Dynamic knowledge graph construction method for power grid resource scheduling

Abstract

The invention belongs to the technical field of power grid resource scheduling, and discloses a dynamic knowledge graph construction method for power grid resource scheduling. By establishing a multi-level scheduling analysis mechanism of scheduling demand identification-power transmission unit screening set construction-target power transmission unit set generation, distance efficiency, voltage quality and user risks are considered, the real-time state of the power grid is dynamically adapted, the accuracy, safety and efficiency of resource allocation are remarkably improved, and the method is suitable for large-scale popularization and application. And a systematic solution is provided for dynamic scheduling of the smart power grid. When the target power transmission unit is identified, the target power transmission unit set is obtained by establishing the multi-level screening rule fusing the geographic position, the power supply capacity, the path node number and the transformation stability, accurate filtering of the power transmission units is achieved, the one-sidedness of traditional single index screening is avoided, and the screening efficiency is improved. And the reliability of the target set and the comprehensive benefit of the scheduling scheme are obviously improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid resource scheduling and relates to a method for constructing a dynamic knowledge graph for power grid resource scheduling. Background Art

[0002] With the advancement of the construction of smart power grids, power grid resource scheduling faces challenges such as diverse user electricity consumption behaviors and dynamic changes in power grid topologies. Traditional scheduling methods mostly rely on static power grid data and single-index analysis, making it difficult to respond in real time to load fluctuations, path losses, and differences in user electricity consumption characteristics, resulting in deficiencies in the efficiency, stability, and risk control of scheduling schemes. How to construct a dynamic scheduling system covering electricity demand identification, power transmission resource screening, path optimization, and priority ranking has become a key issue in improving the utilization rate of power grid resources and power supply reliability.

[0003] Traditional technical solutions do not establish a multi-level scheduling analysis of electricity dispatch demand identification - construction of a screening set of power transmission units - generation of a set of target power transmission units, lacking full-chain quantitative control and dynamic filtering from demand to execution, prone to problems such as lagging demand response, blind path selection, and lack of risk prevention and control, and it is difficult to achieve a dynamic balance of accuracy, security, and efficiency in a complex power grid environment, restricting the improvement of the real-time and refined scheduling capabilities of smart power grids.

[0004] Traditional technical solutions lack an analysis of multi-level screening rules that integrate geographical location, power supply capacity, the number of path nodes, and voltage stability when selecting target power transmission units, which may lead to increased power transmission losses, decreased voltage stability, or impacts on large-area power supply, reducing the rationality of scheduling schemes. Summary of the Invention

[0005] In view of this, to solve the problems proposed in the above background art, a method for constructing a dynamic knowledge graph for power grid resource scheduling is proposed.

[0006] The object of the present invention can be achieved through the following technical solutions: A method for constructing a dynamic knowledge graph for power grid resource scheduling, including: obtaining the electricity consumption behavior data of target electricity-consuming units, including load volatility, peak-valley difference ratio, and load rate, and then identifying the electricity dispatch demands of the target electricity-consuming units.

[0007] When it is identified that the target electricity-consuming unit has electricity dispatch demands, pre-identify power transmission units based on the geographical location of the target electricity-consuming unit to obtain a screening set of power transmission units.

[0008] Obtain the geographical locations and power transmission statuses of each power transmission unit corresponding to the screening set of power transmission units, and construct a power transmission path based on the geographical location of the target electricity-consuming unit and the geographical locations of each power transmission unit.

[0009] Analyze the number of path nodes and the degree of path voltage transformation based on the power transmission path, and at the same time analyze the number of associated power-consuming units and their association degree based on the power transmission state.

[0010] Generate a target power transmission unit set based on the number of path nodes, the degree of path voltage transformation, and the number of associated power-consuming units according to predefined scheduling rules, and further output a power dispatch priority list.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By establishing a multi-level scheduling analysis mechanism of scheduling demand identification - power transmission unit screening set construction - target power transmission unit set generation, the present invention takes into account distance efficiency, voltage quality, and user risk, dynamically adapts to the real-time state of the power grid, significantly improves the accuracy, safety, and efficiency of resource allocation, and provides a systematic solution for the dynamic scheduling of smart grids.

