: Among them, E min When the grid is working, the minimum threshold of energy storage capacity is set. When the grid is not working, E minis 0; specifically, the logarithmic function ln(1+L(t)) reflects the impact of the current load of the power grid. The logarithmic function ensures that when the power grid load is high, the eigenvalue increases, indicating that the system should reduce charging activities to avoid increasing the pressure on the power grid. Correspondingly, when the power grid load is low, the system can increase the charging load; through the hyperbolic tangent function tanh(P pv (t)), indicating the positive impact of photovoltaic power generation on charging scheduling. The higher the power generation, the higher the tanh(P pv (t)) is closer to 1, thus reducing the denominator and increasing the ICSF value, indicating that the system should use more photovoltaic power generation for charging; through the function g(E storage (t)), it is divided into two cases: 1. When the energy storage capacity is greater than or equal to the set minimum threshold E min When the energy storage capacity is lower than E, the function is expressed as the square root of the energy storage capacity, indicating that sufficient energy storage can significantly reduce the pressure on the power grid; 2. When the energy storage capacity is lower than E min When , the function is a negative exponential function, which means that when the energy storage capacity is insufficient, the negative impact on charging scheduling is aggravated, and the system should give priority to reducing charging activities to protect the energy storage unit from being over-consumed;

[0056] like Figure 2 As shown, the present invention is further configured such that the construction logic of the orderly charging management model is:

[0057] A threshold value is judged according to the intelligent charging scheduling characteristics. When the intelligent charging scheduling characteristics are greater than or equal to a preset intelligent charging scheduling threshold value, the charging of the device to be charged is suspended. Specifically, the step of judging the threshold value according to the intelligent charging scheduling characteristics is a dynamic adjustment mechanism for real-time monitoring and adjusting the charging strategy during the charging process to ensure the stability of the power grid, the efficient use of resources, and the safe operation of the equipment and the energy storage unit. When ICSF is greater than or equal to a preset threshold value θ ICSF , it is judged that the current grid load, photovoltaic power generation or energy storage power and other factors have reached a level that is not conducive to continued charging. At this time, in order to protect the stability of the grid, prevent the energy storage unit from over-discharging, or reduce energy waste when the photovoltaic power generation is insufficient, the system will suspend the charging operation of the charging device; when ICSF is less than the preset threshold θ ICSF When the current condition is determined to be suitable for continuing the charging operation, the charging device can continue to charge.

[0058] When the intelligent charging scheduling feature is less than a preset intelligent charging scheduling threshold, the dynamic charging demand feature, the battery health feature and the expected charging duration feature are taken as inputs and normalized;

[0059] According to the normalized dynamic charging demand feature, the battery health feature and the expected charging time feature, a nonlinear feature combination is constructed, and the interactive relationship between the features is mapped by a function; the present invention is further configured such that the calculation logic of the nonlinear feature combination is: Φ i is the nonlinear characteristic combination of the i-th device to be charged, which is used to comprehensively reflect the complex interactive relationship between the charging state characteristics and serves as an important basis for optimizing the charging strategy. The logarithmic function is used to smooth the product of the dynamic charging demand and the battery health characteristics to ensure that the final result is not excessively affected when the characteristic value is too large, thereby improving the stability of the calculation. i The value of is normalized, so sin(π / 2·BHI i ) always remains positive, and when other parameters remain unchanged, BHI i The larger the i The larger the value, the greater the influence of the feature combination when the battery is in good health, indicating that the system can charge more actively at this time; otherwise, the charging rhythm should be slowed down; the expected charging time feature is expressed by the exponential function It is introduced and has a greater impact on the feature combination when the charging time is shorter, prompting the system to complete the charging operation in a short time, ensuring the balance between efficient charging and grid stability.

