Intelligent scheduling and load balancing control method for power supply equipment

Abstract

The invention provides an intelligent scheduling and load balancing control method for power supply equipment, and the method comprises the steps: receiving real-time load data of a plurality of pieces of power supply equipment in a charging region, carrying out the aggregation processing of the data, and generating regional load distribution information; acquiring power grid side power supply constraint data of the charging area; generating a power supply equipment scheduling balance factor according to the regional load distribution information and the power grid side power supply constraint data; calculating a dynamic power distribution threshold value of each power supply device based on the power supply device scheduling balance factor; in response to a detected real-time charging request of target power supply equipment, obtaining charging demand feature data corresponding to the target power supply equipment; generating a target power supply equipment scheduling instruction according to the dynamic power distribution threshold value and the charging demand characteristic data; and in response to determining that the target power supply equipment scheduling instruction represents that charging is allowed, controlling the target power supply equipment to execute charging start-stop operation. The charging efficiency of the power supply equipment and the stability of the power grid can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply systems, and more specifically, to an intelligent scheduling and load balancing control method for power supply equipment. Background Art

[0002] In today's society, with the widespread adoption of electric vehicles, the demand for power supply equipment is growing. Existing power supply equipment typically uses a fixed power distribution method for charging, meaning that each power supply device is allocated a fixed amount of power during charging. While this method is simple, it presents numerous problems in practical applications. First, this fixed power distribution method cannot be flexibly adjusted based on the actual load of the power supply equipment. As a result, when some power supply equipment is lightly loaded, it allocates too much power, resulting in energy waste; while when other power supply equipment is heavily loaded, it allocates insufficient power, affecting charging efficiency. Second, during peak load periods, existing power supply equipment cannot effectively coordinate with the power supply constraints of the grid, which can easily lead to grid overload and affect the stable operation of the grid. Furthermore, when a power supply equipment fails offline, existing technologies are unable to promptly and effectively transfer load and compensate for power, causing some power supply equipment to malfunction, impacting the user's charging experience.

[0003] In the process of implementing the embodiments of the present invention, there are at least the following problems or defects in the existing technology: the power distribution method of the existing power supply equipment is not flexible enough and cannot be dynamically adjusted according to the real-time load conditions and grid power supply constraints, resulting in energy waste and low charging efficiency; during the peak load period of the grid, the power supply relationship between the power supply equipment and the grid cannot be effectively coordinated, which easily causes grid overload; when the power supply equipment fails offline, there is a lack of effective load transfer and power compensation mechanism, which affects the normal operation of the power supply equipment and the user's charging experience. Summary of the Invention

[0004] The present invention provides an intelligent scheduling and load balancing control method for power supply equipment, comprising: Receive real-time load data of multiple power supply devices in the charging area, aggregate the real-time load data, and generate regional load distribution information; Obtaining grid-side power supply constraint data of the charging area; generating a power supply equipment scheduling balancing factor according to the regional load distribution information and the grid-side power supply constraint data; Calculating a dynamic power allocation threshold for each power supply device based on the power supply device scheduling balancing factor; In response to detecting a real-time charging request from a target power supply device, acquiring charging demand characteristic data corresponding to the target power supply device; generating a target power supply equipment scheduling instruction according to the dynamic power allocation threshold and the charging demand characteristic data; In response to determining that the scheduling instruction of the target power supply device indicates that charging is allowed, the target power supply device is controlled to perform a charging start / stop operation.

[0005] Furthermore, the real-time load data includes power data of the power supply equipment and a charging timestamp; and Before acquiring the grid-side power supply constraint data of the charging area, the method further includes: Obtaining historical load curve data of the charging area; Determining a peak load period of a power grid according to the historical load curve data and the charging timestamp; In response to detecting that the current moment belongs to the power grid load peak period, using the current maximum allowable load value of the power grid as power supply constraint data on the power grid side; Generating the power supply equipment scheduling balancing factor includes: Calculating a ratio of a load variance value of the regional load distribution information to a current maximum allowable load value of the power grid; The product of the ratio and a preset load tolerance coefficient is determined as a power supply equipment scheduling balancing factor.

[0006] Furthermore, the calculating of the dynamic power allocation threshold of each power supply device includes: Normalizing the power data sequence of the power supply equipment according to the power supply equipment scheduling balancing factor; Generate a priority ranking of power supply equipment based on the normalized power data sequence of the power supply equipment; In response to determining that the position index of the target power supply device in the power supply device priority ranking meets a preset high priority condition, using the first dynamic power threshold as the dynamic power allocation threshold; In response to determining that the location index does not meet the preset high priority condition, the second dynamic power threshold is used as the dynamic power allocation threshold.

[0007] Furthermore, the charging demand characteristic data includes the remaining battery power and the expected charging time; and The generating of the target power supply equipment scheduling instruction includes: Calculating a minimum charging power requirement based on the difference between the remaining battery power and the target battery capacity; Determining whether the dynamic power allocation threshold is greater than the minimum charging power requirement; In response to a judgment result of yes and the expected charging duration being less than a preset emergency charging duration, generating a first scheduling instruction indicating that maximum power charging is allowed; In response to a yes determination and the expected charging duration being greater than or equal to a preset emergency charging duration, a second scheduling instruction indicating that balanced power charging is permitted is generated.

[0008] Furthermore, controlling the target power supply device to perform a charging start / stop operation includes: In response to the target power supply device scheduling instruction being a first scheduling instruction, setting the output power of the target power supply device to the dynamic power allocation threshold; In response to the target power supply equipment scheduling instruction being a second scheduling instruction, performing the following steps: Obtain the working status of adjacent power supply devices in the charging zone where the target power supply device is located; Calculating the average load rate of the adjacent power supply equipment in working state; generating a partition shared power value according to the average load rate and the dynamic power allocation threshold; Determine the product of the partition shared power value and the preset sharing coefficient as the final output power value; The target power supply device is controlled to perform a charging operation according to the final output power value.

[0009] Furthermore, generating the partition shared power value includes: Detecting whether there is an overloaded power supply device in the working state of the adjacent power supply devices, and extracting excess power data of the overloaded power supply device; In response to the presence of an overloaded power supply, perform the following actions: Obtain real-time temperature data and current fluctuation characteristics of overloaded power supply equipment; If the real-time temperature data is lower than a preset safety threshold and the current fluctuation characteristics are stable, the excess power data is included in the available shared power pool; Calculating a weighted sum of an average load rate of the adjacent power supply devices in working state and the available shared power pool to generate a partition shared power value; In response to there being no overloaded power supply, perform the following operations: Obtain the historical load curve of the charging zone and the current grid electricity price information; Predict the power redundancy in future dispatch periods based on historical load curves; The product of the predicted power margin and the preset basic power is determined as the partition shared power value.

[0010] Furthermore, obtaining real-time load data of multiple power supply devices in the charging area includes: Periodically collect power supply equipment operating parameters, including output voltage, current harmonic distortion rate, and power factor, through the power carrier communication protocol; Perform data cleaning on the collected power supply equipment operating parameters: Detect the instantaneous power change gradient in the operating parameters of the power supply equipment. If the instantaneous power change gradient exceeds the preset mutation threshold, mark the power data at the corresponding time point as data to be verified; Obtain the power supply equipment temperature sensor data and environmental humidity data at the time point corresponding to the data to be verified; When the temperature sensor data is within the limit and the ambient humidity data is lower than the preset humidity threshold, the data to be verified is restored to the original data sequence; Based on the cleaned data, topological grouping is performed according to the physical location of the power supply equipment, and real-time load data including the power supply equipment location label and line loss compensation coefficient is generated in combination with the impedance characteristics of the power grid distribution line.

[0011] Furthermore, it also includes: After controlling the charging start and stop operation, continuously monitor the deviation between the actual output power of the target power supply device and the dynamic power allocation threshold; In response to the deviation value continuously exceeding the preset deviation time, the power supply device control loop correction mechanism is triggered: Obtain the power module heat sink temperature and DC bus voltage fluctuation characteristics of the target power supply equipment; If the heat sink temperature is lower than the critical temperature and the voltage fluctuation characteristics are stable, the output power is gradually increased by adjusting the pulse width modulation duty cycle; If the heat sink temperature approaches the critical temperature, the dynamic derating strategy is enabled and the power compensation mode of the adjacent power supply equipment is simultaneously started; When the actual output power converges to the dynamic power allocation threshold, the power adjustment curve and energy consumption characteristics during the correction process are recorded, and the adaptive control parameter library of the power supply equipment is updated.