[0012] (2) When identifying the target power transmission unit, the present invention obtains the target power transmission unit set by establishing a multi-level screening rule that integrates geographical location, power supply capacity, the number of path nodes, and voltage transformation stability, realizing the precise filtering of power transmission units. This method avoids the one-sidedness of traditional single-index screening, not only shortens the transmission distance and reduces physical losses, but also ensures power supply quality through voltage transformation consistency, and can also avoid the impact on highly sensitive users, significantly improving the reliability of the target set and the comprehensive benefits of the scheduling scheme. Brief Description of the Drawings

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts based on these drawings.

[0014] Figure 1 It is a schematic diagram of the implementation steps of the method of the present invention.

[0015] Figure 2 It is a flow chart of power consumption scheduling demand identification and judgment corresponding to an embodiment provided by the present invention.

[0016] Figure 3 It is a flow chart of target power transmission unit identification and judgment corresponding to an embodiment provided by the present invention. Detailed Embodiments

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] Please refer to Figure 1 As shown, the present invention provides a method for constructing a dynamic knowledge graph for power grid resource scheduling, including: obtaining the power consumption behavior data of a target power consumption unit, including load volatility, peak-valley difference ratio, and load rate, and then identifying the power consumption scheduling requirements of the target power consumption unit.

[0019] In a preferred embodiment of the present invention, the specific analysis method of the power consumption behavior data is as follows: constructing a load-time change curve based on the historical power consumption records of the target power consumption unit within a preset monitoring period, where the abscissa of the change curve is time and the ordinate is load.

[0020] Based on a preset equal-interval time difference, monitoring points are arranged on the abscissa to obtain the loads corresponding to each monitoring point.

[0021] Calculate the absolute difference between the load of each monitoring point and the load of the previous monitoring point to obtain the load fluctuation amount of each monitoring point, compare the load fluctuation amount with a preset load fluctuation amount threshold, count the number of monitoring points greater than the load fluctuation amount threshold, and then calculate the ratio with the total number of monitoring points to obtain the load volatility.

[0022] Compare the loads corresponding to each monitoring point to obtain the maximum load and the minimum load, calculate the difference between the maximum load and the minimum load to obtain the peak-valley difference of the target power consumption unit, and then calculate the ratio with the maximum load to obtain the peak-valley difference ratio.

[0023] Calculate the average value of the loads corresponding to each monitoring point to obtain the average load of the target power consumption unit, and then calculate the ratio with the maximum load to obtain the load rate.

[0024] In a preferred embodiment of the present invention, the specific method for identifying the power consumption scheduling requirements of the target power consumption unit is as follows: compare the load volatility, peak-valley difference ratio, and load rate of the target power consumption unit with preset thresholds respectively.

[0025] Please refer to Figure 2 As shown. If any one of the load volatility, peak-valley difference ratio, and load rate exceeds the corresponding threshold, it is identified that the target power consumption unit has power consumption scheduling requirements.

[0026] It should be noted that through the concise single-index overlimit rule, the scheduling lag caused by multiple condition restrictions is avoided, ensuring the timely response of the power grid to abnormal power consumption states. The overlimit of the load volatility indicates that the power consumption demand fluctuates violently, and the power supply line or energy storage configuration needs to be dynamically adjusted; the overlimit of the peak-valley difference ratio indicates that the peak-valley load difference is too large, which may cause the power grid to be overloaded, and the power supply side scheduling needs to be optimized; the overlimit of the load rate reflects insufficient or overloaded equipment utilization rate, and the power supply capacity needs to be adjusted.

