Method and system for feeding electric quantity of power supply back to power grid, electronic equipment and storage medium

Abstract

The invention provides a method and equipment, and relates to the technical field of energy management and electric power conversion, and the method comprises the steps: calculating the maximum feedback electric quantity and the optimal feedback duration according to the identity information of a feedback equipment node and a feedback capability parameter, and generating a corresponding feedback capability model; inputting the historical power consumption data, the current environment data and the regional activity data in a preset historical time period into a preset machine learning model, and outputting a power grid load curve in a future preset time period; according to a power grid load curve and a feedback capability model, dividing feedback equipment nodes into a plurality of feedback groups and distributing peak-shifting feedback time periods, and generating an initial feedback distribution scheme; and calculating a specific feedback parameter of each feedback equipment node based on the priority sequence and the feedback time window, and distributing the specific feedback parameter to the corresponding feedback equipment node. By implementing the method, power grid voltage and frequency fluctuation caused by excessive concentration of feedback electric quantity can be reduced, and the stability of power grid operation is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy management and power conversion, and particularly relates to a method and system for power supply power feedback to a power grid, an electronic device and a storage medium. BACKGROUND

[0002] With the rapid development of new energy technology, a large number of distributed energy devices (such as photovoltaic power generation systems, energy storage devices, etc.) are connected to the power grid. These distributed energy devices not only provide power support for users, but also can feed back excess power to the power grid through the power feedback function, supplementing power resources for the power grid. This mode of two-way energy flow improves the energy utilization rate of the power grid to some extent and relieves the pressure of traditional centralized power generation. However, the connection of distributed energy devices also makes the operation mode of the power grid more complex, especially in the case of rapid increase in the number of devices. How to effectively manage the power feedback behavior of these devices and ensure the stability of the power grid has become an important challenge in power grid regulation.

[0003] In related technologies, for the problem of power feedback of distributed energy devices, the feedback behavior of the devices is usually managed by fixed rules or static parameters. For example, by setting a unified feedback power limit or a fixed feedback time window, the local fluctuation of the power grid caused by a single device feeding back for a long time and at a high power is avoided. In addition, some technical solutions also detect the state of the power grid by pre-setting a safety threshold of the power grid (such as a voltage and frequency range), and temporarily limit the feedback behavior of some devices when the load of the power grid approaches the threshold. These methods can to some extent cope with the influence of single or small number of devices on the power grid.

[0004] However, with the rapid increase in the number of distributed energy devices, the management method based on fixed rules or static parameters in the prior art may cause voltage or frequency fluctuations of the power grid in the case of simultaneous feedback of multiple devices. Since the prior art fails to coordinate the feedback behavior of multiple devices, the overlap of the feedback period and the feedback power of the devices increases. When too many devices feed back power at the same time, the superposition of the feedback power may exceed the bearing capacity of the power grid, thereby causing instability of the operation of the power grid. SUMMARY

[0005] The present application provides a method and system for power supply power feedback to a power grid, an electronic device and a storage medium, which are used to solve the problem that the voltage or frequency of the power grid may fluctuate when too many devices feed back power to the power grid at the same time, affecting the stability of the overall power grid.

[0006] In a first aspect, the present application provides a method for power supply power feedback to a power grid, applied to a power supply power feedback to a power grid control system, and the method comprises the following steps. Obtaining identity information and feedback capability parameters of a plurality of feedback device nodes; According to the feedback capability parameters, calculating maximum feedback power and optimal feedback duration of each feedback device node, and generating a feedback capability model corresponding to each feedback device node; Inputting historical power consumption data in a preset historical time period, current environmental data and regional activity data into a preset machine learning model, and outputting a power grid load curve and a potential feedback peak period in a future preset time period; According to the power grid load curve and the feedback capability model, dividing the feedback device nodes into a plurality of feedback groups, and assigning a peak-shifting feedback time period to each feedback group to generate an initial feedback allocation scheme, the initial feedback allocation scheme including member devices, feedback time windows and feedback total quantity limits of each feedback group; Based on the priority order of the device nodes in each feedback group and the feedback time window in the initial feedback allocation scheme, calculating specific feedback parameters of each feedback device node, the specific feedback parameters including feedback start and end times, feedback power curves and maximum feedback quantities; According to the specific feedback parameters of each feedback device node, generating a feedback token and distributing it to the corresponding feedback device node.

[0007] Through the above embodiments, the system realizes intelligent grouping and peak-shifting scheduling of feedback devices by obtaining feedback device node information and establishing a feedback capability model, combining machine learning to predict power grid load. This scheme addresses the problem of concentrated power grid load caused by simultaneous feedback of multiple devices by assigning staggered feedback time periods to different feedback groups. At the same time, based on the priority order, the feedback parameters of each node are accurately calculated, and strict control is carried out through the feedback token mechanism, reducing the voltage and frequency fluctuations of the power grid caused by excessive concentration of feedback power, and ensuring the stability of the power grid operation.

[0008] In some embodiments, before the step of dividing the feedback device nodes into a plurality of feedback groups according to the power grid load curve and the feedback capability model, and assigning a peak-shifting feedback time period to each feedback group to generate an initial feedback allocation scheme, it further includes: According to the topology of the power grid, dividing the feedback device nodes into regional feedback groups; Combining the regional feedback groups with the feedback capability parameters and priorities of each feedback device node to generate a feedback capability model in units of regions; When dividing the feedback groups according to the feedback capability model, preferentially grouping within the same region.

