Power distribution network intelligent regulation and control method and system based on load characteristics

Abstract

The invention discloses a power distribution network intelligent regulation and control method and system based on load characteristics, and belongs to the field of power supply.The power distribution network intelligent regulation and control method based on the load characteristics comprises the following steps that historical power utilization load data of users in a power distribution area are collected, and the historical power utilization load data are analyzed through a big data analysis technology; compared with the prior art, the method has the beneficial effects that multi-time-scale accurate prediction of regional new energy power generation and load requirements is realized; based on the prediction result, the power grid operation state is simulated in advance in the simulation model, and the potential voltage out-of-limit risk is accurately positioned; finally, by means of an optimization algorithm, a regulation and control strategy is generated and executed before a voltage problem occurs, traditional post-passive remediation is converted into pre-active defense, voltage out-of-limit is fundamentally avoided, and adaptability and operation safety of a power distribution network to high-proportion new energy access are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of power supply, and particularly relates to an intelligent control method and system for power distribution networks based on load characteristics. Background Technology

[0002] In power distribution networks, traditional voltage regulation mainly relies on on-load tap-changing transformers, capacitor banks, and other equipment. These methods are based on local measurements and mechanical actions, resulting in slow response times (measured in minutes), and their control logic is designed for a unidirectional power flow from the substation to the user.

[0003] However, the integration of a high proportion of renewable energy generation (solar power, wind power, hydropower, etc.) has completely changed this scenario. The rapid and random fluctuations in renewable energy generation require regulation equipment to respond within seconds, which is impossible for traditional mechanical equipment. More importantly, when distributed photovoltaic and other power sources generate large amounts of electricity on the user side, the power flows back into the grid, causing the line voltage to be abnormally high. This causes the control logic based on unidirectional power flow design to fail, or even make incorrect regulation decisions, thus exacerbating the voltage over-limit problem.

[0004] Therefore, traditional voltage regulation methods are difficult to effectively adapt to the random characteristics of new energy power generation and need to be improved. Summary of the Invention

[0005] Therefore, it is necessary to provide a method and system for intelligent control of power distribution networks based on load characteristics to address the above-mentioned problems.

[0006] The present invention is implemented as follows: a method for intelligent control of a power distribution network based on load characteristics, comprising the following steps: Collect historical electricity load data of users in the distribution area, and extract typical production and living load curves and patterns (such as morning and evening peak hours, seasonal changes, etc.) through big data analysis technology to obtain historical electricity load characteristics; obtain information on all new energy power generation in the area, and build a new energy power generation information database, which includes power generation, power generation type, and geographical location. By accessing weather forecast data (such as illuminance, wind speed, and temperature) and combining it with existing new energy power generation information, a power generation prediction model is established to predict the power generation of new energy within a set future time period (such as the next 24 hours). Based on historical electricity load characteristics, a load prediction model is established to predict the total electricity load power of users in the region within a set future time period. The predicted renewable energy power generation within a set future time period is compared with the total electricity load power to calculate the net load power of each node (total electricity load power minus renewable energy power generation). The net load power is then used as the injected power input into the simulation model of the distribution network to predict the voltage level of each node in the distribution network within a set future time period, locate the risk points and time periods of voltage overruns, and obtain the prediction results. Based on the prediction results, control strategies are generated using optimization algorithms (such as the safety-constrained optimal power flow algorithm). For identified risk points and time periods where voltage overruns are likely to occur, the control strategies are issued in advance to the corresponding regulating equipment for execution (e.g., adjusting the tap position of on-load tap changers, controlling the switching of smart capacitors, or guiding distributed power sources to adjust reactive power output).

[0007] In one embodiment, the present invention provides a method for intelligent control of a distribution network based on load characteristics, further comprising: The system monitors the actual operating data of the power distribution network in real time, compares the actual operating data with the prediction results, and updates the prediction and regenerates the control strategy based on the latest actual operating data when the deviation between the two exceeds the first set threshold, so as to ensure the real-time performance and accuracy of the control. It also records the deviation data and periodically calibrates the power generation prediction model and load prediction model offline based on the deviation data.