[0027] It should be noted that the setting basis for the corresponding thresholds of the load volatility, peak-valley difference ratio, and load rate includes: First, the power grid operation standards and equipment safety parameters to ensure system stability; second, the differences in the types of power consumption units, such as the power consumption characteristics of users with different attributes such as industrial and residential; third, the power grid peak shaving ability and resource allocation situation, combined with the power source structure and energy storage capacity; fourth, policy orientation and energy efficiency goals to promote power consumption optimization; fifth, a dynamic adjustment mechanism based on historical data, season and time periods, etc., to ensure the rationality and adaptability of the thresholds and achieve accurate judgment of scheduling requirements.

[0028] When it is identified that the target power consumption unit has a power consumption scheduling requirement, the power supply unit pre-identification is carried out based on the geographical location of the target power consumption unit to obtain the power supply unit screening set.

[0029] In a preferred embodiment of the present invention, the specific construction method of the power supply unit screening set is as follows: Calculate the average value of the loads of each monitoring point to obtain the average load of the current monitoring period.

[0030] Calculate the difference between the rated capacity of the target power consumption unit and the average load to obtain the instant demand load of the target power consumption unit.

[0031] Obtain the geographical locations of each power supply unit, and then compare them with the geographical location of the target power consumption unit to obtain the distances between each power supply unit and the target power consumption unit.

[0032] Compare the available load of each power supply unit with the instant demand load of the target power consumption unit, and at the same time compare the distance between each power supply unit and the target power consumption unit with a preset distance threshold.

[0033] Record the power supply units whose available load is greater than the instant demand load and the distance is less than the preset distance threshold, and then construct the power supply unit screening set.

[0034] It should be noted that the above module screens out power transmission units with dispatch feasibility through the quantitative matching of power demand and power transmission capacity, and combines geographical location. Specifically, the available load capacity of the power transmission unit is compared with the instantaneous demand load capacity of the target power consumption unit, and only the power transmission units with an available load capacity greater than the instantaneous demand load capacity are retained to ensure that the power transmission units have sufficient redundant power to meet the dispatch requirements. At the same time, the distance between the power transmission unit and the target power consumption unit is calculated and compared with the preset distance threshold, and only the power transmission units with a distance less than the preset threshold are retained, and the power transmission units with a short distance are preferentially selected to reduce the power transmission line loss, construction cost and dispatch delay.

[0035] It should be noted that the setting of the preset distance threshold needs to be combined with the power grid layout. For example, if the urban power grid density is high, the threshold can be set to 5 kilometers; the threshold for the rural power grid can be relaxed to 15 kilometers. The essence is to seek a balance between power supply capacity and power transmission cost: too long a distance may lead to high line investment and large losses, and too short a distance may be limited by the local power grid capacity.

[0036] It should be noted that the above content quickly locates feasible objects from all power transmission units in the whole network through double filtering of load capacity and distance, improving the dispatch efficiency.

[0037] Obtain the geographical locations and power transmission statuses of the power transmission units corresponding to the power transmission unit screening set, and construct a power transmission path based on the geographical location of the target power consumption unit and the geographical locations of the power transmission units.

[0038] In a preferred embodiment of the present invention, the specific method for constructing a power transmission path based on the geographical location of the target power consumption unit and the geographical locations of the power transmission units is as follows: identify all lines that can perform power dispatch based on the geographical location of the target power consumption unit and the geographical locations of the power transmission units.

[0039] Obtain the distances corresponding to the lines of each power dispatch, and then compare them, and select the power dispatch line with the smallest distance as the power transmission path.

[0040] It should be noted that on the premise of meeting the power dispatch feasibility, the line with the shortest power transmission distance is selected as the power transmission path to reduce power transmission loss, shorten the dispatch time and reduce the line construction and maintenance costs. The power transmission distance is proportional to the line loss, so the short-distance path is more economical and efficient.

[0041] It should be explained that all feasible lines meet the power transmission capacity requirements, that is, the available load capacity of the power transmission unit has been verified by the screening set, so there is no need to repeatedly check the capacity matching problem, and only focus on distance optimization. In the actual operation process, if new lines are added to the power grid or existing lines fail and are out of service, the set of feasible lines needs to be updated in real time, and the shortest path needs to be recalculated to ensure that the knowledge graph reflects the real-time state of the power grid.