[0009] Through the above embodiment, the system completes the grouping of the feedback devices in the region in priority by taking the power grid topology as the grouping basis, thereby reducing the cross-regional power transmission loss. The scheme establishes a more accurate feedback capability model based on the regional characteristics, enables the feedback power to be consumed nearby, improves the power transmission efficiency, reduces the line burden caused by long-distance power transmission, and thus optimizes the economy and reliability of the overall feedback scheme.

[0010] In some embodiments, before the step of generating an initial feedback allocation scheme by dividing the feedback device nodes into a plurality of feedback groups according to the power grid load curve and the feedback capability model and allocating a peak-shifting feedback time period to each feedback group, the method further comprises: calculating the maximum instantaneous feedback power and the sustainable feedback duration of each feedback device node based on the feedback capability parameters of the feedback device nodes; dividing the feedback device nodes into a high-priority level and a low-priority level according to the priority; summarizing the maximum instantaneous feedback power and the sustainable feedback duration of the feedback device nodes in each priority level to generate a layered feedback capability model, wherein the layered feedback capability model comprises feedback parameters corresponding to the high-priority level devices and the low-priority level devices.

[0011] Through the above embodiment, the system calculates the maximum instantaneous feedback power and the sustainable duration of the devices, and models them according to the priority, thereby realizing the hierarchical management of the feedback resources. The layered model can more accurately reflect the feedback characteristics of devices of different priorities, provide a basis for the system to develop a more reasonable scheduling strategy, and improve the accuracy and flexibility of feedback management.

[0012] In some embodiments, after the step of calculating the specific feedback parameters of each feedback device node based on the priority order of the device nodes in each feedback group and the feedback time window in the initial feedback allocation scheme, the method further comprises: summarizing the specific feedback parameters of each feedback device node to obtain a total feedback power; when the total feedback power exceeds a preset power grid load threshold, allocating the feedback power exceeding the threshold to a virtual battery node for temporary storage; during the low-load period of the power grid, calculating a release power curve of the virtual battery node according to the storage power of the virtual battery node and the power grid load curve, and releasing the stored feedback power to the power grid according to the release power curve.

[0013] Through the above embodiment, the system can store excess power when the feedback power exceeds the threshold by introducing a virtual battery node as a temporary energy storage unit, and release it during the load valley period, thereby achieving peak shaving of feedback power. This dynamic adjustment mechanism avoids waste of feedback power and also provides buffer space for load balancing of the power grid, thereby improving the energy utilization efficiency of the system.

[0014] In some embodiments, after the step of generating a feedback token according to the specific feedback parameters of each feedback device node and distributing it to the corresponding feedback device node, the method further comprises: identifying the digestible feedback power capacity and the location information of the feedback device nodes in the microgrid according to the real-time load model of each power consumption device in the microgrid; calculating the shortest feedback path according to the digestible feedback power capacity and the location information of the feedback device nodes; calculating the microgrid digestion rate according to the digestion amount of feedback power in the microgrid and the amount of power transmitted to the main grid in the execution of the shortest feedback path; adjusting the feedback parameters of the feedback device nodes based on the microgrid digestion rate.

[0015] Through the above embodiment, the system can achieve optimal allocation of feedback power by identifying the digestible capacity of the microgrid and calculating the shortest feedback path. The scheme of dynamically adjusting the feedback parameters based on the microgrid digestion rate improves the local digestion rate, reduces the amount of power returned to the main grid, reduces the impact on the main grid, and makes the feedback scheme more adaptable.

[0016] In some embodiments, after the step of generating a feedback token according to the specific feedback parameters of each feedback device node and distributing it to the corresponding feedback device node, the method further comprises: allocating a basic subsidy coefficient to different feedback periods based on the grid load curve, the basic subsidy coefficient being inversely related to the grid load; obtaining the historical feedback cooperation degree of each feedback device node, and calculating an individual adjustment coefficient according to the historical feedback cooperation degree; calculating the final subsidy amount of the feedback device node in the original feedback period and a plurality of alternative feedback periods by integrating the basic subsidy coefficient and the individual adjustment coefficient, to form a personalized subsidy scheme including a comparison of subsidy amounts; pushing the personalized subsidy scheme to the feedback device nodes, and updating the feedback token according to the response results of the feedback device nodes.

[0017] Through the above embodiments, the system forms a more targeted economic incentive scheme by establishing a dynamic subsidy mechanism based on grid load and combining historical cooperation degree for personalized adjustment. This flexible subsidy strategy can guide the feedback behavior to shift to the ideal time period, improve user participation enthusiasm, and promote the rational use of feedback resources.

[0018] In some embodiments, after the step of generating feedback tokens according to specific feedback parameters of each feedback device node and distributing them to the corresponding feedback device nodes, the system further includes: Obtaining real-time state parameters of the power grid, actual feedback power of each feedback device node, and cumulative feedback power; Comparing the real-time state parameters with expected state parameters predicted based on the grid load curve, and calculating a deviation value; When the deviation value exceeds a preset deviation threshold, re-predicting a new short-term grid load change trend based on the real-time state parameters; Adjusting the initial feedback allocation scheme according to the short-term grid load change trend and the feedback capability model of the feedback device node before executing the feedback token; Updating the feedback token of the corresponding feedback device node according to the adjusted initial feedback allocation scheme, which includes modifying the feedback time window, adjusting the maximum feedback power limit, or temporarily revoking the feedback token.