[0008] In one embodiment, the present invention provides a method for intelligent control of a distribution network based on load characteristics, further comprising: The system continuously compares the real-time power demand values ​​of each node with the predicted demand values ​​based on the output of known power generation equipment (predicted renewable energy power generation plus thermal power generation) to obtain a power deviation curve. When a node's power deviation curve shows a persistent and significant negative deviation (i.e., the actual demand is much lower than the theoretical demand calculated from the known load and known power generation equipment), and the negative deviation exceeds a second set threshold, it is determined that a new renewable energy power generation device has been put into operation on the line to which the node belongs. After the new renewable energy power generation device is confirmed, the device's parameters (such as type, capacity, and location) are updated to the renewable energy power generation information database, and the power generation prediction model and simulation model are reconstructed.

[0009] In one embodiment, the present invention provides a method for intelligent control of a distribution network based on load characteristics, further comprising: A new energy power generation equipment information database is pre-set. The database includes standardized output power characteristic curves of various typical new energy equipment (such as monocrystalline silicon photovoltaic, thin-film photovoltaic, doubly fed wind turbine, etc.) under different environmental conditions (such as light, wind speed, temperature, etc.). When it is determined that a new new energy power generation equipment is put into operation, the current and historical environmental data are synchronized with the observed power deviation curves and matched with the standardized output power characteristic curves. By comparing the fluctuation patterns and environmental response characteristics, one or more inferred new energy power generation equipment types and their confidence levels are output (e.g., "suspected new monocrystalline silicon photovoltaic, confidence level 85%").

[0010] In one embodiment, the present invention provides a method for intelligent control of a distribution network based on load characteristics, further comprising: Based on future weather forecast data and predicted renewable energy power generation (e.g., 3 days), the supply and demand situation of the power distribution network can be judged in advance. When a sustained power generation surplus or shortage is predicted, a dynamic electricity pricing strategy can be generated (e.g., setting low prices during peak power generation periods to incentivize electricity consumption and setting high prices during off-peak power generation periods to suppress electricity consumption).

[0011] In one embodiment, the present invention provides an intelligent control system for a distribution network based on load characteristics, comprising: The information collection module is used to collect historical electricity load data of users in the distribution area. Through big data analysis technology, it extracts typical production and living load curves and patterns (such as morning and evening peak hours, seasonal changes, etc.) to obtain historical electricity load characteristics; it also acquires information on all new energy power generation in the area and builds a new energy power generation information database, which includes power generation, power generation type, and geographical location. The power generation and load forecasting module is used to access weather forecast data (such as illuminance, wind speed, and temperature), combine it with the acquired new energy power generation information, establish a power generation forecasting model, and predict the power generation of new energy within a future set time (such as the next 24 hours); based on historical electricity load characteristics, it establishes a load forecasting model to predict the total electricity load power of users in the region within a future set time. The voltage over-limit prediction module is used to compare the predicted renewable energy power generation within a set future time with the total power load, calculate the net load power of each node (total power load minus renewable energy power generation), input the net load power as the injected power into the simulation model of the distribution network, predict the voltage level of each node in the distribution network within a set future time, locate the risk points and time periods of voltage over-limit, and obtain the prediction results. The regulation strategy generation and execution module is used to generate regulation strategies based on prediction results and optimization algorithms (such as the safety constraint optimal power flow algorithm). For identified risk points and time periods where voltage overruns are likely to occur, the regulation strategies are issued in advance to the corresponding regulation equipment for execution (e.g., adjusting the tap position of on-load tap changers, controlling the switching of smart capacitors, or guiding distributed power sources to adjust reactive power output, etc.).

[0012] In one embodiment, the present invention provides an intelligent control system for distribution networks based on load characteristics, further comprising: The model calibration module is used to monitor the actual operating data of the power distribution network in real time, compare the actual operating data with the prediction results, and when the deviation between the two exceeds a first set threshold, update the prediction based on the latest actual operating data and regenerate the control strategy to ensure the real-time performance and accuracy of the control; record the deviation data and periodically calibrate the power generation prediction model and load prediction model offline based on the deviation data.

[0013] In one embodiment, the present invention provides an intelligent control system for distribution networks based on load characteristics, further comprising: The new equipment acquisition module continuously compares the real-time power demand value of each node with the predicted demand value based on the output of known power generation equipment (predicted new energy power generation plus thermal power generation) to obtain the power deviation curve. When a node's power deviation curve shows a persistent and significant negative deviation (i.e., the actual demand is much lower than the theoretical demand calculated from the known load and known power generation equipment), and the negative deviation exceeds a second set threshold, it is determined that a new new energy power generation equipment has been put into operation on the line to which the node belongs. After the new new energy power generation equipment is confirmed, the parameters of the equipment (such as type, capacity, and location) are updated to the new energy power generation information database, and the power generation prediction model and simulation model are reconstructed.