[0060] According to the calculated nonlinear feature combination, the optimal weight of the feature of the device to be charged is determined by an iterative optimization algorithm to maximize the overall charging efficiency; the present invention is further configured to determine the optimal weight of the feature of the device to be charged by an iterative optimization algorithm, and the calculation logic is: Among them, ω i is the nonlinear characteristic combination Φ of the i-th device to be charged i The optimal weight of L is the target loss function, ∑ j Φ j represents the sum of the nonlinear characteristic combinations of j devices to be charged; specifically, Represents the nonlinear characteristic combination Φ of the i-th device to be charged i The partial derivative of the target loss function L reflects the nonlinear feature combination Φ i The extent of the impact on the overall loss; Is a smooth adjustment factor used to dynamically adjust the weight ω i The final value of It represents the i-th nonlinear feature combination Φ iThe proportion of the total feature combination of all devices to be charged; the goal of the above calculation logic is to optimize the charging strategy by minimizing the loss function. To achieve this goal, it is necessary to know the nonlinear feature combination Φ i The impact on the loss function is calculated by calculating the partial derivative To determine, the larger the partial derivative value, the more nonlinear the feature combination Φ i The more significant the impact on the loss function, the higher the weight should be;

[0061] The present invention is further configured such that the calculation logic of the target loss function L is: Among them, P max is the maximum power supply of the power grid, η is the target stability value of the power grid load, and its value range is (0,1]; the target loss function L is used to measure the deviation of the system from the power grid load and the utilization of renewable energy during the charging process. By minimizing the loss function, the charging scheduling strategy of the system is optimized; the target loss function optimizes the dynamic balance between the power grid load level and the photovoltaic power generation and energy storage power, thereby reducing excessive dependence on the power grid; through the optimization of the target loss function, it can ensure the stability of the power grid under different load conditions and avoid the instability of the power system caused by load fluctuations.

[0062] The device charging priority is calculated according to the nonlinear feature combination and the corresponding optimal weight, and a device charging priority table is generated according to the descending order of the device charging priority. The charging priority table is used to perform orderly charging management on the devices to be charged. The present invention is further configured to calculate the device charging priority according to the nonlinear feature combination and the corresponding optimal weight, and the calculation logic is: Among them, Priority i is the device charging priority of the i-th device to be charged, ∑ j ω j ·Φ j is the sum of the product of the nonlinear feature combination and the optimal weight of the j devices to be charged; specifically, the device charging priority is used to determine the order of the devices in the charging process, thereby achieving orderly charging management; Is a regulation factor based on the device battery health characteristic BHI i Dynamic adjustments are made to ensure that devices in better health are appropriately promoted in charging priority; by calculating the charging priority of each device, the system can scientifically allocate charging resources to ensure that high-priority devices are charged first when resources are limited, thereby optimizing charging efficiency and helping to effectively manage power resources in a multi-device environment to avoid inefficient charging or waste of resources. At the same time, the priority calculation formula introduces The item is adjusted according to the battery health status to ensure that the long-term service life of the device is taken into account in actual applications. For devices with good battery health, their priority will be appropriately increased to ensure that they can get sufficient charge, thereby extending the battery life.