[0042] Analyze the number of path nodes and the degree of path voltage transformation based on the power transmission path, and analyze the number of associated power consumption units and their association degrees based on the power transmission state.

[0043] In a preferred embodiment of the present invention, the specific analysis method for the degree of path voltage transformation: Statistically analyze the rated voltage of the high-voltage side and the rated voltage of the low-voltage side of each transformer in the power transmission path corresponding to each power transmission unit, and then calculate the ratio to obtain the transformation ratio of each transformer.

[0044] Calculate the standard deviation of the transformation ratios of each transformer in the power transmission path corresponding to each power transmission unit to obtain the path voltage transformation index of each power transmission unit.

[0045] In a preferred embodiment, the specific calculation formula for the path voltage transformation index: , where represents the standard deviation of the transformation ratios of each transformer in the power transmission path corresponding to each power transmission unit, represents the power transmission unit number, , represents the number of power transmission units, represents the transformation ratios of each transformer in the power transmission path corresponding to each power transmission unit, represents the average value of the transformation ratios of each transformer in the power transmission path corresponding to each power transmission unit.

[0046] It should be noted that the purpose of analyzing the degree of path voltage transformation: 1. For a path with a small standard deviation of the transformation ratio, the voltage transformation process is more uniform, which can reduce voltage quality problems caused by voltage fluctuations and ensure the safety of electrical equipment. 2. Inconsistent transformation ratios may lead to the superposition of multi-stage voltage transformation losses. For example, the total loss of multiple voltage transformations from high voltage, medium voltage to low voltage is higher than that of single voltage transformation. Screen low-loss paths through the path voltage transformation index to improve the energy efficiency of the power grid.

[0047] Exemplarily, assume that the transformer transformation ratio data of two power transmission paths are as follows: The transformation ratio of path A is 10:1, 10:1, 10:1, and the transformation ratio of path B is 12:1, 8:1, 10:1. The path voltage transformation index of path A is 0, and the path voltage transformation index of path B is 1.63. According to the rule, the voltage transformation degree of path A is better, and the corresponding power transmission unit has more advantages in screening and priority ranking.

[0048] In a preferred embodiment of the present invention, the specific analysis method for the number of associated power consumption units and their association degrees is as follows: Obtain the power consumption units and their loads powered by each power transmission unit.

[0049] Based on the power consumption units with loads greater than the preset load threshold and the pre-identified special power consumption units, construct the corresponding sets of associated power consumption units, and statistically analyze the number of associated power consumption units corresponding to the sets of associated power consumption units.

[0050] It should be noted that the associated electricity-consuming units refer to the electricity-consuming units directly powered by the power-sending unit. A large number of associated electricity-consuming units means that dispatching the power-sending unit may affect more users. Especially when there are multiple high-load or special users, a careful assessment is required.

[0051] It should be noted that the correlation degree measures the degree of electricity dependence between the associated electricity-consuming units and the power-sending unit. The larger the value, the higher the proportion of electricity consumption of the user, and the more significant the impact of dispatching on it.

[0052] The correlation degree of each associated electricity-consuming unit is calculated by taking the ratio of the load of each associated electricity-consuming unit to the output load of the power-sending unit. If the electricity consumption correlation degree of a certain user is large, it means that the power-sending unit has a strong power supply exclusivity for it, and dispatching may lead to a decrease in the power stability of the user, and the power supply stability of the user needs to be guaranteed preferentially.

[0053] Based on the number of path nodes, the degree of path voltage transformation, and the number of associated electricity-consuming units, a set of target power-sending units is generated according to predefined dispatching rules, and a power dispatching priority list is further output.

[0054] In a preferred embodiment of the present invention, the specific method for generating the set of target power-sending units is as follows: The number of path nodes, the path voltage transformation index, and the number of associated electricity-consuming units of each power-sending unit are respectively compared with the preset path node number threshold, path voltage transformation index threshold, and associated electricity-consuming unit number threshold.