[0014] In one embodiment, the present invention provides an intelligent control system for distribution networks based on load characteristics, further comprising: The new equipment matching module is used to pre-set a new energy power generation equipment information database. The database includes standardized output power characteristic curves of various typical new energy equipment (such as monocrystalline silicon photovoltaic, thin-film photovoltaic, doubly fed wind turbine, etc.) under different environmental conditions (such as light, wind speed, temperature, etc.). When it is determined that a new new energy power generation equipment is put into operation, the current and historical environmental data are synchronized with the observed power deviation curves and matched with the standardized output power characteristic curves. By comparing the fluctuation patterns and environmental response characteristics, one or more inferred new energy power generation equipment types and their confidence levels are output (e.g., "suspected new monocrystalline silicon photovoltaic, confidence level 85%").

[0015] In one embodiment, the present invention provides an intelligent control system for distribution networks based on load characteristics, further comprising: The electricity price control module is used to predict the supply and demand situation of the distribution network in advance based on future weather forecast data and predicted renewable energy power generation (e.g., 3 days). When a sustained power generation surplus or shortage is predicted, dynamic electricity price strategies are generated (e.g., setting low prices during peak power generation periods to incentivize electricity consumption and setting high prices during off-peak power generation periods to suppress electricity consumption).

[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: By integrating weather forecast data, historical electricity load data, and new energy power generation information, this invention achieves accurate multi-timescale prediction of regional new energy power generation and load demand; furthermore, based on the prediction results, it simulates the grid operation status in advance in the simulation model, accurately locating potential voltage limit exceedance risks; finally, relying on optimization algorithms, it generates and executes control strategies before voltage problems occur, transforming traditional reactive remediation into proactive defense, fundamentally avoiding voltage limit exceedances, and significantly improving the adaptability and operational safety of the distribution network to high-proportion new energy access. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the first part of a method for intelligent control of power distribution networks based on load characteristics, provided in an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of the second part of a method for intelligent control of power distribution networks based on load characteristics, provided in an embodiment of the present invention.

[0019] Figure 3 This is a schematic flowchart of the third part of an intelligent control method for power distribution networks based on load characteristics, provided in an embodiment of the present invention.

[0020] Figure 4 This is a schematic flowchart of the fourth part of an intelligent control method for power distribution networks based on load characteristics, provided in an embodiment of the present invention.

[0021] Figure 5 This is a schematic diagram of the fifth part of a method for intelligent control of power distribution networks based on load characteristics, provided in an embodiment of the present invention.

[0022] Figure 6 This is a schematic diagram of the first part of an intelligent control system for a power distribution network based on load characteristics, provided as an embodiment of the present invention.

[0023] Figure 7 This is a schematic diagram of the second part of an intelligent control system for a power distribution network based on load characteristics, provided as an embodiment of the present invention.

[0024] Figure 8 This is a schematic diagram of the third part of an intelligent control system for power distribution networks based on load characteristics, provided as an embodiment of the present invention.

[0025] Figure 9 This is a schematic diagram of the fourth part of an intelligent control system for power distribution networks based on load characteristics, provided as an embodiment of the present invention.

[0026] Figure 10 This is a schematic diagram of the fifth part of an intelligent control system for a power distribution network based on load characteristics, provided as an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0028] It is understood that the terms "first," "second," etc., used in this application may be used herein to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are used only to distinguish one element from another. For example, without departing from the scope of this application, a first script may be referred to as a second script, and similarly, a second script may be referred to as a first script.

[0029] In one embodiment, such as Figure 1 As shown, a smart control method for distribution networks based on load characteristics includes the following steps: Step S1: Collect historical electricity load data of users in the distribution area, and extract typical production and living load curves and patterns (such as morning and evening peak hours, seasonal changes, etc.) through big data analysis technology to obtain historical electricity load characteristics; obtain all new energy power generation information in the area, and build a new energy power generation information database, which includes power generation, power generation type, and geographical location. Step S2: Access weather forecast data (such as illuminance, wind speed, and temperature), combine it with the acquired new energy power generation information, establish a power generation prediction model, and predict the new energy power generation within a set future time period (such as the next 24 hours); based on historical electricity load characteristics, establish a load prediction model to predict the total electricity load power of users in the region within a set future time period. Step S3: Compare the predicted new energy power generation within a set future time period with the total power load, calculate the net load power of each node (total power load minus new energy power generation), and input the net load power as the injected power into the simulation model of the distribution network to predict the voltage level of each node in the distribution network within a set future time period, locate the risk points and time periods of voltage overruns, and obtain the prediction results. Step S4: Based on the prediction results, generate control strategies using optimization algorithms (such as the safety-constrained optimal power flow algorithm). For identified risk points and time periods where voltage overruns are likely to occur, issue the control strategies to the corresponding control equipment in advance for execution (e.g., adjust the tap position of the on-load tap changer, control the switching of smart capacitors, or guide distributed power sources to adjust reactive power output, etc.).