[0012] Furthermore, it also includes: When a power supply device offline fault is detected, perform the following steps: Obtain the device identification of the faulty power supply device and the last set of operating parameters before it went offline; Locate the distribution line and topology weight value of the device in the regional load distribution information based on the device identification; Recalculate the dynamic power allocation threshold of adjacent power supply devices based on the topology weight value and generate load transfer instructions; Broadcasting the updated dynamic power allocation threshold to the non-offline power supply devices; Start the redundant power module in the partition where the faulty power supply device is located; Generate an offline charging code for the faulty power supply device and store it on the server; In response to the user scanning the verification QR code of the faulty power supply device, the encrypted charging code is transmitted to the faulty power supply device via near field communication; If the device verifies that the encrypted charging code matches the pre-stored offline charging code, the charging port is unlocked and the local charging mode is started.

[0013] Furthermore, the local charging mode includes: Perform charging operations in an offline state and cache charging process data to the device's memory; When the network connection is restored, the cached data is automatically uploaded to the server to generate a billing order; If the network does not recover within the preset time, an operation and maintenance alarm is triggered and the dynamic power sharing mechanism is activated: Analyze the topology of the power distribution circuit to which the faulty power supply equipment belongs; Calculate the remaining available capacity of the circuit, where the remaining available capacity is equal to the total capacity of the circuit minus the sum of the real-time power of all normal power supply devices; The power quota of the faulty power supply device is distributed to the normal power supply devices on the same circuit according to the weight, and their dynamic power allocation thresholds are updated.

[0014] The above embodiments of the present invention have at least the following beneficial effects: 1. By aggregating the real-time load data of multiple power supply devices in the charging area and combining it with the power supply constraint data on the grid side to generate the power supply device scheduling balancing factor, the dynamic allocation of power to the power supply devices is achieved, effectively improving the charging efficiency and solving the problems of energy waste and low charging efficiency caused by the fixed power allocation of power supply devices in the existing technology.

[0015] 2. During peak load periods on the power grid, the power distribution of power supply equipment can be flexibly adjusted based on the power supply constraint data on the grid side and regional load distribution information, thus avoiding grid overload, ensuring the stable operation of the grid, and solving the problem in existing technologies that power supply equipment cannot effectively coordinate with the power supply relationship of the grid.

[0016] 3. When a power supply device fails to operate offline, load transfer and power compensation can be carried out in a timely manner to ensure the normal operation of other power supply devices. At the same time, the local charging mode is unlocked for the faulty power supply device to ensure the user's charging needs, solving the problem of the lack of an effective response mechanism when power supply equipment fails in the existing technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which: Figure 1 A flow chart of an intelligent scheduling and load balancing control method for power supply equipment provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0018] The technical solutions of this application will be described clearly and completely below, in conjunction with the accompanying drawings. It should be understood that the described embodiments represent only a portion of the embodiments of this application, and not all of them. The components of this application, generally described and illustrated in the drawings herein, may be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of the claimed application, but rather merely represents selected embodiments of this application. All other embodiments derived by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application. It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. Furthermore, in the description of this application, the terms "first," "second," etc., are used solely to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0019] like Figure 1 As shown, the present application proposes an intelligent scheduling and load balancing control method for power supply equipment, including: S1. Receive real-time load data of multiple power supply devices in a charging area, aggregate the real-time load data, and generate regional load distribution information; S2. Obtaining grid-side power supply constraint data of the charging area; S3. Generate a power supply equipment scheduling balancing factor based on the regional load distribution information and the grid-side power supply constraint data; S4. Calculating a dynamic power allocation threshold for each power supply device based on the power supply device scheduling balancing factor; S5. In response to detecting a real-time charging request from a target power supply device, acquiring charging demand characteristic data corresponding to the target power supply device; S6. Generate a target power supply equipment scheduling instruction based on the dynamic power allocation threshold and the charging demand characteristic data; S7: In response to determining that the scheduling instruction of the target power supply device indicates that charging is allowed, control the target power supply device to perform a charging start / stop operation.

[0020] Real-time load data refers to the instantaneous power, current, and voltage parameters generated by power supply equipment during operation. This can be achieved by periodically collecting the output voltage, current harmonic distortion, and power factor of power supply equipment using a power carrier communication protocol. By monitoring changes in power supply equipment electrical parameters in real time, this provides a fundamental data source for load balancing control. Aggregation processing involves integrating and analyzing real-time load data from multiple power supply equipment. This can be achieved through data cleansing, topology grouping, and line loss compensation algorithms. By eliminating abnormal data interference and taking into account the impedance characteristics of the power grid's distribution lines, load distribution information accurately reflects the overall power consumption status of the region. Grid-side power supply constraint data refers to the maximum safe power supply capacity that the grid can provide during the current period. This data can be determined by obtaining historical load curve data and combining it with the grid's current maximum allowable load. This ensures that power supply equipment scheduling plans align with the grid's actual power supply capabilities and prevents overload. The power supply equipment scheduling balancing factor is a quantitative parameter that reflects the degree of match between the regional load distribution and the grid's power supply capacity. This factor can be determined by calculating the ratio of the regional load variance to the grid's maximum allowable load and combining it with a preset load tolerance factor. This factor is used to dynamically adjust the priority of power supply equipment power allocation strategies. Among them, the dynamic power allocation threshold refers to the variable power upper limit set for each power supply device based on the real-time grid constraints and regional load status. Specifically, it can be achieved by normalizing the power data sequence of the power supply device and combining it with a priority sorting algorithm, so that the power allocation can meet the global balance requirements and adapt to the operating status of individual power supply devices. Among them, the charging demand characteristic data refers to the battery status and time requirement parameters contained in the user's charging request. Specifically, it can be obtained by analyzing the remaining battery power, the expected charging time and the target battery capacity difference, and is used to match the user's actual charging needs while meeting the grid constraints. Among them, the charging start and stop operation refers to the action of controlling the power supply device to execute power output or interruption according to the dispatch instruction. Specifically, it can be achieved by adjusting the pulse width modulation duty cycle or enabling the dynamic derating strategy to ensure that the power supply device performs power adjustment within the safety threshold. The core innovation of this application lies in the formation of a closed-loop control link from real-time data collection to power allocation execution by building a multi-dimensional data fusion and dynamic feedback mechanism. By generating a power supply equipment dispatch balancing factor, the grid constraints are dynamically linked to the regional load distribution. This is then double-checked with charging demand characteristic data to achieve adaptive adjustment of the power allocation threshold of the power supply equipment in response to the grid status and user demand, effectively coordinating the dynamic balance between the grid's carrying capacity and user charging demand.

[0021] The working process and principle of this application are as follows: first, by receiving real-time load data from multiple power supply devices in the charging area, this data is aggregated and processed to generate regional load distribution information. This step can fully reflect the overall load status of the charging area and provide a data basis for subsequent scheduling decisions.

[0022] Next, the grid-side power supply constraint data for the charging area is obtained. This step takes into account the actual power supply capacity of the grid and ensures that the scheduling plan will not cause grid overload.

[0023] Then, based on regional load distribution information and grid-side power supply constraint data, a power supply equipment dispatch balancing factor is generated. This factor is a key parameter that mathematically links regional load distribution with grid constraints, providing a quantitative basis for dynamic adjustment.

[0024] Based on the PSU scheduling balancing factor, the dynamic power allocation threshold of each PSU is calculated. This step ensures that the power allocation of each PSU takes into account both global balance and individual differences.

[0025] When a real-time charging request is detected from a target power supply device, the system obtains the charging demand characteristic data of the corresponding target power supply device, which contains the user's specific charging demand information.

[0026] Based on the dynamic power allocation threshold and charging demand characteristics, the target power supply equipment dispatch instructions are generated. This step is double-checked to ensure that user needs are met while maintaining system stability.