[0019] Through the above embodiments, the system realizes closed-loop control of feedback management by real-time monitoring of grid state and feedback execution, timely adjusting the prediction model and allocation scheme when a large deviation is found. This dynamic response mechanism can quickly respond to changes in grid operating conditions, ensuring that the feedback scheme always meets actual needs and improving the reliability and adaptability of the system.

[0020] In a second aspect, the present application provides a power supply power feedback grid control system, which includes one or more processors and a memory; The memory is coupled to the one or more processors, and the memory is used to store computer program codes, which include computer instructions. The one or more processors invoke the computer instructions to enable the system to implement the method of the power supply power feedback grid provided in the above embodiments, which will not be repeated here.

[0021] In a third aspect, the present application provides a computer readable storage medium, which includes instructions that, when executed on a power supply power feedback grid control system, enable the system to implement the method of the power supply power feedback grid provided in the above embodiments, which will not be repeated here.

[0022] In a fourth aspect, the present application provides a computer program product, which, when running on a power supply electricity feedback power grid control system, enables the system to implement the method of the power supply electricity feedback power grid provided by the above-mentioned embodiments, which will not be described here again.

[0023] The one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By obtaining feedback device node information to establish a feedback capability model, combining machine learning to predict power grid load for intelligent grouping, the peak shaving scheduling of feedback devices is realized. The scheme allocates staggered time periods for different feedback groups and accurately calculates feedback parameters based on priority, strictly controls through the feedback token mechanism, effectively solves the problem of power grid voltage and frequency fluctuations caused by multiple devices feedback at the same time, and ensures the stability of power grid operation.

[0024] 2. A multi-level regulation system including priority layering, virtual battery nodes and real-time monitoring is established. The characteristics of different priority devices are reflected through the layered model, the virtual battery is used to realize peak clipping and valley filling, and the prediction model and allocation scheme are quickly adjusted through real-time monitoring. This dynamic adjustment mechanism not only ensures the efficient use of feedback electricity, but also provides safety protection for system operation, forming a complete closed-loop control system.

[0025] 3. Combined with the topology of the power grid, the priority grouping within the region is realized, the power distribution is optimized by identifying the consumable capacity of the microgrid and the shortest feedback path, and a dynamic subsidy mechanism is established to guide the feedback behavior. This regional coordination strategy not only reduces transmission losses across regions and improves local consumption efficiency, but also optimizes the temporal and spatial distribution of feedback resources through economic incentives, making the entire feedback system run more economically and reliably. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 is a flowchart of a method of a power supply electricity feedback power grid in the embodiments of the present application; Figure 2 is another flowchart of a method of a power supply electricity feedback power grid in the embodiments of the present application; Figure 3 is a schematic diagram of an entity device structure of a power supply electricity feedback power grid control system in the embodiments of the present application. DETAILED DESCRIPTION

[0027] The terminology used in the following embodiments of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used in the description of the application and the appended claims, the singular forms "a," "an" and "the" are intended to include both singular and plural forms, unless the context clearly indicates otherwise. It will be further understood that the terms "and / or," as used in the description of the present application, refers to any one or more of the listed items, optionally including zero.

[0028] Hereinafter, the terms "first", "second" are only for the purpose of description, and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise specified.

[0029] For the convenience of understanding, the method provided by the present embodiment is described in the flow. Please refer to Figure 1 , a flowchart of a method for feeding back power of a power supply to a power grid in an embodiment of the present application.

[0030] S101, obtain the identity information and feedback capability parameters of a plurality of feedback device nodes.

[0031] Among them, the feedback device node represents the distributed energy device accessing the power grid, such as photovoltaic power generation system, energy storage battery, electric vehicle charging pile, etc.; the identity information refers to the parameter identifying the uniqueness of the device, including device ID, model, manufacturer, installation location coordinates, topological node number accessing the power grid, etc.; the feedback capability parameter is used to represent the related index of the power feedback capability of the device, including rated feedback power, maximum feedback power, allowed continuous feedback time, charging and discharging efficiency, current remaining power, etc.

[0032] Specifically, the system establishes a connection with the feedback device node through wired (such as RS485, Ethernet) or wireless (such as 4G / 5G, WiFi, LoRa) communication protocol, and periodically or triggeredly sends parameter query instructions to the device. After the device receives the instruction, it encapsulates its own identity information and feedback capability parameters and returns them to the control system through the communication link. The control system checks and analyzes the received data, stores it to the device information database after eliminating invalid data, and forms a device file.

[0033] S102, calculate the maximum feedback power and the best feedback time of each feedback device node according to the feedback capability parameters, and generate the feedback capability model corresponding to each feedback device node.

[0034] Wherein, the maximum feedback power refers to the maximum total amount of electrical energy that the device can safely feedback to the power grid under the current device state; the optimal feedback duration is used to represent the duration of the device feedback at the optimal efficiency; the feedback capability model refers to the mathematical model formed by structuring and packaging the feedback capability parameters, maximum feedback power, optimal feedback duration, and other information of the device, which is used to represent the feedback capability characteristics of the device.