[0030] The construction of the power generation prediction model first relies on a historical database of renewable energy power generation information, as well as synchronously collected meteorological data (such as sunshine, wind speed, and temperature). Based on this, machine learning algorithms (such as LSTM and XGBoost) are used for training to establish a complex nonlinear mapping relationship between meteorological conditions and power generation. By analyzing fluctuation patterns in historical data, the model learns the output characteristics of different power generation types (such as photovoltaics and wind turbines) under specific weather conditions, thereby enabling high-precision quantitative predictions of power generation within a given future time period.

[0031] Load forecasting models are built upon collected historical electricity load data, extracting typical production and residential load curves and patterns. These models typically combine time series analysis algorithms (such as ARIMA and Prophet) or machine learning models, using historical load values, date type (weekday / holiday), weather temperature, and seasonal variations as key input features for training. By capturing the periodic variations in load on daily, weekly, and yearly scales and their correlation with external factors, the model can predict the overall regional electricity load power within a given future timeframe.

[0032] The core of the simulation model is to establish a digital mirror of the power distribution network, based on the network topology, line parameters (such as resistance and reactance), and nameplate parameters of equipment such as transformers and capacitors. The model uses the net load power of each node as the injected power and calculates key electrical state quantities such as voltage levels and branch power flows for all nodes in the entire network by solving power flow equations based on physical laws (such as using forward-backward substitution or the Newton-Raphson method). This allows for the simulation and prediction of the power grid's operation over a given period of time.

[0033] In one embodiment, such as Figure 2 As shown, a smart control method for distribution networks based on load characteristics also includes: Step S5: Monitor the actual operating data of the power distribution network in real time, compare the actual operating data with the prediction results, and when the deviation between the two exceeds the first set threshold, update the prediction based on the latest actual operating data and regenerate the control strategy to ensure the real-time performance and accuracy of the control; record the deviation data and periodically calibrate the power generation prediction model and load prediction model offline based on the deviation data.

[0034] For example, based on a weather forecast predicting that the photovoltaic power generation in this area will be 800kW between 12:00 PM and 1:00 PM, while the load power will be 500kW, resulting in a net load power of -300kW (power backflow). The simulation model predicts that the voltage at a certain line terminal node will rise to 1.08 pu at this time, close to the voltage limit of 1.10 pu, but not yet exceeding the limit. Therefore, the system does not generate a strong control command in step S4, but only makes a preventative adjustment.

[0035] At 12:00 noon that day, real-time monitoring by the system revealed that the sky had actually turned cloudy, resulting in an actual photovoltaic power generation of only 400kW in the area, while the load power was 480kW, and the actual net load power was -80kW. Simultaneously, the actual voltage at this node was 1.05 pu.

[0036] The system compares the actual operating data (net load -80kW, voltage 1.05 pu) with the predicted results (net load -300kW, predicted voltage 1.08 pu). The calculation reveals that the deviation in net load power far exceeds the first set threshold (e.g., the set threshold is 150kW). The system then updates the prediction, regenerates the control strategy, records the deviation data, and calibrates the model.

[0037] In one embodiment, such as Figure 3 As shown, a smart control method for distribution networks based on load characteristics also includes: Step S6: Continuously compare the real-time power demand value of each node with the predicted demand value based on the output of known power generation equipment (predicted new energy power generation plus thermal power generation) to obtain the power deviation curve. When a node's power deviation curve shows a persistent and significant negative deviation (e.g., the duration reaches 3 days, the deviation exceeds 10%) (i.e., the actual demand is much lower than the theoretical demand calculated from the known load and known power generation equipment), and the negative deviation exceeds the second set threshold, it is determined that a new new energy power generation equipment has been put into operation on the line to which the node belongs. After the new new energy power generation equipment is confirmed, the parameters of the equipment (such as type, capacity, location) are updated to the new energy power generation information database, and the power generation prediction model and simulation model are reconstructed.