[0027] Finally, if it is determined that the target power supply device scheduling instruction indicates that charging is allowed, the system controls the target power supply device to execute charging start and stop operations. This forms a complete control chain from data acquisition to execution feedback.

[0028] In this way, the present application realizes closed-loop control of intelligent scheduling of power supply equipment, effectively coordinating the dynamic balance between the grid's carrying capacity and user charging needs.

[0029] As a preferred embodiment, the solution of this application is specifically implemented as follows: In a charging area containing multiple power supply devices, the real-time load data of each power supply device is first periodically collected through the power carrier communication protocol. This data includes parameters such as the output voltage, current, and power of the power supply device.

[0030] After data cleaning and preprocessing, the collected real-time load data is aggregated and analyzed. This analysis uses different algorithms depending on the scenario. To assess the overall load level, a weighted average algorithm (weighted by the percentage of the rated power of the power supply equipment) is used to generate a regional average power indicator. To identify load distribution patterns, a K-means clustering algorithm (using the power supply equipment power value as the feature vector) is used to generate information reflecting the distribution of high- and low-load clusters in the charging area.

[0031] At the same time, the system uses a real-time communication interface with the grid dispatch center to obtain grid-side power supply constraint data for the current charging area. This data may include information such as substation load rate and distribution line capacity.

[0032] Next, the system uses a neural network model (consisting of one input layer, two hidden layers, and one output layer). The input layer receives regional load distribution information (load variance, average power) and grid-side power supply constraint data (maximum allowable load, load tolerance factor). The output layer generates dispatch balancing factors for power supply equipment. Training uses a mean squared error loss function, using grid stability indicators from historical dispatch data as labels. This model dynamically generates dispatch balancing factors based on real-time input data.

[0033] Based on the generated scheduling balancing factor, the system calculates the dynamic power allocation threshold for each power supply device. The calculation process takes into account factors such as the power supply device's rated power and historical utilization rate, and uses an adaptive algorithm to adjust the threshold.

[0034] When a user initiates a charging request, the system obtains the user's charging demand characteristic data, such as remaining battery power, expected charging time, etc. This data is transmitted to the charging management platform through the user's mobile phone application or vehicle system.

[0035] The system uses the CART decision tree algorithm. The root node determines whether the dynamic power allocation threshold is greater than the minimum charging power requirement. If it is satisfied, the child node branches based on the comparison between the expected charging time and the preset emergency time to generate the first or second scheduling instruction. If it is not satisfied, a charge rejection instruction is generated.

[0036] Finally, if the decision allows charging, the system sends control instructions to the target power supply device via industrial Ethernet or 5G network to execute charging start and stop operations. During the charging process, the system continuously monitors the charging status and adjusts the charging power in real time as needed.

[0037] The present application further proposes that real-time load data includes power data of power supply equipment and charging timestamp; before obtaining grid-side power supply constraint data of the charging area, the method also includes: obtaining historical load curve data of the charging area; determining the peak load period of the grid based on the historical load curve data and the charging timestamp; in response to detecting that the current moment belongs to the peak load period of the grid, using the current maximum allowable load value of the grid as the grid-side power supply constraint data; generating a power supply equipment scheduling balancing factor includes: calculating the ratio of the load variance value of the regional load distribution information to the current maximum allowable load value of the grid; and determining the product of the ratio and the preset load tolerance coefficient as the power supply equipment scheduling balancing factor. Among them, the historical load curve data obtains the daily load curve within the past set period through the power grid dispatching system. The set period can be 30 days to 90 days, and the charging timestamp contains the charging start time and duration. The peak load period of the power grid is determined by comparing the average power value of the same time period every day in the historical load curve. For example, the period when the average power exceeds 85% of the total capacity of the region for three consecutive hours is marked as the peak period. The preset load tolerance factor value range is set to 0.8 to 1.2, and the specific value is dynamically adjusted according to the overload protection threshold of the distribution line. The load variance value is obtained by calculating the sum of the squares of the deviations between the power data of each power supply device and the average power of the region, and is used to characterize the degree of dispersion of the charging load in the region. Specifically, after acquiring historical load curve data for the charging area, the charging timestamps are matched with the time coordinates of the historical load curves to identify peak periods with periodic characteristics. When the current time is detected as a peak period, the current maximum allowable load value of the grid is set as the power supply constraint baseline value, which is typically lower than the grid load limit during off-peak periods. The load variance value of the regional load distribution information reflects the degree of dispersion of the power distribution of power supply equipment, and its square root represents the load fluctuation amplitude. The ratio of this fluctuation amplitude to the current maximum allowable load value of the grid is defined as the grid load stress coefficient. By multiplying this ratio by a preset load tolerance coefficient, a dispatch balancing factor is generated that dynamically adjusts the power allocation threshold. For example, when the load variance value reaches 50% of the maximum allowable load value and the tolerance coefficient is 1.0, the dispatch balancing factor is 0.5, triggering load reduction control of the power supply equipment. This constraint generation mechanism, based on the fusion of historical patterns and real-time data, not only avoids the short-sightedness of dispatch strategies caused by relying solely on instantaneous data, but also establishes a dynamic balance model between grid carrying capacity and charging demand through the ratio of variance to maximum load.

[0038] As a preferred embodiment, the solution of this application is specifically implemented as follows: Real-time load data includes power data from power supply devices and charging timestamps. Before obtaining grid-side power supply constraint data for the charging area, historical load curve data for the charging area is first obtained. Based on this historical load curve data and charging timestamps, peak grid load periods are determined. For example, by analyzing the historical load curves for the past 30 days, peak grid load periods between 6:00 PM and 10:00 PM are identified.

[0039] When it is detected that the current moment belongs to the peak load period of the power grid, the current maximum allowable load value of the power grid is used as the power supply constraint data of the power grid side. For example, during the peak period, the current maximum allowable load value of the power grid is set to 10MW.

[0040] The process of generating the power supply equipment scheduling balancing factor includes the following steps: First, the ratio of the load variance value of the regional load distribution information to the current maximum allowable load value of the power grid is calculated. The load variance value reflects the degree of dispersion of the power supply equipment load distribution within the region. Furthermore, the product of this ratio and the preset load tolerance factor is determined as the power supply equipment scheduling balancing factor. For example, if the calculated ratio is 0.05 and the preset load tolerance factor is 1.2, the power supply equipment scheduling balancing factor is 0.06.

[0041] The present application further proposes a method for calculating the dynamic power allocation threshold of each power supply device, including: normalizing the power data sequence of the power supply device according to the power supply device scheduling balancing factor; generating the power supply device priority ranking based on the normalized power data sequence of the power supply device; and using the first or second dynamic power threshold according to whether the location index meets the preset high priority condition.

[0042] Normalization can use the range method to map power data to the 0-1 range. For example, the maximum value in the power data sequence of the power supply equipment is set to 1 and the minimum value is set to 0. Prioritization of power supply equipment can be achieved through a weighted scoring algorithm, in which the power supply equipment scheduling balancing factor is used as a weight coefficient in the calculation. For example, the normalized power value and the balancing factor are weighted and summed at a weight ratio of 0.6:0.4.

[0043] The power supply equipment priority is sorted in descending order by score. The score is calculated as follows: Score = 0.6 × normalized power + 0.4 × scheduling balancing factor. Normalized power = device real-time power / rated power. A higher score gives a higher priority. The default high priority condition is for devices ranked in the top 20% by score.

[0044] The preset high-priority condition is: in a priority queue sorted in descending order by score, for a power supply device with a location index ≤ 20% of the total number of devices, the first dynamic power threshold can be set to 80% of the rated power, and the second dynamic power threshold can be set to 50% of the rated power. The location index can be determined using a binary search method for rapid location determination. For example, the high-priority condition is triggered when the power supply device's sequence number in the sorting order is less than or equal to the total number of devices multiplied by 0.2.