[0035] The system extracts the feedback capability parameters of each feedback device node from the device information database, such as the rated feedback power Pn, the current remaining power E, the allowed depth of discharge DOD, the maximum continuous operating temperature Tmax, etc. Optionally, for battery-type devices, the maximum feedback power Emax can be calculated by Emax = E × DOD; for photovoltaic power generation systems, the maximum feedback power is dynamically determined by factors such as current light intensity, component efficiency, and inverter capacity. The optimal feedback duration t is calculated according to the rated power and maximum feedback power of the device, i.e. t = Emax / Pn, and combined with the heat dissipation characteristics and historical operation data of the device, the calculation result is corrected to avoid overload operation of the device. Finally, the calculated maximum feedback power, optimal feedback duration, and original parameters are integrated to generate the feedback capability model of each device, which is stored in the model database.

[0036] S103, input the historical power consumption data in the preset historical time period, the current environmental data and the regional activity data into the preset machine learning model, and output the power grid load curve and potential feedback peak period in the future preset time period.

[0037] Wherein, the historical power consumption data refers to the active power, reactive power, voltage, frequency, and other operating data of the power grid in the past period (such as the past 1 month, 1 year), as well as the time-sharing statistical data of power consumption in each region; the environmental data is used to represent natural environmental factors that affect the power grid load and distributed energy output, including temperature, humidity, light intensity, wind speed, precipitation, etc., for example, light intensity affects photovoltaic power generation, temperature affects residential air conditioning load; the regional activity data refers to information related to social activities in the region where the power grid is located, including workday / weekend arrangements, holidays, large event holding times, industrial production shift plans, etc.

[0038] Specifically, the system obtains historical power consumption data of a preset historical time period (such as the past 30 days) from a power grid data center, including the total network active power of each hour, the load distribution of each region, and the like. At the same time, current environmental data is obtained from an interface of a meteorological department, such as hourly weather forecasts (temperature, light intensity, wind speed, and the like) for the next 24 hours. Regional activity data is obtained from a public data platform, such as upcoming holidays, production suspension or expansion plans of factories in the region, and the like. After preprocessing these data, a feature vector is formed. Then the feature vector is input into a preset machine learning model (such as a trained LSTM model), and the model outputs the power grid load prediction value for a preset time period (such as the next 24 hours) through the calculation of multiple layers of neural networks, generating a continuous load curve. At the same time, the model identifies time periods with low load (lower than a set threshold value, such as 60% of the average load) and high distributed energy output according to the load curve and environmental data, as potential feedback peak periods.

[0039] S104, dividing the feedback device nodes into multiple feedback groups according to the power grid load curve and the feedback capability model, and assigning a peak-shifting feedback time period to each feedback group to generate an initial feedback allocation scheme.

[0040] Among them, the feedback group represents a grouping formed by aggregating feedback device nodes with similar feedback characteristics or geographical locations; the peak-shifting feedback time period refers to a time period assigned to different feedback groups that does not overlap or is in a load valley, so as to avoid high power grid load caused by simultaneous feedback of multiple groups; the initial feedback allocation scheme is used to represent a scheduling scheme containing member device lists of each feedback group, exclusive feedback time windows (such as 08:00-10:00), total feedback amount limits (such as total feedback power in the group ≤ 500 kWh), and the like.

[0041] Specifically, the system first analyzes the power grid load curve to identify multiple load valley periods (such as night 00:00-06:00, noon 12:00-14:00) and peak periods. Then, according to the device parameters (such as maximum feedback power, optimal duration) in the feedback capability model, combined with the device priority (such as high-priority devices need to be arranged for feedback first), the devices are divided into several feedback groups. And when grouping, the "same type of device first" principle is followed, for example, the energy storage battery and the photovoltaic device are divided into separate groups. Subsequently, a peak-shifting time period is assigned to each group to ensure that the total feedback power of multiple groups in the same period does not exceed the maximum feedback capacity allowed by the power grid. Finally, an initial scheme containing group members, time periods, and total amount limits is generated and stored in a scheduling database.

[0042] S105, calculating the specific feedback parameters of each feedback device node based on the priority order of the device nodes in each feedback group and the feedback time window in the initial feedback allocation scheme.

[0043] The system obtains the time window and total amount limit of each feedback group from the initial scheme, and combines the priority order of the devices in the group to perform parameter calculation. For high-priority devices, the front section of the time window or the power peak period is preferentially allocated to ensure that their feedback requirements are met; for low-priority devices, the remaining period or the power valley period is allocated. In specific calculation, according to the maximum feedback power of the device and the length of the time window, the total amount of feedback in the group is proportionally allocated according to the priority. For example, the total feedback amount limit of a group is 100 kWh, and there are 2 high-priority devices (A and B) and 3 low-priority devices (C, D and E) in the group, then the high-priority devices allocate 60 kWh (30 kWh each), and the low-priority devices allocate 40 kWh (8 kWh each). At the same time, according to the optimal feedback time length and power curve of the device, the start and end time and the power output slope are adjusted to avoid device overload.

[0044] S106, generate a feedback token according to the specific feedback parameters of each feedback device node, and distribute it to the corresponding feedback device node.