[0038] Suppose a power distribution network has a line supplying power to area A. Historical data analysis predicts that the typical daytime load in area A under clear weather conditions is approximately 500kW. A 100kW photovoltaic power station has been registered on this line. Based on the weather forecast (clear weather), the system predicts its daytime power generation will be 95kW.

[0039] Forecasted demand calculation: Forecasted demand at this node = Known load (500kW) - Known generated capacity forecast (95kW) = 405kW. The expected power flowing into this line from the upstream grid is 405kW.

[0040] During several consecutive days of clear weather, monitoring showed that the real-time power demand at this node remained stable at around 250kW during periods of ample daytime sunshine, significantly lower than the predicted demand of 405kW. The power deviation was calculated as: Real-time value (250kW) - Predicted value (405kW) = -155kW. This is a negative deviation. This negative deviation (-155kW) not only persists but its absolute value also exceeds a preset second threshold (e.g., set at 100kW).

[0041] Based on its persistent and significant characteristics, the phenomenon was determined to be neither a random measurement error nor a short-term load fluctuation. The most reasonable explanation is that a new, unreported renewable energy power generation unit has been put into operation on this line. This unknown unit is generating electricity, and its power (approximately 155kW) is offsetting part of the load demand, resulting in the actual power injected from the main grid being far lower than expected. Therefore, the new renewable energy power generation unit was identified, the database was updated, and the model was reconstructed.

[0042] In one embodiment, such as Figure 4 As shown, a smart control method for distribution networks based on load characteristics also includes: Step S7: A new energy power generation equipment information database is preset. The database includes standardized output power characteristic curves of various typical new energy equipment (such as monocrystalline silicon photovoltaic, thin-film photovoltaic, doubly fed wind turbine, etc.) under different environmental conditions (such as light, wind speed, temperature, etc.). When it is determined that a new new energy power generation equipment is put into operation, the current and historical environmental data are synchronized with the observed power deviation curves and matched with the standardized output power characteristic curves. By comparing the fluctuation patterns and environmental response characteristics, one or more inferred new energy power generation equipment types and their confidence levels are output (e.g., "suspected new monocrystalline silicon photovoltaic, confidence level 85%").

[0043] Matching analysis requires extracting key features from the power curves. For example, for photovoltaics, power output is highly synchronized with illuminance, peaking at midday and zero when there is no sunlight; for wind turbines, power output has a cubic relationship with wind speed, with distinct cut-in and cut-out wind speed thresholds. The system uses algorithms (such as dynamic time warping, correlation coefficient calculation, or machine learning classifiers) to calculate the similarity between the observed curve and various standard feature curves at different time scales. Finally, based on the overall match of feature matching, the system outputs one or more most likely device types and provides a quantified confidence level. For example, the observed curve has a correlation of 0.95 with the characteristic curve of monocrystalline silicon photovoltaics, while its correlation with the wind turbine curve is only 0.1, therefore the output is "suspected monocrystalline silicon photovoltaic, confidence level 85%".

[0044] In one embodiment, such as Figure 5 As shown, a smart control method for distribution networks based on load characteristics also includes: Step S8: Based on future weather forecast data and predicted new energy power generation (e.g., 3 days), the supply and demand situation of the power distribution network is judged in advance. When a continuous power generation surplus or shortage is predicted, a dynamic electricity price strategy is generated (e.g., setting a low price during peak power generation periods to incentivize electricity consumption, and setting a high price during off-peak power generation periods to suppress electricity consumption).

[0045] Based on the weather forecast for the next three days, the system predicts the following power grid operation tomorrow (a working day): From 12:00 to 14:00, due to clear weather, regional photovoltaic power generation will have a huge surplus, resulting in a significantly negative net load, which may lead to a risk of voltage spikes exceeding limits. However, from 18:00 to 20:00 in the evening, photovoltaic power generation will drop sharply, while residential load will surge, predicting a severe power shortage.

[0046] The system predicts a sustained and regular generation surplus and shortage tomorrow. To fundamentally smooth the net load curve from the load side, the system automatically generates a dynamic electricity pricing strategy for that future day, for example: Peak power generation period (12:00-14:00): Set a low electricity price, such as 0.3 yuan / kWh, to incentivize users to increase electricity consumption during this period (e.g., start electric vehicle charging, run washing machines, or start non-urgent production processes in factories).

[0047] Off-peak electricity generation period (18:00-20:00): Set peak electricity price, such as 1.2 yuan / kWh, to suppress concentrated electricity consumption by users during this period (for example, encourage users to postpone the use of high-power appliances such as dryers).