[0045] Specifically, the power data series of power supply devices are normalized using the range method to eliminate power dimension differences among different power supply devices. This normalized data is then fed into a weighted scoring model, which, combined with the power supply device scheduling balancing factor, generates a prioritized list. The top 20% of power supply devices are assigned to a high-priority group, and their dynamic power allocation threshold is raised to 80% of their rated power, ensuring sufficient power resources for high-demand devices. The remaining power supply devices use a threshold of 50% of their rated power, limiting allocation to prevent grid overload. When grid load fluctuates, the power supply device scheduling balancing factor adjusts the weight coefficients in real time, dynamically updating the priority ranking. For example, during peak load periods, the balancing factor increases the suppression weight for low-power power supply devices, placing more power supply devices in the low-priority group and reducing overall power consumption. This dynamic threshold allocation mechanism enables the system to automatically adapt to grid status changes while maintaining load balance, reducing power supply device response time to milliseconds and improving scheduling efficiency by approximately 35% compared to traditional fixed threshold methods.

[0046] As a preferred embodiment, the solution of this application is specifically implemented as follows: When calculating the dynamic power allocation threshold for each PSU, the power data sequence of each PSU is first normalized based on the PSU scheduling balancing factor. Specifically, the real-time power data of each PSU is divided by the rated power of the PSU, and then multiplied by the PSU scheduling balancing factor to obtain the normalized power data sequence.

[0047] Next, based on the normalized power data sequence, a priority ranking of the power supply devices is generated. For example, a weighted average method can be used to weight the normalized power value with factors such as the power supply device's historical usage frequency and user ratings to obtain a comprehensive score for each power supply device, and then sort them from high to low based on the score.

[0048] Then, in response to determining that the position index of the target power supply device in the power supply device priority ranking meets the preset high priority condition, the first dynamic power threshold is used as the dynamic power allocation threshold. For example, the power supply devices in the top 30% of the ranking can be set as high priority, and for these power supply devices, a higher first dynamic power threshold, such as 90% of the rated power, is used.

[0049] Accordingly, in response to determining that the location index does not meet the preset high priority condition, the second dynamic power threshold is used as the dynamic power allocation threshold. For power supply devices with lower priority, a lower second dynamic power threshold is used, such as 70% of the rated power.

[0050] Furthermore, these thresholds can be dynamically adjusted based on the real-time load of the power supply device. For example, if a high-priority power supply device is detected to be underloaded for an extended period, its dynamic power allocation threshold can be temporarily lowered to allocate excess power to other devices with higher charging needs.

[0051] The present application further proposes that the charging demand characteristic data includes the remaining battery power and the expected charging time; generate a target power supply equipment scheduling instruction, including: calculating the minimum charging power requirement based on the difference between the remaining battery power and the target battery capacity; judging whether the dynamic power allocation threshold is greater than the minimum charging power requirement; in response to the judgment result being yes and the expected charging time being less than the preset emergency charging time, generating a first scheduling instruction indicating that maximum power charging is allowed; in response to the judgment result being yes and the expected charging time being greater than or equal to the preset emergency charging time, generating a second scheduling instruction indicating that balanced power charging is allowed. Among them, the calculation of the minimum charging power requirement is achieved by multiplying the difference between the remaining battery power and the target battery capacity by the charging efficiency coefficient per unit time, for example, the difference is multiplied by the power conversion coefficient corresponding to the 0.2C rate. In the comparison and judgment of the dynamic power allocation threshold and the minimum charging power requirement, when the dynamic power allocation threshold is more than 1.2 times the minimum requirement, subsequent hierarchical regulation is allowed. The preset emergency charging time is set as a configurable parameter, for example, it is set to 15 minutes or 30 minutes according to the regional power grid load characteristics. When generating the first scheduling instruction, the maximum power is limited to the smaller value of the dynamic power allocation threshold and the rated power of the power supply equipment; when generating the second scheduling instruction, the balanced power is calculated by the ratio of the dynamic power allocation threshold to the number of current online power supply equipment. Specifically, when a charging request is detected, the difference between the remaining battery charge and the target battery capacity is first used to calculate the minimum charging power requirement, ensuring that the user's vehicle reaches the minimum drivable charge level within a preset timeframe. A comparison of the dynamic power allocation threshold with the minimum requirement determines whether charging is permitted. If the threshold exceeds the minimum requirement, a tiered decision is made based on the expected charging duration. If the expected charging duration is less than the preset emergency duration, indicating an urgent need for recharging, fast charging is permitted at the maximum available power. If the expected charging duration exceeds the preset value, a balanced power allocation strategy is implemented, distributing the available power proportionally across multiple power supply devices. During this process, the dynamic power allocation threshold is dynamically generated based on regional load distribution and grid constraints, ensuring that emergency charging instructions do not overload the grid. In non-emergency scenarios, balanced allocation dynamically adjusts the power of each power supply device to achieve a smooth overall load transition. This ensures a rapid response to emergency requests while avoiding resource monopoly in non-emergency scenarios, improving the accuracy of matching power allocation strategies with actual user needs.

[0052] As a preferred embodiment, the solution of this application is specifically implemented as follows: Charging demand characteristic data includes the remaining battery charge and the expected charging time. When generating the target power supply equipment dispatch instruction, the minimum charging power requirement is first calculated based on the difference between the remaining battery charge and the target battery capacity. For example, if the electric vehicle battery capacity is 60kWh and the current remaining charge is 15kWh, the minimum charging power requirement can be calculated as (60kWh - 15kWh) / 4h = 11.25kW, where 4h is the preset baseline charging time.

[0053] Furthermore, a determination is made as to whether the dynamic power allocation threshold is greater than the minimum charging power requirement. If so, and the desired charging duration is less than the preset emergency charging duration, a first scheduling instruction is generated, indicating that maximum power charging is permitted. For example, if the dynamic power allocation threshold is 15kW, which is greater than the minimum charging power requirement of 11.25kW, and the user's desired charging duration is 2 hours, which is less than the preset emergency charging duration of 3 hours, the system will generate a first scheduling instruction allowing charging at a maximum power of 15kW.

[0054] Conversely, if the judgment result is yes and the expected charging time is greater than or equal to the preset emergency charging time, a second scheduling instruction is generated to indicate that balanced power charging is allowed. For example, if the user's expected charging time is 4 hours, which is greater than the preset 3-hour emergency charging time, the system will generate a second scheduling instruction to allow 11.25kW balanced power charging.

[0055] The present application further proposes controlling the target power supply device to perform charging start and stop operations, including: in response to the target power supply device scheduling instruction being a first scheduling instruction, setting the output power of the target power supply device to a dynamic power allocation threshold; in response to the target power supply device scheduling instruction being a second scheduling instruction, executing the following steps: obtaining the working status of adjacent power supply devices in the charging partition where the target power supply device is located; calculating the average load rate of the working status of the adjacent power supply devices; generating a partition shared power value based on the average load rate and the dynamic power allocation threshold; determining the product of the partition shared power value and the preset sharing coefficient as the final output power value; and controlling the target power supply device to perform charging operations according to the final output power value.

[0056] Among them, the acquisition of the working status of adjacent power supply equipment can be achieved by periodically collecting real-time power data of adjacent power supply equipment through the power carrier communication protocol, for example, collecting the output voltage and current harmonic distortion rate once every 30 seconds. The calculation of the average load rate adopts the sliding window weighted average algorithm, in which the data weights of the last three sampling cycles are set to 0.6, 0.3, and 0.1 to ensure accurate perception of real-time load characteristics. The generation of partition shared power value needs to be combined with the linear combination of dynamic power allocation threshold and average load rate, for example, through the formula

[0057] in is the dynamic power allocation threshold, is the average load factor, is the baseline power. α is in the range of 0.7-0.9, and β is in the range of 0.3-0.5. The preset sharing coefficient is dynamically adjusted based on fluctuations in grid electricity prices, set to 0.8 during peak electricity price periods and 0.95 during off-peak periods, preserving grid safety margins while improving resource utilization.