[0045] The system generates structured data (such as JSON format) for the specific feedback parameters of each device, including parameter details and timestamps. Use asymmetric encryption algorithm (such as RSA) to sign the data and generate feedback token. Then push the token to the device end through the communication channel corresponding to the device (such as 4G, WiFi). After the device receives the token, it verifies the validity of the signature, and if it passes, it is stored in the local cache and waits for the feedback period to trigger execution. If the verification fails or the token is not received within the timeout period, the feedback operation is refused and the system is reported.

[0046] In addition, the system can also collect the voltage, current, power factor and other data of each electrical equipment in real time through the smart meter, sensor and other devices in the micro-grid, generate real-time load model combined with the equipment type and the preset load characteristic model, aggregate and calculate the total real-time power consumption of the micro-grid, subtract the real-time output of the distributed power supply, obtain the consumable feedback power capacity, extract the installation address latitude and longitude, power grid node number and other location information of the feedback equipment node from the equipment information database; then abstract the micro-grid topology structure into a graph model composed of nodes and edges, take the line transmission loss parameter or physical length as the edge weight, determine the starting point and ending point according to the location of the feedback equipment node and the consumable node, calculate the shortest feedback path with the minimum transmission loss and the highest efficiency by using the shortest path algorithm such as Dijkstra algorithm, and preferentially select the line with low loss and large remaining capacity; when the feedback operation is performed at the feedback equipment node, measure the feedback power consumption by the smart meter of the electrical equipment in the micro-grid, measure the power transmitted to the main grid by the gateway metering device of the micro-grid and the main grid interconnection point, calculate the micro-grid digestion rate by dividing the digestion amount by the total feedback power; finally, set the digestion rate target threshold, if the actual digestion rate is lower than the threshold, appropriately increase the feedback power or prolong the time, if the actual digestion rate is higher than the threshold and close to the upper limit of the capacity, reduce the power or shorten the time, preferentially adjust the low priority equipment parameters, and update the feedback token and issue it to the corresponding node.

[0047] Optionally, the system can also analyze the power grid load curve, divide the future preset time period into multiple feedback periods and mark the load level, according to the preset load-subsidy mapping rule, assign the basic subsidy coefficient inversely related to the power grid load to the periods of different load levels such as trough, flat section and peak, for example, if the load in the trough period is lower than 60% of the average load, the subsidy coefficient is set to 1.5 yuan / kWh, the flat section is 1.0 yuan / kWh, and the peak is 0.5 yuan / kWh, and stored in the subsidy strategy database; then extract the historical feedback data of each feedback device node from the device operation log database, including the deviation of actual feedback start and end time from the plan, power curve matching degree, power completion rate and other indicators, calculate the comprehensive matching degree score according to the preset weight, and then convert the score to an individual adjustment coefficient of 0.5-1.5 through linear transformation, for example, if the matching degree score is 80 points, the coefficient is 0.8x0.5+0.5=0.9; then generate the original feedback period and multiple alternative feedback periods (such as selecting the 3 lowest load non-original periods) for each device node, calculate the final subsidy amount according to the product of the basic subsidy coefficient and the individual adjustment coefficient of each period, for example, the original period is the flat section (1.0 yuan / kWh), the individual adjustment coefficient is 0.9, the final subsidy is 0.9 yuan / kWh, and the final subsidy of a certain alternative trough period (1.5 yuan / kWh) is 1.35 yuan / kWh, forming a personalized subsidy scheme including the load level, subsidy coefficient and amount comparison of each period; finally, send the scheme to the device node through 4G SMS, APP push and other ways, receive the device response result (such as selecting a certain alternative period), verify the validity, modify the time window and feedback total amount limit in the feedback token according to the selected period, generate a new token using RSA encryption and issue it, and if it is refused, maintain the original token parameters.

[0048] In the above embodiment, the system realizes intelligent grouping and peak shifting of feedback devices by obtaining feedback device node information and establishing a feedback capability model, combining with machine learning to predict power grid load. This scheme allocates staggered feedback periods for different feedback groups to address the problem of concentrated power grid load caused by simultaneous feedback of multiple devices. At the same time, the feedback parameters of each node are accurately calculated based on the priority order, and strict control is carried out through the feedback token mechanism, reducing the voltage and frequency fluctuations of the power grid caused by excessive concentration of feedback power, and ensuring the stability of the power grid operation.

[0049] The method provided by the embodiment is further described in more detail below. Please refer to Figure 2 , another flowchart of the method of power supply power feedback to the power grid in the embodiment of the present application.

[0050] S201, divide the feedback device nodes into regions according to the topology of the power grid, and generate multiple regional feedback groups.

[0051] The system first obtains power grid topology data from the power grid management platform, including substation location, power transmission line direction, power distribution area division, and other information. For newly connected devices, through their installation location coordinates (such as GPS data) or connected power grid node numbers, they are matched to the corresponding power distribution area. Then, according to the "near access" principle, devices under the same substation or power distribution line are divided into a regional feedback group. For example, all photovoltaic power stations and energy storage devices connected to substation A are divided into regional group A, and devices connected to substation B are divided into regional group B. The division results are stored in the regional management database and associated with device information.

[0052] S202, combine the regional feedback group with the feedback capability parameters and priority of each feedback device node to generate a feedback capability model in units of regions.