[0048] In one embodiment, such as Figure 6 As shown, a power distribution network intelligent control system based on load characteristics includes: Information collection module 1 is used to collect historical electricity load data of users in the distribution area, and extract typical production and living load curves and patterns (such as morning and evening peak hours, seasonal changes, etc.) through big data analysis technology to obtain historical electricity load characteristics; acquire all new energy power generation information in the area, and build a new energy power generation information database, which includes power generation, power generation type, and geographical location. The power generation and load forecasting module 2 is used to access weather forecast data (such as illuminance, wind speed, and temperature), combine it with the acquired new energy power generation information, establish a power generation forecasting model, and predict the power generation of new energy within a set time period (such as the next 24 hours); based on historical electricity load characteristics, it establishes a load forecasting model to predict the total electricity load power of users in the region within a set time period. The voltage over-limit prediction module 3 is used to compare the predicted new energy power generation power with the total power load power within a set future time, calculate the net load power of each node (total power load power minus new energy power generation power), input the net load power as the injected power into the simulation model of the distribution network, predict the voltage level of each node in the distribution network within a set future time, locate the risk points and time periods of voltage over-limit, and obtain the prediction results. The regulation strategy generation and execution module 4 is used to generate regulation strategies based on the prediction results and optimization algorithms (such as the safety constraint optimal power flow algorithm). For the identified risk points and time periods where voltage over-limit will occur, the regulation strategies are issued in advance to the corresponding regulation equipment for execution (e.g., adjusting the tap position of the on-load tap changer, controlling the switching of smart capacitors, or guiding distributed power sources to adjust reactive power output, etc.).

[0049] The purpose of collecting historical electricity load data in the information collection module 1 is to extract load characteristics and understand regional electricity consumption patterns; the purpose of building a new energy power generation information database is to comprehensively understand the current status of distributed power sources.

[0050] The power generation and load forecasting module 2 incorporates weather forecasts and establishes power generation and load forecasting models to quantitatively predict the power grid's operating status within a specified future timeframe. This transforms uncontrollable renewable energy generation and user load from completely random variables into predictable values ​​with certain patterns, providing input for subsequent analysis.

[0051] The voltage over-limit prediction module 3 inputs the predicted net load power into the simulation model to predict node voltage, aiming to proactively locate risk points and time periods before voltage over-limits actually occur. This directly solves the problem of abnormal voltage rises caused by the inability to predict reverse power flow in traditional methods.

[0052] The regulation strategy generation and execution module 4 generates and issues regulation strategies in advance based on prediction results to overcome the slow response speed of traditional regulation equipment. By optimizing the algorithm to calculate the optimal settings and instructing the equipment to act in advance, it realizes the transformation from passive response to active intervention. Moreover, controlling the equipment to act in advance also reduces the requirements on the equipment, and can still avoid voltage overshoot even without meeting the second-level response requirement.

[0053] In one embodiment, such as Figure 7 As shown, a power distribution network intelligent control system based on load characteristics also includes: Model calibration module 5 is used to monitor the actual operating data of the power distribution network in real time, compare the actual operating data with the prediction results, and when the deviation between the two exceeds the first set threshold, update the prediction based on the latest actual operating data and regenerate the control strategy to ensure the real-time performance and accuracy of the control; record the deviation data and periodically calibrate the power generation prediction model and load prediction model offline based on the deviation data.

[0054] In model calibration module 5, when the deviation between the actual operating data and the prediction results exceeds a first set threshold, the prediction and strategy are updated based on the latest data, which can prevent control failure caused by inaccurate initial model predictions. Simultaneously, periodically calibrating the power generation prediction model and load prediction model offline is to continuously improve the long-term prediction accuracy of the models using historical deviation data, thereby achieving system self-optimization.

[0055] In one embodiment, such as Figure 8 As shown, a power distribution network intelligent control system based on load characteristics also includes: The new equipment acquisition module 6 is used to continuously compare the real-time power demand value of each node with the predicted demand value based on the output of known power generation equipment (predicted new energy power generation plus thermal power generation) to obtain the power deviation curve. When a node's power deviation curve shows a continuous and significant negative deviation (i.e., the actual demand is much lower than the theoretical demand calculated from the known load and known power generation equipment), and the negative deviation exceeds the second set threshold, it is determined that a new new energy power generation equipment has been put into operation on the line to which the node belongs. After the new new energy power generation equipment is confirmed, the parameters of the equipment (such as type, capacity, and location) are updated to the new energy power generation information database, and the power generation prediction model and simulation model are reconstructed.