[0058] Specifically, when a second scheduling instruction is detected, the operating parameters of adjacent devices in the partition where the target power supply device is located, including power output values ​​and temperature sensor data, are first obtained through the topology grouping mechanism. The collection of the operating status of adjacent power supply devices must be linked to the power supply device priority sorting mechanism in the preceding claims. For example, the load data of high-priority power supply devices is assigned a weight factor of 1.2 when calculating the average load rate. The calculation of the average load rate must exclude abnormal offline devices and only count the data of power supply devices that are operating normally for three consecutive sampling periods. In the process of generating the partition shared power value, the dynamic power allocation threshold is used as a basic constraint condition and dynamically superimposed with the real-time average load rate. For example, when the average load rate exceeds 75%, a 10% derating of the shared power value is automatically triggered. The determination of the final output power value requires a secondary constraint through a preset sharing coefficient. This coefficient is dynamically adjusted according to the impedance characteristics of the distribution line. For example, for every 0.1Ω / km increase in line impedance, the sharing coefficient is correspondingly reduced by 0.05. Through the synergistic effect of the above steps, while ensuring the power supply constraints on the grid side, dynamic load balancing is achieved between adjacent power supply equipment, avoiding the risk of distribution line tripping caused by local overload, and at the same time enabling charging power distribution to have dynamic adjustment capabilities at the minute level.

[0059] As a preferred embodiment, the solution of this application is specifically implemented as follows: When controlling the target power supply device to perform the charging start and stop operation, first respond to the target power supply device scheduling instruction. When the scheduling instruction is the first scheduling instruction, set the output power of the target power supply device to the dynamic power allocation threshold. When the scheduling instruction is the second scheduling instruction, perform the following steps: Obtain the operating status of adjacent power supply devices in the charging zone where the target power supply device is located. For example, the operating parameters such as current, voltage, and power of adjacent power supply devices can be collected in real time through the communication network.

[0060] Calculate the average load factor of the adjacent power supply devices in their working state. Specifically, the average load factor can be obtained by taking the arithmetic average of the ratio of the actual output power of each adjacent power supply device to its rated power.

[0061] The partition shared power value is generated based on the average load rate and the dynamic power allocation threshold. For example, the product of the average load rate and the dynamic power allocation threshold can be used as the basis for the partition shared power value, and then adjusted appropriately based on the actual load situation.

[0062] The final output power value is determined by multiplying the zone shared power value by a preset sharing coefficient. The preset sharing coefficient can be determined based on factors such as historical data of the charging zone and grid capacity, and is used to balance charging demand and grid carrying capacity.

[0063] Finally, the target power supply device is controlled to perform charging operations according to the final output power value. The control system of the power supply device can adjust the charging current and voltage based on the calculated final output power value to achieve precise power output control.

[0064] The present application further proposes generating a partition shared power value, including: detecting whether there is an overloaded power supply device in the working state of adjacent power supply devices, and extracting the excess power data of the overloaded power supply device; in response to the existence of an overloaded power supply device, performing the following operations: obtaining the real-time temperature data and current fluctuation characteristics of the overloaded power supply device; if the real-time temperature data is lower than a preset safety threshold and the current fluctuation characteristics are stable, incorporating the excess power data into the available shared power pool; calculating the average load rate of the working state of the adjacent power supply equipment and the weighted sum of the available shared power pool to generate a partition shared power value; in response to the absence of an overloaded power supply device, performing the following operations: obtaining the historical load curve of the charging partition and the power grid electricity price information of the current period; predicting the power redundancy of the future scheduling period based on the historical load curve; and determining the product of the predicted power redundancy and the preset basic power as the partition shared power value.

[0065] Overloaded power supply equipment is detected by real-time monitoring of its output current to see if it exceeds the rated current threshold. The rated current threshold can be set to 105%-120% of the equipment nameplate value. Preset safety thresholds for real-time temperature data are based on the heat dissipation performance of the power supply equipment. For example, a temperature sensor is used to collect heat sink temperature, and a safe state is determined when the temperature is below 80°C. Current fluctuations are considered stable if the amplitude is within 5% and the frequency is below 10Hz. The weighting coefficient of the available shared power pool is dynamically adjusted based on the remaining life of the power supply equipment and line impedance. For example, the weight of equipment with a remaining life of less than 50% of its design life is reduced by 30%. Historical load curves are predicted using time series analysis, and the power redundancy calculation period can be set to 15 minutes to 1 hour in the future. The preset base power is determined based on the capacity of the distribution line and the current grid electricity price tier. For example, during periods of low electricity prices, the base power is increased to 90% of the line capacity.

[0066] Specifically, when an overloaded device is detected among adjacent power supply equipment, its excess power data is first extracted. For example, if a power supply has a rated power of 60 kW but actually outputs 70 kW, the excess power is 10 kW. The system then obtains the device's real-time temperature data. If the temperature sensor indicates a radiator temperature of 75°C and a current fluctuation of 3%, the device is deemed safe for sharing and the 10 kW excess power is added to the available shared power pool. Considering the average load factor of 60% for adjacent power supply equipment and assuming a weight of 0.4 for the available shared power pool, the partition's shared power value is calculated as 60% × 0.6 + 10 kW × 0.4 = 36 kW + 4 kW = 40 kW. This calculation method maintains basic load balancing capabilities while effectively utilizing excess power resources. In a scenario without overloaded devices, by analyzing the load curves of the charging partition during the same period over the past week, a power margin of 20 kW is predicted for the next half hour. Given the current off-peak electricity price, the base power is set at 80% of the line capacity, resulting in a partition-shared power value of 20 kW × 80% = 16 kW. Through the above mechanism, the dual goals of risk avoidance and resource reuse can be achieved in overload scenarios, while the economic efficiency of the power grid can be improved and load fluctuations can be adapted in normal scenarios.

[0067] As a preferred embodiment, the solution of this application is specifically implemented as follows: Generating the partition shared power value includes the following steps: First, check whether any adjacent power supply devices are overloaded. This can be determined by comparing the actual output power of the power supply device with its rated power. If the actual output power exceeds the rated power, the device is considered overloaded. For any detected overloaded power supply device, its excess power data is extracted.

[0068] If there is an overloaded power supply device, do the following: Acquire real-time temperature data and current fluctuation characteristics of overloaded power supply equipment. Temperature data can be collected using a built-in temperature sensor, and current fluctuation characteristics can be obtained by analyzing the time domain waveform of the power supply equipment's output current.

[0069] Determine whether the real-time temperature data is below the preset safety threshold and whether the current fluctuation characteristics are stable. The safety threshold can be set to 80°C, and the criterion for judging the stable current fluctuation characteristics can be that the fluctuation amplitude does not exceed 5% of the rated current.

[0070] If the above conditions are met, the excess power data is included in the available shared power pool. The available shared power pool can be implemented using a queue data structure, and new excess power data is queued in sequence.

[0071] The partition's shared power value is calculated by adding the weighted sum of the average load factor of adjacent PSUs and the available shared power pool. The weighting factor can be dynamically adjusted based on factors such as the PSU's physical location and line loss.

[0072] If there is no overloaded power supply equipment, do the following: Obtain the historical load curve for the charging zone and the current grid electricity price information. The historical load curve can use the data from the last 7 days, and the grid electricity price information can be obtained through the real-time query interface of the power company.

[0073] Predict the power redundancy during future dispatch periods based on historical load curves. This can be done using time series analysis or machine learning algorithms, such as ARIMA models or LSTM neural networks.

[0074] The product of the predicted power margin and the preset base power is determined as the partition shared power value. The preset base power can be set according to the overall capacity of the charging partition.

[0075] The present application further proposes obtaining real-time load data of multiple power supply devices in the charging area, including: periodically collecting the operating parameters of the power supply equipment, including output voltage, current harmonic distortion rate and power factor, through the power carrier communication protocol; performing data cleaning processing on the collected operating parameters of the power supply equipment: detecting the instantaneous power change gradient in the operating parameters of the power supply equipment, and if the instantaneous power change gradient exceeds a preset mutation threshold, marking the power data at the corresponding time point as data to be verified; obtaining the power supply equipment temperature sensor data and ambient humidity data at the time point corresponding to the data to be verified; when the temperature sensor data is within the limit and the ambient humidity data is lower than the preset humidity threshold, restoring the data to be verified to the original data sequence; based on the cleaned data, topological grouping is performed according to the physical location of the power supply equipment, and real-time load data including the power supply equipment location label and the line loss compensation coefficient is generated in combination with the impedance characteristics of the power grid distribution line.