[0063] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0064] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0065] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0066] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0067] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0068] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0069] In the several embodiments provided in the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0070] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0071] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0072] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0073] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An orderly charging management system based on deep learning algorithm, characterized in that: include: A distributed energy module is used to supply power to the charging device, wherein the distributed energy module includes a photovoltaic power generation unit and an energy storage unit; the photovoltaic power generation unit is used to convert sunlight into electrical energy to directly supply power to the charging device or store it in the energy storage unit; A data acquisition module is used to acquire power grid data, distributed energy module data and status data of all connected devices to be charged; A feature extraction module, used to preprocess the power grid data, distributed energy module data and status data of all connected devices to be charged, and construct charging state features according to the preprocessed power grid data, distributed energy module data and status data of all connected devices to be charged, wherein the charging state features include dynamic charging demand features, battery health features, expected charging time features and intelligent charging scheduling features; An orderly charging management module, used to input the characteristics of the state to be charged into a trained orderly charging management model, output a device charging priority table, and perform orderly charging management on all connected devices to be charged according to the device charging priority table; The charging status features are constructed based on the preprocessed grid data, distributed energy module data and the status data of all connected charging devices, including: Calculate dynamic charging demand characteristics based on current remaining power and connection time; Calculating a battery health characteristic based on the number of charge and discharge cycles and the battery temperature; Calculate expected charging time characteristics according to the maximum capacity and charging power of the battery; Calculating intelligent charging scheduling characteristics according to the grid load level, the photovoltaic power generation and the energy storage power; The calculation logic of the dynamic charging demand feature is: Among them, DCDI i is the dynamic charging demand characteristics of the i-th device to be charged, R i is the current remaining power of the i-th device to be charged, T i is the connection time of the i-th device to be charged, α is the correction coefficient, the value range is [0,1], σ( ) is the Sigmoid function, which is used to limit the result between 0 and 1; The calculation logic of the battery health feature is: Among them, BHI i is the battery health characteristic of the i-th device to be charged, C i is the number of charge and discharge cycles of the i-th device to be charged, D i is the battery temperature of the i-th device to be charged, β and γ are correction coefficients used to correct the influence of the number of charge and discharge cycles and battery temperature, and the value range is [0,1]; The calculation logic of the expected charging time feature is: Among them, ECT i is the expected charging time characteristic of the i-th device to be charged, C max is the maximum battery capacity of the i-th device to be charged, S i is the charging power of the i-th device to be charged; The calculation logic of the intelligent charging scheduling feature is: ICSF = ln(1+L(t)) / (tanh(P pv (t)+g(E storage (t)))), where ICSF is the intelligent charging scheduling feature, L(t) is the grid load level at time t, and P pv (t) is the photovoltaic power generation at time t, tanh( ) is the hyperbolic tangent function, E storage (t) is the energy storage capacity at time t, g(E storage (t)) is the influence function of the energy storage capacity at time t, which is a nonlinear piecewise function based on the energy storage capacity. It is used to quantify the impact of the energy storage unit on charging scheduling when the capacity is within the set threshold range. g(E storage The calculation logic of (t)) is: Among them, E min When the grid is working, the minimum threshold of energy storage capacity is set. When the grid is not working, E min is 0; The construction logic of the orderly charging management model is: Performing a threshold judgment according to the intelligent charging scheduling characteristics, and suspending the charging of the device to be charged when the intelligent charging scheduling characteristics are greater than or equal to a preset intelligent charging scheduling threshold; When the intelligent charging scheduling feature is less than a preset intelligent charging scheduling threshold, the dynamic charging demand feature, the battery health feature and the expected charging duration feature are taken as inputs and normalized; According to the normalized dynamic charging demand feature, the battery health feature and the expected charging time feature, a nonlinear feature combination is constructed, and the interaction relationship between the features is mapped by a function; According to the calculated nonlinear feature combination, the optimal weight of the features of the device to be charged is determined through an iterative optimization algorithm to maximize the overall charging efficiency; Calculate the device charging priority according to the nonlinear feature combination and the corresponding optimal weight, generate a device charging priority table according to the descending order of the device charging priority, and perform orderly charging management on the devices to be charged according to the device charging priority table; The calculation logic of the nonlinear feature combination is: Φ i is the nonlinear characteristic combination of the i-th device to be charged; The optimal weights of the features of the device to be charged are determined through an iterative optimization algorithm. The calculation logic is: Among them, ω i is the nonlinear characteristic combination Φ of the i-th device to be charged i The optimal weight of L is the target loss function, ∑ j Φ j represents the sum of the nonlinear characteristic combinations of j devices to be charged; The device charging priority is calculated based on the nonlinear feature combination and the corresponding optimal weight, and the calculation logic is: Among them, Priority i is the device charging priority of the i-th device to be charged, ∑ j ω j ·Φ j is the sum of the product of the nonlinear feature combination and the optimal weight of the j devices to be charged; The calculation logic of the objective loss function L is: Among them, P max is the maximum power supply of the power grid, η is the target stable value of the power grid load, and its value range is (0,1].

2. The orderly charging management system based on deep learning algorithm according to claim 1 is characterized in that: The grid data includes the grid load level; the distributed energy module data includes photovoltaic power generation and energy storage capacity; the status data of the device to be charged includes the number of charge and discharge cycles, maximum battery capacity, charging power, current remaining power, connection time and battery temperature.

Images