[0056] The new equipment detection module 6 addresses the regulatory challenge of monitoring unregistered renewable energy generation equipment connected to the distribution network. By continuously comparing real-time power demand with predicted demand, a persistent and significant negative deviation exceeding a second preset threshold indicates that new renewable energy generation equipment has been put into operation. This mechanism ensures the system can dynamically detect unplanned changes in network topology and power generation resources and promptly update the renewable energy generation information database to maintain the accuracy of the power generation prediction and simulation models.

[0057] In one embodiment, such as Figure 9 As shown, a power distribution network intelligent control system based on load characteristics also includes: The new equipment matching module 7 is used to preset a new energy power generation equipment information database. The new energy power generation equipment information database includes standardized output power characteristic curves of various typical new energy equipment (such as monocrystalline silicon photovoltaic, thin film photovoltaic, doubly fed wind turbine, etc.) under different environmental conditions (such as light, wind speed, temperature, etc.). When it is determined that a new new energy power generation equipment is put into operation, the current and historical environmental data are synchronized with the observed power deviation curve, and matched and analyzed with the standardized output power characteristic curve. By comparing the fluctuation pattern and environmental response characteristics, one or more inferred new energy power generation equipment types and their confidence levels are output (e.g., "suspected new monocrystalline silicon photovoltaic, confidence level 85%").

[0058] The new equipment matching module 7 is designed to identify the type of new renewable energy power generation equipment as accurately as possible in the absence of reported information. By synchronizing environmental data with the observed power deviation curve and performing matching analysis with standardized output power characteristic curves in a pre-set renewable energy power generation equipment information database, the system can infer the equipment type and its confidence level. This provides key technical parameters for incorporating unknown equipment into the power generation prediction model, completing a closed loop from discovering its existence to identifying its attributes.

[0059] In one embodiment, such as Figure 10 As shown, a power distribution network intelligent control system based on load characteristics also includes: The electricity price control module 8 is used to predict the supply and demand situation of the distribution network in advance based on future weather forecast data and predicted new energy power generation (e.g., 3 days). When it is predicted that there will be a continuous power generation surplus or shortage, a dynamic electricity price strategy is generated (e.g., setting a low price during peak power generation to incentivize electricity consumption, and setting a high price during off-peak power generation to suppress electricity consumption).

[0060] Module 8, the electricity price control module, introduces load-side management methods in addition to source-side regulation. Based on the assessment of future supply and demand, it generates a dynamic electricity price strategy, aiming to incentivize users to adjust their electricity consumption behavior using price signals. Setting low prices during periods of predicted generation surplus can stimulate electricity consumption, while setting high prices during periods of predicted generation shortage can suppress electricity consumption, thereby smoothing the regional net load curve, reducing the risk of voltage exceeding limits at the source, and improving the local absorption capacity of renewable energy generation.

[0061] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps. The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0063] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0064] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A power distribution network intelligent regulation method based on load characteristics, characterized in that, The power distribution network intelligent regulation method based on load characteristics comprises the following steps: Collecting historical power load data of users in the power distribution area, refining typical production and living load curves and rules through big data analysis technology to obtain historical power load characteristics; obtaining all new energy power generation information in the area, constructing a new energy power generation information database, and the new energy power generation information including power generation capacity, power generation type, and geographic location; Accessing weather forecast data, combining the obtained new energy power generation information, establishing a power generation power prediction model to predict the new energy power generation power in a future set time; based on the historical power load characteristics, establishing a load prediction model to predict the overall power load power of users in the area in the future set time; Comparing the predicted new energy power generation power and the overall power load power in the future set time, calculating the net load power of each node, inputting the net load power as injection power into the simulation model of the power distribution network, predicting the voltage level of each node in the power distribution network in the future set time, positioning the risk points and time periods where voltage overrun occurs, and obtaining the prediction results; According to the prediction results, generating a regulation strategy based on an optimization algorithm, for the risk points and time periods where voltage overrun will occur, the regulation strategy is issued in advance to the corresponding adjustment equipment for execution.

2. The method of claim 1, wherein, Further comprising: Real-time monitoring of actual operation data of the power distribution network, comparing the actual operation data with the prediction results, when the deviation exceeds a first set threshold, updating the prediction based on the latest actual operation data and regenerating the regulation strategy to ensure the real-time and accuracy of the control; recording the deviation data and periodically calibrating the power generation power prediction model and the load prediction model based on the deviation data.