[0055] The predefined dispatching rule is: The number of path nodes is less than or equal to the path node number threshold.

[0056] The path voltage transformation index is less than or equal to the path voltage transformation index threshold.

[0057] The number of associated electricity-consuming units is less than or equal to the associated electricity-consuming unit number threshold.

[0058] Please refer to Figure 3 As shown, when the power-sending unit simultaneously meets the three conditions in the predefined dispatching rules, the power-sending unit is identified as a target power-sending unit, and then a set of target power-sending units is constructed.

[0059] It should be noted that the above content excludes high-risk or inefficient paths through the rigid matching of three key indicators with preset thresholds, ensuring that the power-sending units included in the set meet the requirements of dispatching feasibility and stability.

[0060] It should be noted that the number of path nodes refers to the total number of transmission line nodes included in the power transmission path, such as equipment nodes like substations and switch stations. The more the number of nodes, the higher the path complexity, which may lead to an increase in power transmission losses and the probability of faults. The path voltage transformation index refers to the standard deviation of the transformation ratios of each transformer in the power transmission path, reflecting the consistency of voltage transformation. The higher the index, the more unstable the voltage transformation process, which may cause voltage fluctuations or an increase in losses. The number of associated power-consuming units refers to the number of high-load or special power-consuming units supplied by the power transmission unit. The more the number, the greater the scope of dispatching influence and the higher the risk.

[0061] It should be noted that the above content quickly filters out invalid options through hard thresholds, narrows the scope of subsequent priority sorting, and reduces the computational complexity; ensures that the power transmission units within the target set all have the characteristics of low risk and high stability, laying a foundation for generating a reliable dispatching plan.

[0062] It should be noted that the setting basis for the thresholds of the number of path nodes, the path voltage transformation index, and the number of associated power-consuming units: 1. Grid operation efficiency and reliability: The threshold of the number of path nodes refers to the hierarchical limit of the transmission line. Too many nodes are likely to increase losses and fault risks, and it is usually set according to the grid planning standards; the threshold of the path voltage transformation index is based on the requirement of the consistency of transformer transformation ratios. Too high an index may cause voltage fluctuations, and it needs to be set in combination with equipment parameters and power supply quality standards.

[0063] 2. Control of the user influence range: The threshold of the number of associated power-consuming units is based on the dispatching risk tolerance. Too many numbers may cause a large-scale impact, and it needs to be dynamically adjusted in combination with user types to give priority to ensuring critical loads.

[0064] 3. Historical data and simulation verification: By analyzing the index distribution of qualified paths in historical dispatching cases and combining the dispatching effects under different thresholds through grid simulation, the threshold values are optimized to ensure that the selected power transmission units have both stability and feasibility.

[0065] 4. Dynamic adjustment mechanism: The thresholds can be flexibly corrected according to the real-time status such as the peak and valley of the grid load and equipment maintenance. For example, when the load is at a peak, the threshold of the number of nodes is tightened to reduce path losses and ensure power supply efficiency.

[0066] In a preferred embodiment of the present invention, the specific method of the power dispatching priority list is as follows: Calculate the power dispatching feasibility evaluation index of each target power transmission unit according to the number of path nodes, the path voltage transformation index, and the number of associated power-consuming units of each target power transmission unit in the target power transmission unit set by using the power dispatching feasibility evaluation index calculation formula.

[0067] Arrange the power dispatching feasibility evaluation indices in descending order to generate an output power dispatching priority list, which includes the name of the target power transmission unit, geographical location, and power dispatching feasibility evaluation index.

[0068] It should be noted that by arranging the evaluation indices of each target power transmission unit from large to small, the higher the index, the stronger the dispatching feasibility and the higher the priority. Outputting in list form can visually display the differences in dispatching priorities. Through multi-index weighted calculation, the limitations of a single index are avoided, and factors such as path complexity, user impact, and voltage stability are comprehensively balanced.