[0076] The power carrier communication protocol can use the PRIME standard or the G3-PLC protocol for data collection, and the periodic collection interval can be set to 5-30 seconds. The preset mutation threshold is dynamically adjusted based on the rated power of the power supply equipment. For example, a power supply with a rated power of 120kW can be set to a mutation threshold of 15kW / s. Temperature sensor data out-of-limit judgment is based on the equipment's safe operating temperature range, and the ambient humidity threshold can be set to 85%RH. Topology grouping is spatially matched between the power supply equipment's GPS coordinates and the power grid's GIS system. Line impedance characteristics are calculated by calculating the resistivity and length of the distribution line, and the line loss compensation coefficient can be derived based on Ohm's law.

[0077] Specifically, when collecting multiple physical quantity parameters through the power carrier communication protocol, the output voltage is used to calculate the line voltage drop, the current harmonic distortion rate reflects the power quality, and the power factor is used to correct the conversion error between apparent power and active power. When it is detected that the power change gradient exceeds the threshold, the temperature and humidity data are combined to verify the cause of the abnormality: if the temperature is within the limit and the humidity is normal, it is determined to be a pseudo-anomaly caused by environmental interference, and the data to be verified is reinserted into the original sequence. After the cleaned data is grouped by physical location, each group of data is superimposed with the compensation coefficient calculated based on the corresponding line impedance characteristics. For example, when the line impedance is 0.5Ω, the compensation coefficient can be set to (1+0.5*I 2 / P), where I is the line current and P is the transmitted power. The resulting real-time load data eliminates noise interference and corrects for line loss errors, ensuring that regional load distribution information accurately reflects actual power demand and provides reliable input for subsequent dynamic power allocation.

[0078] As a preferred embodiment, the solution of this application is specifically implemented as follows: Multiple power supply devices are installed within the charging area, each equipped with a power carrier communication module, temperature sensor, and humidity sensor. Using the power carrier communication protocol, the system periodically collects power supply device operating parameters, including output voltage, current harmonic distortion, and power factor. The collection period can be set to 5 seconds.

[0079] Perform data cleaning on the collected operating parameters. First, detect the instantaneous power change gradient in the power supply equipment operating parameters. For example, calculate the power change rate between two adjacent sampling points. If the change rate exceeds a preset mutation threshold (e.g., 20% / second), mark the power data at the corresponding time point as pending verification data.

[0080] Next, the system obtains the power supply device temperature sensor data and ambient humidity data at the time point corresponding to the data to be verified. If the temperature sensor data does not exceed the preset limit (e.g., 80°C) and the ambient humidity data is below the preset humidity threshold (e.g., 85% RH), the system restores the data to the original data sequence. Otherwise, the data point is discarded.

[0081] Based on the cleaned data, the system topologically groups power supply devices by their physical location. For example, power supply devices within a 50-meter distance are grouped together. Incorporating the impedance characteristics of the power grid's distribution lines, real-time load data is generated for each power supply device, including a location tag and a line loss compensation coefficient. The location tag can be expressed in latitude and longitude coordinates, and the line loss compensation coefficient is calculated based on the distance from the power supply device to the distribution transformer and the line type.

[0082] The present application further proposes to continuously monitor the deviation between the actual output power of the target power supply device and the dynamic power allocation threshold after controlling the charging start and stop operation; in response to the deviation value (actual output power - absolute value of the dynamic power allocation threshold) continuously exceeding the preset deviation duration, trigger the power supply device control loop correction mechanism: obtain the power module heat sink temperature and DC bus voltage fluctuation characteristics of the target power supply device; if the heat sink temperature is lower than the critical temperature and the voltage fluctuation characteristics are stable, gradually increase the output power by adjusting the pulse width modulation duty cycle; if the heat sink temperature is close to the critical temperature, enable the dynamic derating strategy and synchronously start the power compensation mode of the adjacent power supply device; when the actual output power converges to the dynamic power allocation threshold, record the power adjustment curve and energy consumption characteristics during the correction process, and update the adaptive control parameter library of the power supply device.

[0083] The preset deviation duration for deviation monitoring can be set between 5 and 30 seconds. For example, a 10-second threshold is used in scenarios with frequent grid load fluctuations. The critical temperature of the power module heat sink is set based on the upper temperature limit of the IGBT module. For example, a temperature threshold of 120°C is used, and a derating strategy is triggered when the temperature reaches 115°C. The DC bus voltage fluctuation characteristics can be determined by calculating the voltage root mean square deviation. For example, a fluctuation amplitude not exceeding ±2% of the rated voltage is considered stable. When adjusting the pulse width modulation duty cycle, each adjustment is controlled within the range of 1%-3%, for example, gradually increasing the output power in 2% steps. The dynamic derating strategy can be implemented by reducing the power by 5% every 5 seconds. At the same time, power compensation for adjacent power supply units is achieved by redistributing the power quota. For example, the power reduction of the derating unit is proportionally distributed among the three adjacent power supply units.

[0084] Specifically, after the power supply equipment starts and stops, the actual output power data is collected in real time via current sensors and voltage sampling circuits. The difference between the actual output power and the dynamic power allocation threshold is calculated. If the deviation value continuously exceeds the allowable range for a preset period of time, a correction mechanism is triggered. At this time, the temperature sensor collects the temperature data of the power module heat sink, and the voltage detection module analyzes the DC bus voltage fluctuation characteristics. If the temperature does not exceed the safety threshold and the voltage fluctuation is within the stable range, the output power is gradually increased by adjusting the duty cycle of the PWM drive signal, for example, by increasing the duty cycle by 2% at a time and maintaining it for 30 seconds to observe the effect. If the temperature approaches the critical value, a derating strategy is immediately initiated, for example, reducing the current power by 10%. Simultaneously, power compensation instructions are sent to adjacent power supply equipment, causing them to increase their power output by 3.3% to maintain overall grid load balance. During the correction process, the power adjustment curve is recorded as a time-power correspondence table, and the energy consumption characteristics are integrated to obtain the energy consumption per unit time. Finally, the historical parameters such as response time, temperature change rate, and voltage recovery speed in the adaptive control parameter library are updated to optimize the decision algorithm of the subsequent control strategy. Therefore, when power deviation occurs, the output can be dynamically adjusted according to the equipment status, and the grid stability can be maintained through a collaborative compensation mechanism. At the same time, the accumulated correction data provides a basis for continuous optimization of the system.

[0085] As a preferred embodiment, the solution of the present application is specifically implemented as follows: after the power supply device completes the charging start and stop operation, the actual output power data is collected in real time through the Hall sensor and the voltage sampling circuit, and the difference is calculated with the dynamic power allocation threshold. When the power deviation value exceeds the preset deviation for 5 seconds continuously, the correction mechanism is triggered. At this time, the temperature sensor collects the temperature of the power module heat sink, and the ripple coefficient of the DC bus voltage is captured by the high-speed ADC module. If the heat sink temperature is lower than the critical temperature of 85°C and the voltage ripple coefficient is maintained within the range of ±2%, the duty cycle of the pulse width modulation signal is increased in steps of 0.5%, so that the output power gradually approaches the target threshold. When the temperature reaches 80°C, the dynamic derating strategy is activated, and the output power is reduced at a rate of 1kW per second. At the same time, a power compensation instruction is sent to the adjacent power supply equipment to increase its output power by the corresponding derating amount. After the correction is completed, the time-power correspondence and temperature rise curve during the power adjustment process are stored in the database, and the adaptive control parameters are optimized based on the gradient descent algorithm.

[0086] The present application further proposes that when an offline failure of a power supply device is detected, the following steps are performed: obtaining the device identification of the faulty power supply device and the last set of operating parameters before going offline; locating the distribution line and topology weight value to which it belongs in the regional load distribution information based on the device identification; recalculating the dynamic power allocation threshold of the adjacent power supply device according to the topology weight value, and generating a load transfer instruction; broadcasting the updated dynamic power allocation threshold to the non-offline power supply devices; starting the redundant power module of the partition where the faulty power supply device is located; generating an offline charging code corresponding to the faulty power supply device and storing it on the server side; in response to the user scanning the verification QR code of the faulty power supply device, transmitting the encrypted charging code to the faulty power supply device side via near-field communication; if the device side verifies that the encrypted charging code matches the pre-stored offline charging code, unlocking the charging interface and starting the local charging mode.