3. The method of claim 1, wherein, Further comprising: Continuously comparing the real-time power demand value of each node with the predicted demand value based on the output of the known power generation equipment to obtain a power deviation curve, when a persistent and significant negative deviation is found in the power deviation curve of a node, and the negative deviation exceeds a second set threshold, it is determined that the line to which the node belongs has a new new energy power generation equipment put into operation; after the new new energy power generation equipment is confirmed, the parameters of the equipment are updated to the new energy power generation information database, and the power generation power prediction model and the simulation model are reconstructed.

4. The method of claim 3, wherein, Further comprising: A new energy power generation equipment information library is preset, which records the standardized output power characteristic curves of various types of new energy typical equipment under different environmental conditions; When it is determined that a new new energy power generation equipment is put into operation, the environmental data of the current and historical periods and the observed power deviation curve are synchronized and matched with the standardized output power characteristic curve for matching analysis, and by comparing the fluctuation mode and environmental response characteristics, one or more types of new energy power generation equipment and their confidence are output.

5. The method of claim 1 to 4, wherein, Further comprising: Based on the future weather forecast data and the predicted new energy power generation power, the supply and demand situation of the power distribution network is judged in advance, when it is predicted that there will be a continuous power generation surplus or shortage, a dynamic electricity price strategy is generated.

6. A power distribution network intelligent regulation system based on load characteristics, characterized in that, Comprising: An information collection module is configured to collect historical power consumption load data of users in a power distribution area, extract typical production and living load curves and rules through big data analysis technology, and obtain historical power consumption load characteristics; obtain all new energy power generation information in the area, and construct a new energy power generation information database, wherein the new energy power generation information includes power generation capacity, power generation type, and geographical position; A power generation and load prediction module is configured to access weather forecast data, combine the obtained new energy power generation information, establish a power generation power prediction model, and predict new energy power generation power in a future set time; Based on the historical power consumption load characteristics, a load prediction model is established to predict the overall power consumption load power of users in the area in the future set time; A voltage overrun prediction module is configured to compare the predicted new energy power generation power and the overall power consumption load power in the future set time, calculate the net load power of each node, input the net load power as injection power into a simulation model of the power distribution network, predict the voltage level of each node in the power distribution network in the future set time, locate the risk points and time periods where voltage overrun occurs, and obtain a prediction result; A regulation strategy generation and execution module is configured to generate a regulation strategy based on an optimization algorithm according to the prediction result, and execute the regulation strategy in advance for the identified risk points and time periods where voltage overrun will occur.

7. The load characteristic based power distribution network intelligent regulation system of claim 6, wherein, Further comprising: A model calibration module is configured to monitor actual operation data of the power distribution network in real time, compare the actual operation data with the prediction result, update the prediction and regenerate the regulation strategy based on the latest actual operation data when the deviation between the actual operation data and the prediction result exceeds a first set threshold, to ensure the real-time and accuracy of the control, and record the deviation data and periodically calibrate the power generation power prediction model and the load prediction model offline based on the deviation data.

8. The load characteristic based power distribution network intelligent regulation system of claim 6, wherein, Further comprising: A new device learning module is configured to continuously compare real-time power demand values of each node with predicted demand values based on known power generation device outputs, obtain a power deviation curve, and determine that a line to which the node belongs has a new new energy power generation device put into operation when a persistent and significant negative deviation of the power deviation curve of a certain node is found and the negative deviation exceeds a second set threshold; after the new new energy power generation device is confirmed, parameters of the device are updated to the new energy power generation information database, and the power generation power prediction model and the simulation model are reconstructed.

9. The load characteristic based power distribution network intelligent regulation system of claim 8, wherein, Further comprising: A new device matching module is configured to preset a new energy power generation device information library that records standardized output power characteristic curves of various types of new energy typical devices under different environmental conditions; when it is determined that a new new energy power generation device is put into operation, the environmental data of the current and historical periods and the observed power deviation curve are synchronized and matched with the standardized output power characteristic curves for matching analysis, and one or more types of the presumed new energy power generation devices and their confidence levels are output by comparing fluctuation patterns and environmental response characteristics.

10. The load characteristic based power distribution network intelligent regulation system according to any one of claims 6 to 9, characterized in that, Further comprising: The electricity price regulation module is used for judging the supply-demand situation of the power distribution network in advance based on future weather forecast data and predicted new energy power generation, and generating a dynamic electricity price strategy when it is predicted that there will be a continuous power generation surplus or shortage.

Images