[0053] The system obtains the member list of each regional feedback group from the regional management database, and extracts the feedback capability parameters (such as rated power Pn, maximum feedback power Emax) and priority level (such as high, medium, and low) of each member from the device information database. Statistical analysis is performed on the parameters in the region, and the total maximum feedback power E1 of the region is calculated as E1 = ∑Emax, and the average optimal feedback duration t1 is calculated as t1 = (∑t × wi) / ∑wi (where wi is the priority weight of the device, the high priority weight is 2, the medium priority is 1, and the low priority is 0.5). At the same time, the proportion of high priority devices in the region is calculated, and a regional priority distribution histogram is generated. Finally, these data are packaged into a regional feedback capability model and stored in the model database for calling during grouping and scheduling.

[0054] S203, calculate the maximum instantaneous feedback power and sustainable feedback duration of each feedback device node based on the feedback capability parameters of the feedback device node.

[0055] The system extracts the feedback capability parameters of each device from the device information database, such as the rated capacity C (kWh) of the battery and the maximum discharge rate D (C-rate), then the maximum instantaneous feedback power P = C × D (kW). The sustainable feedback duration t is calculated based on the maximum feedback power Emax and the rated feedback power Pn, that is, t = Emax / Pn, and is corrected in combination with the heat dissipation capacity and historical operation data of the device. For example, if the temperature of the device will exceed the safety threshold after continuous feedback for 2 hours, then the sustainable duration is taken as the smaller value of 2 hours and Emax / Pn.

[0056] S204, divide the feedback device nodes into high priority level and low priority level according to priority.

[0057] The system assigns priority levels to each device according to preset priority rules. For example, the rules are set as follows: energy storage devices (such as battery energy storage systems) are high priority, photovoltaic inverters and electric vehicle charging piles are low priority; devices that have signed an emergency frequency modulation agreement with the grid are high priority, and ordinary user devices are low priority. For special scenarios (such as grid frequency fluctuations), devices participating in frequency modulation are automatically promoted to the highest priority. The division result is stored in the device information database and bound to the device ID.

[0058] S205, the maximum instantaneous feedback power and sustainable feedback duration of the feedback device node in each priority level are summarized to generate a hierarchical feedback capability model.

[0059] The system extracts the maximum instantaneous feedback power and sustainable feedback duration data of all devices in each priority level (such as high priority level and low priority level) from the device information database. For high priority level devices, calculate their total maximum instantaneous power Phigh_total=∑Pinstant_high and average sustainable duration thigh_avg=∑Emax_high∑tsustain_high×Emax_high (weighted by power) ; for low priority level devices, perform the same statistics to obtain Plow_total and tlow_avg. Encapsulate the two-layer parameters as a hierarchical model, containing {Phigh_total,thigh_avg,Plow_total,tlow_avg}, and store it in the model database for calling when formulating the scheduling strategy.

[0060] S206, after calculating the specific feedback parameters of each feedback device node, the specific feedback parameters of each feedback device node are summarized to obtain the total feedback power.

[0061] The system traverses the specific feedback parameters of all feedback device nodes, extracts the power curve data of each device in the corresponding feedback period (such as one sampling point every 15 minutes). Add all device power values at the same time point to obtain the total feedback power Ptotal(t)=∑Pi(t) at that time point, forming a total power curve that changes over time. For example, there are 10 devices in a certain period, and the power curve of each device is a piecewise constant, then the total power is the power accumulated at each corresponding time. The summary result is stored in the scheduling result database for subsequent load threshold checking.

[0062] S207, when the total feedback power exceeds the preset grid load threshold, the excess feedback power is allocated to the virtual battery node for temporary storage.

[0063] The system first compares the total feedback power curve Ptotal(t) with the preset threshold Pthreshold, identifies the period set Tover of Ptotal(t) > Pthreshold. For each over-standard period t∈Tover, calculate the excess power ΔP(t)=Ptotal(t)-Pthreshold, and integrate it to get the excess power ΔE=∫ToverΔP(t)dt. Then, the system starts from the high priority device, and reduces the feedback power of part of the device according to the priority order, allocates the reduced power ΔE to the virtual battery node storage, and records the storage amount Evirtual of the virtual battery at the same time. For example, the feedback power of the low priority device is preferentially reduced to ensure the feedback demand of the high priority device.

[0064] S208, during the valley period of the power grid load, the release power curve of the virtual battery node is calculated according to the storage power of the virtual battery node and the load curve of the power grid, and the stored feedback power is released to the power grid according to the release power curve.

[0065] The system first identifies the valley period Tvalley according to the load prediction curve of the power grid, and obtains the current storage power Evirtual of the virtual battery. Then, combined with the load curve Pload(t) in the valley period and the maximum charging power Pcharge_max allowed by the power grid, the release power curve Prelease(t) of the virtual battery is calculated. The calculation method is usually uniform release or optimization release based on electricity price, for example, the storage power is evenly distributed in the valley period: Prelease(t)=Evirtual / Tduration, where Tduration is the length of the valley period. Finally, the system sends a release instruction to the virtual battery node, gradually releases the power to the power grid according to the power curve, and updates the storage amount in real time until it is zero.

[0066] S209, after the feedback token is executed at the feedback device node, the real-time state parameters of the power grid, the actual feedback power of each feedback device node and the cumulative feedback power are obtained.