[0069] It should be noted that when identifying the target power transmission units in the present invention, by establishing a multi-level screening rule that integrates geographical location, power supply capacity, number of path nodes, and voltage transformation stability, a set of target power transmission units is obtained, realizing the precise filtering of power transmission units. This method avoids the one-sidedness of traditional single-index screening, not only shortens the transmission distance and reduces physical losses, but also ensures the power supply quality through voltage transformation consistency, and can also avoid the impact on highly sensitive users, significantly improving the reliability of the target set and the comprehensive benefits of the dispatching scheme.

[0070] In a preferred embodiment of the present invention, the calculation formula of the power dispatching feasibility evaluation index can be: Using the formula Calculate the power dispatching feasibility evaluation indices of each target power transmission unit , represents the number of the target power transmission unit, , represents the number of target power transmission units, where , , respectively represent the number of path nodes, path voltage transformation index, and number of associated power consumption units of each target power transmission unit, represents the maximum value of the association degrees of the associated power consumption units of each target power transmission unit, , , respectively represent the preset reference number of path nodes, reference path voltage transformation index, and reference number of associated power consumption units, respectively represent the influence factors corresponding to the preset number of path nodes, path voltage transformation index, and number of associated power consumption units.

[0071] It should be noted that the construction idea of the above formula: Based on multi-index quantitative weighting and reverse optimization logic, by integrating core dispatching elements such as path complexity, voltage stability, and user impact risk, it is transformed into a comparable comprehensive evaluation index. Specifically: 1. The more path nodes, the more complex the path, and the higher the loss and fault risk. Through Achieve a positive mapping where the fewer the number of implementation nodes, the higher the metric value. 2. The larger the path voltage transformation index, the worse the voltage stability. By Achieve a positive mapping where the smaller the implementation index, the higher the metric value. Specifically, low fluctuations correspond to high stability. 3. The larger the number of electricity-consuming units or the higher the maximum correlation degree, the greater the scope of dispatching influence and the higher the risk. 4. Through dynamic adjustment of weights, the formula is adapted to different dispatching objectives. Through the forward transformation and standardization of reverse indicators, the scientific superposition of indicators in different dimensions is realized, ensuring the rationality and interpretability of the priority ranking.

[0072] It should be noted that the setting basis for the number of reference path nodes, the reference path voltage transformation index, and the number of reference associated electricity-consuming units: 1. Grid planning standards: Refer to the requirements for the levels of transmission lines and transformer configurations in industry specifications. For example, in urban power grids, it is usually required that the number of path nodes does not exceed 10 to control complexity, and the voltage transformation index needs to be lower than 0.8 to ensure voltage stability.

[0073] 2. Historical optimal data: Analyze the index distribution of efficient paths in historical dispatching cases, and take the average value or minimum value as a reference. For example, the average number of nodes in historical high-quality paths is 6 - 8, ensuring that the reference value has practical feasibility.

[0074] 3. Equipment operation parameters: Combine the rated voltage level of the transformer and the load capacity of the transmission line to avoid the reference value exceeding the safety boundary of the equipment. For example, the voltage transformation index refers to the allowable fluctuation range of the transformer turns ratio.

[0075] 4. Risk control objectives: The number of associated electricity-consuming units refers to the tolerance limit of the power grid for the scope of influence of a single dispatching. For example, usually no more than 5 high-load users are allowed to prevent large-area power supply fluctuations.

[0076] 5. Dynamic correction mechanism: Adjust in real time according to the peak and valley of the grid load, the proportion of new energy access, etc. For example, when the load is at a low valley, the reference value of the number of nodes is relaxed to improve equipment utilization rate, ensuring that the reference value adapts to changes in the grid operation state.

[0077] In the above formula, the influence factor corresponding to the number of associated electricity-consuming units needs to be adjusted in real time according to the standby power supply configuration of the associated units and the particularity of the associated electricity-consuming units.