[0087] The device identifier can be a composite identifier consisting of the distribution line code and the device's physical address, such as a string consisting of an 8-bit line code and a 6-bit address code. Pre-offline operating parameters can include power output values, temperature sensor readings, and communication status flags. The power output value acquisition window can be set to between 30 seconds and 5 seconds before the fault is triggered. Topology weights can be calculated based on the distribution line capacity. For example, if the capacity of the line where the faulty device resides accounts for 15% of the total regional capacity, its topology weight can be set to 0.15.

[0088] The generation of load transfer instructions can be combined with the real-time load rate of adjacent devices. For example, when the current load rate of an adjacent device is 60%, its dynamic power allocation threshold can be increased to 1.2 times the original threshold. The startup of the redundant power module can call the preset backup power group, whose capacity configuration can be 10%-15% of the partition rated power. The generation of offline charging codes can use a hash algorithm to encrypt the device identification and timestamp, such as the 32-byte data generated by the SHA-256 algorithm. The near-field communication transmission distance can be limited to within 10 cm, and the verification process of the encrypted charging code can include digital signature verification and time validity check, where the time window can be set to be valid within 24 hours after generation.

[0089] Specifically, when a power supply device offline fault is detected, the acquisition of the device identification enables the system to trace its topological position in the power distribution network. By locating the distribution line to which it belongs and the topological weight value, the power gap caused by the device being offline can be accurately calculated. The dynamic power allocation threshold of adjacent power supply devices is recalculated based on the topological weight value. For example, the original power quota of the faulty device is allocated to the devices on the same line according to the weight ratio, so that 15% of the quota of the faulty device in line A is allocated to devices B and C on the same line. When the weight of device B is 0.6 and the weight of device C is 0.4, B will receive an additional quota of 9% and C will receive an additional quota of 6%. The broadcast of the load transfer instruction triggers the online devices to synchronously adjust the power threshold to ensure regional load balancing.

[0090] The startup of the redundant power module compensates for the power gap by connecting a parallel backup power supply group. For example, when the faulty device originally bears a 50kW load, the startup capacity of the redundant module is 7.5kW to maintain the basic power supply. The generation and storage of the offline charging code establishes a trusted association between the offline device and the server. After the user scans the QR code, it triggers the near-field communication to transmit encrypted data. The device side realizes two-way authentication by comparing the pre-stored hash value. The startup of the local charging mode can call the pre-offline operating parameters in the device cache. For example, charging is performed at 80% of the rated power last recorded before the failure, and the charging time and energy consumption data are recorded in the local storage. When the network is restored, the cached data is uploaded to the server to generate a complete billing order. If it is not restored after the timeout, the operation and maintenance response is triggered to realize the closed loop of fault handling.

[0091] As a preferred embodiment, the solution of the present application is specifically implemented as follows: when a power supply device fails offline, the unique identification code of the faulty device and the operating parameters recorded before offline, including output power, current value and temperature data, are first obtained through the device management platform. Based on the identification code, the physical connection relationship of the distribution line to which it belongs is queried in the power grid topology database to determine the weight coefficient of the device in the regional load distribution. The power upper limit value of the adjacent power supply equipment is dynamically adjusted according to the weight coefficient. For example, 30% of the load borne by the original faulty device is distributed to the three adjacent devices in the same circuit according to the line impedance ratio. The updated power allocation threshold is broadcast to all online devices through the power carrier communication protocol, triggering the automatic adjustment program of its power module. At the same time, the backup power unit in the distribution cabinet where the faulty device is located is activated. The unit is connected to the power grid through a parallel contactor to maintain the total capacity of the circuit. The server generates an encrypted charging code containing the device serial number and timestamp, and writes it into the near-field communication chip. After the user uses a mobile terminal to scan the device's QR code, the terminal transmits the encrypted charging code to the power supply equipment control board via Bluetooth. After the control board compares the pre-stored hash value to complete the verification, it drives the electromagnetic lock to release the charging gun plug and switches to local control mode, performing charging operations based on the cached historical power data.

[0092] This application further proposes a local charging mode including: performing charging operations in an offline state and caching charging process data to the device-side memory; when the network connection is restored, automatically uploading the cached data to the server to generate a billing order; if the network is not restored within the preset time, triggering the operation and maintenance alarm instruction and activating the dynamic power sharing mechanism: parsing the topology of the distribution circuit to which the faulty power supply equipment belongs; calculating the remaining available capacity of the circuit, which is equal to the total capacity of the circuit minus the sum of the real-time power of all normal power supply equipment; allocating the power quota of the faulty power supply equipment to the normal power supply equipment on the same circuit according to the weight, and updating its dynamic power allocation threshold.

[0093] When charging is performed offline, the device-side memory can utilize non-volatile memory chips, such as ferroelectric memory or flash memory, with data retention set to at least 72 hours. Cached data includes charging start and end times, power curves, and user identification. Data can be stored in encrypted binary files to prevent tampering. Automatic upload upon network recovery is triggered by a heartbeat packet detection mechanism. The server-side billing order generation logic includes timestamp verification and a power integration algorithm, for example, accumulating charging volume in minutes. The preset duration can be set to 15-30 minutes based on network stability, with a countdown triggered by a hardware timer. Distribution circuit topology analysis is based on a pre-stored database of line connection relationships and combines current direction sensor data to verify actual physical connections. Remaining available capacity calculation utilizes a real-time power sampling frequency, for example, collecting instantaneous power values ​​from healthy power supply devices once per second, filtering them through a sliding window, and summing them. The weight allocation mechanism is determined based on the ratio of the power supply device's rated power to its current load factor. For example, a device with a rated power of 20kW and a load factor of 60% receives a higher weight. A dynamic programming algorithm is used to ensure that the power addition does not exceed the circuit capacity threshold.

[0094] Specifically, when the power supply equipment enters the offline state, the charging interface maintains the basic charging function through the local control unit, and the voltage, current and time data during the charging process are encrypted and written to the device-side memory. If the network is restored within the preset time, the stored data is fully uploaded to the server through the breakpoint resume protocol. The server parses the data and generates a billing order containing the charging period and power consumption details. When the network interruption lasts for more than the preset time, the hardware alarm module sends an alarm signal containing the device location information to the operation and maintenance terminal, and activates the distribution circuit analysis program at the same time. At this time, the line topology analysis module calls the pre-stored electrical connection relationship map to identify the upstream and downstream nodes of the circuit where the faulty equipment is located, and the current sensor collects the power data of each node in real time. The remaining capacity calculation module uses the total capacity of the circuit as the base value. For example, a 400A distribution circuit is converted to 280kW. The allocable capacity is obtained by deducting the total real-time power of all online devices. When redistributing power quotas, the weight coefficients are dynamically adjusted based on the rated power and load status of the equipment. For example, the 30kW quota of the faulty equipment is distributed in a ratio of 1:2:3 to three equipment in the same circuit with a load rate below 80%, increasing their dynamic power allocation thresholds by 5kW, 10kW, and 15kW, respectively. This process broadcasts update instructions to the controllers of each power supply device via the CAN bus to ensure that power adjustments are executed synchronously and the total load of the circuit is always below the safety limit. As a result, the power resources released during the offline fault are reintegrated and utilized, avoiding idle distribution capacity and preventing overload of adjacent equipment. At the same time, the overall balance of the system is maintained through dynamic threshold updates.

[0095] As a preferred embodiment, the solution of the present application is specifically implemented as follows: when the power supply device is in an offline state, the charging operation can still be performed continuously, and the voltage, current and time data generated during the charging process are encrypted and stored in the built-in flash memory module of the device. After the network link between the power supply device and the control center is rebuilt, the data in the storage module is transmitted to the billing server via the SSL / TLS protocol. The server parses the data packet to generate a billing order containing the charging start and end time and energy consumption value. If the network interruption lasts for more than 24 hours, the system activates the sound and light alarm device and sends a work order request to the operation and maintenance terminal, and starts the distribution circuit topology analysis program at the same time. The program calls the power grid GIS database to obtain the circuit connection relationship, calculates the remaining capacity of the circuit based on the real-time collected power data of the normal power supply equipment, and distributes the rated power of the offline power supply equipment to the online equipment in the same circuit according to the inverse ratio of the impedance of each node. The updated power threshold is sent to each power supply equipment controller via power carrier communication.