[0067] The system collects the state parameters of the power grid and the feedback data of the devices in real time through sensors (such as PMU synchronous phasor measurement units) deployed in the power grid and smart meters at the device end. The data is transmitted to the data center of the control system through high-speed communication network (such as 5G, optical fiber), and stored in the real-time database after format conversion and verification. For example, for each feedback device node, the average value of the actual feedback power and the increment of the cumulative feedback power are recorded every 15 minutes, and the voltage and frequency values of the power grid at that time are also recorded.

[0068] S210, compare the real-time state parameters with the expected state parameters based on the load curve prediction of the power grid, and calculate the deviation value.

[0069] The system extracts the latest real-time state parameters (e.g. actual power Preal(t), actual voltage Vreal(t) at the current time t) from the real-time database, and obtains the expected parameters (predicted power Ppred(t), predicted voltage Vpred(t)) at the corresponding time from the prediction result database. For each parameter, calculate its deviation value, for example, power deviation ΔP = Ppred(t) | Preal(t) - Ppred(t) | x 100%, voltage deviation ΔV = Vnominal | Vreal(t) - Vpred(t) | x 100% (Vnominal is the rated voltage). Store the deviation value of each parameter to the monitoring database.

[0070] S211, when the deviation value exceeds the preset deviation threshold, re-predict the new short-term power grid load change trend based on the real-time state parameters.

[0071] The system triggers the short-term prediction process, extracts the real-time state parameters (e.g. minute-level data of power, voltage, frequency) at the current time and 1 hour before as new input features from the real-time database. Call lightweight machine learning models (e.g. ARIMA model or real-time updated LSTM model) to predict the power grid load in the next 2 hours and generate a short-term load curve Pshort(t).

[0072] S212, adjust the initial feedback allocation scheme according to the short-term power grid load change trend and the feedback capability model of the feedback device node before executing the feedback token.

[0073] The system analyzes the short-term load curve to determine the feedback period and power range that need to be adjusted. For example, if the predicted load will significantly increase after 1 hour, adjust the feedback group allocation in that period in advance to reduce the feedback power during the high load period. When adjusting, preferentially adjust the feedback parameters of low-priority devices, such as extending their feedback start and end times, reducing the peak value of the power curve, while ensuring that the maximum feedback power and sustainable duration of the device are not exceeded. After adjustment, recalculate the total feedback power of each feedback group to ensure that the power grid load threshold is not exceeded.

[0074] S213, update the feedback token of the corresponding feedback device node according to the adjusted initial feedback allocation scheme.

[0075] The system generates a new feedback parameter (such as a modified time window, an adjusted maximum power value) according to the adjusted scheme, generates a new token using the same encryption method as generating the original token, and sends an update instruction and the new token through a communication channel (such as 4G or Bluetooth) corresponding to the device. After receiving the update instruction, the device verifies the validity of the signature of the new token. If the verification is passed, the locally stored token parameters are replaced, and the feedback operation is performed according to the new parameters. If the verification fails or an invalidation instruction is received, the feedback is stopped, and the status is reported.

[0076] The power supply power feedback power grid control system of the embodiment of the application is applied to an electronic device, Figure 3 An architectural schematic diagram of an electronic device suitable for implementing the embodiment of the application is shown.

[0077] It should be noted that, Figure 3 The electronic device shown is only an example and should not limit the functions and use range of the embodiment of the application.

[0078] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions (computer programs) or controlled by instructions (computer programs) related hardware, and the instructions can be stored in a computer readable storage medium and loaded and executed by a processor. The electronic device of the embodiment includes a storage medium and a processor, wherein the storage medium stores a plurality of instructions, and the instructions can be loaded by the processor to execute any step of the method provided by the embodiment of the application.

[0079] Specifically, the storage medium and the processor are directly or indirectly electrically connected to realize the transmission or interaction of data. For example, the elements can be electrically connected to each other through one or more signal lines. The storage medium stores computer execution instructions for realizing the data access control method, including at least one software function module stored in the storage medium in the form of software or firmware. The processor executes various function applications and data processing by running the software program and the module stored in the storage medium. The storage medium can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. Among them, the storage medium is used to store programs, and the processor executes the programs after receiving execution instructions.

[0080] Further, the software program and the module in the storage medium can also include an operating system, which can include various software components and / or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and can communicate with various hardware or software components to provide a running environment for other software components. The processor can be an integrated circuit chip with signal processing capability. The processor described above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc., which can realize or execute the methods, steps and logic flow diagrams disclosed in the embodiments. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0081] Due to the instructions stored in the storage medium, the steps in any method provided by the embodiments of the present application can be executed, and thus the beneficial effects of any method provided by the embodiments of the present application can be achieved. Details are described in the foregoing embodiments, which will not be repeated here.

[0082] The above merely describes preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any changes or replacements within the technical scope disclosed by the present application, which can be easily thought by those skilled in the art, should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for power supply power feedback to power grid, applied to power supply power feedback to power grid control system, characterized in that, The method comprises: obtaining identity information and feedback capability parameters of a plurality of feedback device nodes; calculating the maximum feedback power and the optimal feedback duration of each feedback device node according to the feedback capability parameters, and generating a feedback capability model corresponding to each feedback device node; inputting historical power consumption data in a preset historical time period, current environmental data and regional activity data into a preset machine learning model, and outputting a power grid load curve and a potential feedback peak period in a future preset time period; dividing the feedback device nodes into a plurality of feedback groups according to the power grid load curve and the feedback capability model, and assigning a peak-shifting feedback time period to each feedback group to generate an initial feedback allocation scheme, wherein the initial feedback allocation scheme comprises member devices, feedback time windows and feedback total quantity limits of each feedback group; calculating specific feedback parameters of each feedback device node based on the priority order of the device nodes in each feedback group and the feedback time windows in the initial feedback allocation scheme, wherein the specific feedback parameters include feedback start and end times, feedback power curves and maximum feedback quantities; generating feedback tokens according to the specific feedback parameters of each feedback device node and distributing the feedback tokens to the corresponding feedback device nodes.