[0078] It should be noted that the setting basis for the influence factors corresponding to the number of path nodes, the path voltage transformation index, and the number of associated electricity-consuming units: The influence factors corresponding to the number of path nodes, the path voltage transformation index, and the number of associated electricity-consuming units are set based on the influence degree of the indicators on dispatching efficiency and safety, the priority of user needs, the real-time state of the power grid, the verification of the effects of historical dispatching cases, and policy guidance. By comprehensively considering the actual effects of various factors on the dispatching results, weights are dynamically allocated to ensure that the multi-index evaluation system conforms to the operation requirements of the power grid, achieving the balanced optimization of path complexity, voltage stability, and user influence risk.

[0079] It should be noted that the present invention establishes a multi-level dispatching analysis mechanism for dispatching demand identification - power transmission unit screening set construction - target power transmission unit set generation, taking into account distance efficiency, voltage quality and user risk, dynamically adapting to the real-time state of the power grid, significantly improving the accuracy, safety and efficiency of resource allocation, and providing a systematic solution for the dynamic dispatching of smart grids.

[0080] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar ways to replace them. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should fall within the protection scope of the present invention.

Claims

1. A method for constructing a dynamic knowledge graph for power grid resource scheduling, characterized in that, Including: Obtain the electricity consumption behavior data of the target electricity-consuming unit, including load volatility, peak-valley difference ratio, and load rate, and identify the electricity dispatching requirements of the target electricity-consuming unit; When it is identified that the target electricity-consuming unit has electricity dispatching requirements, pre-identify the power supply units based on the geographical location of the target electricity-consuming unit to obtain a screening set of power supply units; Obtain the geographical locations and power supply statuses of the power supply units corresponding to the screening set of power supply units, and construct a power supply path based on the geographical location of the target electricity-consuming unit and the geographical locations of the power supply units; Analyze the number of path nodes and the degree of voltage transformation of the path based on the power supply path, and analyze the number of associated electricity-consuming units and their association degrees based on the power supply status; Generate a set of target power supply units based on the number of path nodes, the degree of voltage transformation of the path, and the number of associated electricity-consuming units according to predefined dispatching rules, and output a list of power dispatching priorities.

2. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 1, wherein: The specific analysis method of the electricity consumption behavior data: Construct a load-time change curve based on the historical electricity consumption records of the target electricity-consuming unit within a preset monitoring period, where the abscissa of the change curve is time and the ordinate is load; Arrange monitoring points on the abscissa based on a preset equal-interval time difference, and obtain the load corresponding to each monitoring point; Calculate the absolute difference between the load of each monitoring point and the load of the previous monitoring point to obtain the load fluctuation amount of each monitoring point, compare the load fluctuation amount with a preset load fluctuation amount threshold, count the number of monitoring points greater than the load fluctuation amount threshold, and then calculate the ratio with the total number of monitoring points to obtain the load volatility; Compare the loads corresponding to each monitoring point to obtain the maximum load and the minimum load, calculate the difference between the maximum load and the minimum load to obtain the peak-valley difference of the target electricity-consuming unit, and then calculate the ratio with the maximum load to obtain the peak-valley difference ratio; Calculate the average value of the loads corresponding to each monitoring point to obtain the average load of the target electricity-consuming unit, and then calculate the ratio with the maximum load to obtain the load rate.

3. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 1, characterized in that: The specific method for identifying the electricity dispatching requirements of the target electricity-consuming unit is as follows: Compare the load volatility, peak-valley difference ratio, and load rate of the target electricity-consuming unit with preset thresholds respectively; If any of the load volatility, peak-valley difference ratio, and load rate exceeds the corresponding threshold, it is identified that the target electricity-consuming unit has electricity dispatching requirements.

4. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 2, wherein: The specific construction method of the screening set of power supply units is as follows: Calculate the average load of the current monitoring period by calculating the average value of the loads of each monitoring point; Calculate the immediate demand load of the target electricity-consuming unit by calculating the difference between the rated capacity of the target electricity-consuming unit and the average load; Obtain the geographical locations of each power supply unit, and then compare them with the geographical location of the target electricity-consuming unit to obtain the distance between each power supply unit and the target electricity-consuming unit; Compare the available load of each power supply unit with the immediate demand load of the target electricity-consuming unit, and at the same time compare the distance between each power supply unit and the target electricity-consuming unit with a preset distance threshold; Record the power supply units with available load greater than the immediate demand load and distance less than the preset distance threshold, and then construct a screening set of power supply units.

5. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 1, characterized in that: The specific method for constructing a power supply path based on the geographical location of the target electricity-consuming unit and the geographical locations of the power supply units is as follows: Identify all the lines that can perform power dispatching based on the geographical locations of the target power-consuming units and those of each power-sending unit; Obtain the distances corresponding to the lines of each power dispatching, and then make a comparison to select the power dispatching line with the minimum distance as the power transmission path.

6. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 1, characterized in that: The specific analysis method for the voltage transformation degree of the path: Statistically calculate the rated voltages of the high-voltage side and the low-voltage side of each transformer in the power transmission path corresponding to each power-sending unit, and then calculate the ratio to obtain the transformation ratio of each transformer; Calculate the standard deviation of the transformation ratios of each transformer in the power transmission path corresponding to each power-sending unit to obtain the path voltage transformation index of each power-sending unit.

7. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 6, characterized in that: The specific analysis method for the number of associated power-consuming units and their association degrees is as follows: Obtain the power-consuming units and their loads that are transmitted by each power-sending unit; Based on the power-consuming units with loads greater than the preset load threshold and the pre-identified special power-consuming units, construct the corresponding sets of associated power-consuming units, and statistically calculate the number of associated power-consuming units corresponding to the sets of associated power-consuming units; Calculate the ratio of the load of each associated power-consuming unit to the output load of the power-sending unit to obtain the association degree of each associated power-consuming unit.

8. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 7, wherein: The specific method for generating the set of target power-sending units is as follows: Compare the number of path nodes, the path voltage transformation index, and the number of associated power-consuming units of each power-sending unit with the preset thresholds of the number of path nodes, the path voltage transformation index, and the number of associated power-consuming units respectively; The predefined dispatching rule is: The number of path nodes is less than or equal to the threshold of the number of path nodes; The path voltage transformation index is less than or equal to the threshold of the path voltage transformation index; The number of associated power-consuming units is less than or equal to the threshold of the number of associated power-consuming units; When a power-sending unit simultaneously meets the three conditions in the predefined dispatching rule, identify this power-sending unit as a target power-sending unit, and then construct a set of target power-sending units.

9. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 7, wherein: The specific method for the power dispatching priority list is as follows: Calculate the power dispatching feasibility evaluation index of each target power-sending unit in the set of target power-sending units according to the number of path nodes, the path voltage transformation index, and the number of associated power-consuming units of each target power-sending unit by using the power dispatching feasibility evaluation index calculation formula; Arrange the power dispatching feasibility evaluation indexes in descending order, and then generate an output power dispatching priority list, where the power dispatching priority list includes the name, geographical location, and power dispatching feasibility evaluation index of the target power-sending unit.

10. The dynamic knowledge graph construction method for power grid resource scheduling according to claim 9, wherein: The power dispatching feasibility evaluation index calculation formula can be: Using the formula calculate the power dispatching feasibility evaluation index of each target power transmission unit , represents the number of the target power transmission unit, , represents the number of target power transmission units, where , , respectively represent the number of path nodes, path voltage transformation index and the number of associated power consumption units of each target power transmission unit, represents the maximum value of the association degree of each associated power consumption unit of each target power transmission unit, , , respectively represent the preset reference number of path nodes, reference path voltage transformation index and reference number of associated power consumption units, respectively represent the influence factors corresponding to the preset number of path nodes, path voltage transformation index and number of associated power consumption units; In the above formula, the influence factor corresponding to the number of associated power-consuming units needs to be adjusted in real time according to the standby power supply configuration of the associated units and the particularity of the associated power-consuming units.

Images