[0096] Through the above technical solutions, this application achieves dual guarantees for offline charging service continuity and grid capacity utilization efficiency. The topology-driven dynamic power redistribution mechanism effectively avoids idle distribution circuit capacity and ensures that the power transfer process complies with the physical constraints of the grid. The combination of offline charging code verification and encrypted transmission of cached data ensures the security and integrity of billing data while maintaining service availability.

[0097] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A method for intelligent scheduling and load balancing control of power supply equipment, characterized in that: include: Receive real-time load data of multiple power supply devices in the charging area, aggregate the real-time load data, and generate regional load distribution information; Obtaining grid-side power supply constraint data of the charging area; generating a power supply equipment scheduling balancing factor according to the regional load distribution information and the grid-side power supply constraint data; Calculating a dynamic power allocation threshold for each power supply device based on the power supply device scheduling balancing factor; In response to detecting a real-time charging request from a target power supply device, acquiring charging demand characteristic data corresponding to the target power supply device; generating a target power supply equipment scheduling instruction according to the dynamic power allocation threshold and the charging demand characteristic data; In response to determining that the scheduling instruction of the target power supply device indicates that charging is allowed, the target power supply device is controlled to perform a charging start / stop operation.

2. The method according to claim 1, characterized in that The real-time load data includes power data of the power supply device and a charging timestamp; before obtaining the grid-side power supply constraint data of the charging area, the method further includes: Obtaining historical load curve data of the charging area; Determining a peak load period of a power grid according to the historical load curve data and the charging timestamp; In response to detecting that the current moment belongs to the power grid load peak period, using the current maximum allowable load value of the power grid as power supply constraint data on the power grid side; Generating the power supply equipment scheduling balancing factor includes: Calculating a ratio of a load variance value of the regional load distribution information to a current maximum allowable load value of the power grid; The product of the ratio and a preset load tolerance coefficient is determined as a power supply equipment scheduling balancing factor.

3. The method according to claim 2, characterized in that Calculating the dynamic power allocation threshold of each power supply device includes: Normalizing the power data sequence of the power supply equipment according to the power supply equipment scheduling balancing factor; Generate a priority ranking of power supply equipment based on the normalized power data sequence of the power supply equipment; In response to determining that the position index of the target power supply device in the power supply device priority ranking meets a preset high priority condition, using the first dynamic power threshold as the dynamic power allocation threshold; In response to determining that the location index does not meet the preset high priority condition, the second dynamic power threshold is used as the dynamic power allocation threshold.

4. The method according to claim 3, characterized in that The charging demand characteristic data includes the remaining battery power and the expected charging time; as well as The generating of the target power supply equipment scheduling instruction includes: Calculating a minimum charging power requirement based on the difference between the remaining battery power and the target battery capacity; Determining whether the dynamic power allocation threshold is greater than the minimum charging power requirement; In response to a yes determination and the expected charging duration being less than a preset emergency charging duration, generating a first scheduling instruction indicating that maximum power charging is allowed; In response to a yes determination and the expected charging duration being greater than or equal to a preset emergency charging duration, a second scheduling instruction indicating that balanced power charging is permitted is generated.

5. The method according to claim 4, characterized in that The controlling the target power supply device to perform a charging start / stop operation includes: In response to the target power supply device scheduling instruction being a first scheduling instruction, setting the output power of the target power supply device to the dynamic power allocation threshold; In response to the target power supply equipment scheduling instruction being a second scheduling instruction, performing the following steps: Obtain the working status of adjacent power supply devices in the charging zone where the target power supply device is located; Calculating the average load rate of the adjacent power supply equipment in working state; generating a partition shared power value according to the average load rate and the dynamic power allocation threshold; Determine the product of the partition shared power value and the preset sharing coefficient as the final output power value; The target power supply device is controlled to perform a charging operation according to the final output power value.

6. The method according to claim 5, characterized in that Generating the partition shared power value includes: Detecting whether there is an overloaded power supply device in the working state of the adjacent power supply devices, and extracting excess power data of the overloaded power supply device; In response to the presence of an overloaded power supply, perform the following actions: Obtain real-time temperature data and current fluctuation characteristics of overloaded power supply equipment; If the real-time temperature data is lower than a preset safety threshold and the current fluctuation characteristics are stable, the excess power data is included in the available shared power pool; Calculating a weighted sum of an average load rate of the adjacent power supply devices in working state and the available shared power pool to generate a partition shared power value; In response to there being no overloaded power supply, perform the following operations: Obtain the historical load curve of the charging zone and the current grid electricity price information; Predict the power redundancy in future dispatch periods based on historical load curves; The product of the predicted power margin and the preset basic power is determined as the partition shared power value.

7. The method according to claim 1, characterized in that The acquiring of real-time load data of multiple power supply devices in the charging area includes: Periodically collect power supply equipment operating parameters, including output voltage, current harmonic distortion rate, and power factor, through the power carrier communication protocol; Perform data cleaning on the collected power supply equipment operating parameters: Detect the instantaneous power change gradient in the operating parameters of the power supply equipment. If the instantaneous power change gradient exceeds the preset mutation threshold, mark the power data at the corresponding time point as data to be verified; Obtain the power supply equipment temperature sensor data and environmental humidity data at the time point corresponding to the data to be verified; When the temperature sensor data is within the limit and the ambient humidity data is lower than the preset humidity threshold, the data to be verified is restored to the original data sequence; Based on the cleaned data, topological grouping is performed according to the physical location of the power supply equipment, and real-time load data including the power supply equipment location label and line loss compensation coefficient is generated in combination with the impedance characteristics of the power grid distribution line.

8. The method according to claim 1, characterized in that Also includes: After controlling the charging start and stop operation, continuously monitor the deviation between the actual output power of the target power supply device and the dynamic power allocation threshold; In response to the deviation value continuously exceeding the preset deviation duration, a power supply device control loop correction mechanism is triggered: Obtain the power module heat sink temperature and DC bus voltage fluctuation characteristics of the target power supply equipment; If the heat sink temperature is lower than the critical temperature and the voltage fluctuation characteristics are stable, the output power is gradually increased by adjusting the pulse width modulation duty cycle; If the heat sink temperature approaches the critical temperature, the dynamic derating strategy is enabled and the power compensation mode of the adjacent power supply equipment is simultaneously started; When the actual output power converges to the dynamic power allocation threshold, the power adjustment curve and energy consumption characteristics during the correction process are recorded, and the adaptive control parameter library of the power supply equipment is updated.

9. The method according to claim 1, characterized in that Also includes: When a power supply device offline fault is detected, perform the following steps: Obtain the device identification of the faulty power supply device and the last set of operating parameters before it went offline; Locate the distribution line and topology weight value of the device in the regional load distribution information based on the device identification; Recalculate the dynamic power allocation threshold of adjacent power supply devices based on the topology weight value and generate load transfer instructions; Broadcasting the updated dynamic power allocation threshold to the non-offline power supply devices; Start the redundant power module in the partition where the faulty power supply device is located; Generate an offline charging code for the faulty power supply device and store it on the server; In response to the user scanning the verification QR code of the faulty power supply device, the encrypted charging code is transmitted to the faulty power supply device via near field communication; If the device verifies that the encrypted charging code matches the pre-stored offline charging code, the charging port is unlocked and the local charging mode is started.

10. The method according to claim 9, characterized in that The local charging mode includes: Perform charging operations in an offline state and cache charging process data to the device's memory; When the network connection is restored, the cached data is automatically uploaded to the server to generate a billing order; If the network does not recover within the preset time, an operation and maintenance alarm is triggered and the dynamic power sharing mechanism is activated: Analyze the topology of the power distribution circuit to which the faulty power supply equipment belongs; Calculate the remaining available capacity of the circuit, where the remaining available capacity is equal to the total capacity of the circuit minus the sum of the real-time power of all normal power supply devices; The power quota of the faulty power supply device is distributed to the normal power supply devices on the same circuit according to the weight, and their dynamic power allocation thresholds are updated.

Images