2. The method of claim 1, wherein, Before the step of dividing the feedback device nodes into a plurality of feedback groups according to the power grid load curve and the feedback capability model, and assigning a peak-shifting feedback time period to each feedback group to generate an initial feedback allocation scheme, the method further comprises: dividing the feedback device nodes into a plurality of regional feedback groups according to the topology of the power grid; combining the regional feedback groups with the feedback capability parameters and priorities of the feedback device nodes to generate a feedback capability model in units of regions; when dividing the feedback groups according to the feedback capability model, preferentially grouping within the same region.

3. The method of claim 1, wherein, Before the step of dividing the feedback device nodes into a plurality of feedback groups according to the power grid load curve and the feedback capability model, and assigning a peak-shifting feedback time period to each feedback group to generate an initial feedback allocation scheme, the method further comprises: calculating the maximum instantaneous feedback power and the sustainable feedback duration of each feedback device node based on the feedback capability parameters of the feedback device nodes; dividing the feedback device nodes into a high-priority level and a low-priority level according to the priorities; summarizing the maximum instantaneous feedback power and the sustainable feedback duration of the feedback device nodes in each priority level to generate a hierarchical feedback capability model, wherein the hierarchical feedback capability model comprises feedback parameters corresponding to the high-priority level devices and the low-priority level devices.

4. The method of claim 1, wherein, After the step of calculating the specific feedback parameters of each feedback device node based on the priority order of the device nodes in each feedback group and the feedback time windows in the initial feedback allocation scheme, the method further comprises: summarizing the specific feedback parameters of each feedback device node to obtain a total feedback power; when the total feedback power exceeds a preset power grid load threshold, allocating the excess feedback power to a virtual battery node for temporary storage; During a low-load period of the power grid, a release power curve of the virtual battery node is calculated according to a storage power of the virtual battery node and a load curve of the power grid, and stored feedback power is released to the power grid according to the release power curve.

5. The method of claim 1, wherein, The method further comprises, after the step of generating the feedback token according to the specific feedback parameter of each feedback device node and distributing the feedback token to the corresponding feedback device node: identifying a digestible feedback power capacity and location information of the feedback device node in the micro-grid according to a real-time load model of each power consumption device in the micro-grid; calculating a shortest feedback path according to the digestible feedback power capacity and the location information of the feedback device node; calculating a micro-grid digestion rate according to a digestion amount of the feedback power in the micro-grid and an amount of power transmitted to the main grid in the execution of the shortest feedback path; adjusting the feedback parameter of the feedback device node based on the micro-grid digestion rate.

6. The method of claim 1, wherein, The method further comprises, after the step of generating the feedback token according to the specific feedback parameter of each feedback device node and distributing the feedback token to the corresponding feedback device node: allocating a basic subsidy coefficient for different feedback periods based on the load curve of the power grid, the basic subsidy coefficient being inversely related to the load of the power grid; obtaining a historical feedback cooperation degree of each feedback device node, and calculating an individual adjustment coefficient according to the historical feedback cooperation degree; calculating a final subsidy amount of the feedback device node in the original feedback period and a plurality of alternative feedback periods by integrating the basic subsidy coefficient and the individual adjustment coefficient, to form a personalized subsidy scheme including a comparison of the subsidy amounts; pushing the personalized subsidy scheme to the feedback device node, and updating the feedback token according to a response result of the feedback device node.

7. The method of claim 1, wherein, The method further comprises, after the step of generating the feedback token according to the specific feedback parameter of each feedback device node and distributing the feedback token to the corresponding feedback device node: obtaining a real-time state parameter of the power grid, an actual feedback power of each feedback device node, and a cumulative feedback power; comparing the real-time state parameter with an expected state parameter predicted based on the load curve of the power grid, to calculate a deviation value; when the deviation value exceeds a preset deviation threshold, re-predicting a new short-term power grid load change trend based on the real-time state parameter; adjusting an initial feedback allocation scheme according to the short-term power grid load change trend and a feedback capability model of the feedback device node before the execution of the feedback token; updating the feedback token of the corresponding feedback device node according to the adjusted initial feedback allocation scheme, the updating including modifying a feedback time window, adjusting a maximum feedback power limit, or temporarily revoking the feedback token.

8. A power flow feedback grid control system, characterized by, The system comprises one or more processors and a memory; The memory is coupled to the one or more processors, and the memory is configured to store computer program codes, the computer program codes comprising computer instructions, and the one or more processors are configured to invoke the computer instructions to cause the system to perform the method of any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are run on a power supply power feedback grid control system, the system is caused to perform the method of any one of claims 1-7.

10. A computer program product, characterised in that, When the computer program product is run on a power supply grid feedback control system, it causes the system to perform the method of any one of claims 1-7.

Images