A data collection-based power market transaction dynamic settlement method

Abstract

The present application relates to the field of power system automation, and more particularly to a power market transaction dynamic settlement method based on data collection, comprising: collecting and aligning the operation state of the power generation side, the state of charge of energy storage and environmental characteristics, constructing the power kinetic curve of the power generation side, identifying the fluctuation inertia period of the power purchase side and intercepting environmental characteristics, and extracting the central feature through clustering analysis to correct the period, extracting the basic and extreme power nodes on the kinetic curve based on the corrected period, calculating the coordinate distance in the unified coordinate system, establishing the power coordination interval, using the approximation algorithm to obtain the minimum power deviation, and dynamically allocating the basic and matching power for steady-state clearing. Through multi-source data fusion and intelligent algorithm, the present application significantly improves the accuracy of power transaction settlement, resource allocation efficiency and system operation stability.

Description

Technical Field

[0001] This invention relates to the field of power system automation, and specifically to a dynamic settlement method for power market transactions based on data acquisition. Background Technology

[0002] With the deepening of power system reform, the marketization and refinement of electricity trading have become key pathways to improve energy allocation efficiency. In the electricity market operation system, the settlement process, as the value realization link between the generation and purchasing sides, directly impacts the enthusiasm of market participants and the overall stability of the power system in terms of accuracy and real-time performance. Therefore, constructing a scientific electricity trading settlement mechanism plays a crucial supporting role in balancing grid load, optimizing energy dispatch, and ensuring fair market transactions.

[0003] Among them, the dynamic settlement method based on data acquisition aims to build a settlement model that can flexibly respond to changes in supply and demand by monitoring the operating characteristics, electricity load, and environmental impact parameters of each node in the power system from multiple dimensions. Current technical directions mainly focus on using multi-source data acquisition methods to conduct collaborative analysis of the output capacity of the generation side and the consumption behavior of the electricity purchase side, thereby achieving the technical goal of evolving electricity trading from a static cycle to a dynamic real-time process through refined power allocation and settlement algorithms.

[0004] However, existing technologies largely rely on static settlement models with fixed cycles, making it difficult to adapt to the dynamic fluctuations in electricity demand on the purchasing side and the real-time changes in the operating status of generating units on the generating side. Traditional methods often exhibit isolated characteristics in the data acquisition phase, lacking spatiotemporal alignment and collaborative analysis between the generating side, the purchasing side, and environmental parameters, resulting in inaccurate power generation curves and inconsistent data dimensions. Simultaneously, the identification of electricity demand fluctuation patterns lacks scientific basis, particularly in the division and correction of fluctuation inertia periods, which does not fully consider the guiding role of clustering characteristics, making it difficult to achieve accurate correlation of multi-source heterogeneous data. Furthermore, the lack of a unified coordinate system mapping mechanism in the settlement process leads to insufficient accuracy in establishing power coordination intervals, and the deviation calculation in the power allocation process lacks a dynamic benchmark, making it impossible to adjust the allocation ratio according to real-time operating status. Consequently, the settlement accuracy and efficiency of steady-state clearing in the electricity market cannot meet current requirements. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a dynamic settlement method for electricity market transactions based on data acquisition. This method overcomes the problems in existing technologies, such as isolated acquisition and analysis of monitored electricity data, lack of spatiotemporal alignment and collaborative analysis of generation, purchase, and environmental parameters, and insufficient accuracy and efficiency in predicting electricity data due to the lack of correlation between the fluctuation inertia period division of electricity data affected by the environment and specific environments.

[0006] To achieve the above objectives, the present invention provides a dynamic settlement method for electricity market transactions based on data acquisition, comprising: After collecting and aligning the data on the operating status of the generation-side system, the state of charge of the energy storage system on the power purchase side, and the environmental impact characteristics, the equivalent kinetic energy characteristics of the generation-side unit are obtained based on the operating status of the generation-side system, and the power kinetic energy curve of the generation side is established. Electricity consumption data from the electricity purchase side is collected and processed for fluctuation analysis to identify the basic electricity consumption and the fluctuation inertia period. Based on the fluctuation inertia period, environmental impact characteristic data is extracted and processed to determine several environmental impact characteristic data groups. The environmental impact feature data group is clustered to extract the cluster center features corresponding to each environmental impact feature data group, and the fluctuation inertia period is corrected based on the cluster center features. The power kinetic energy curve is truncated based on the corrected fluctuation inertia period to obtain the basic operating power node and the extreme operating power node on the power generation side. Based on the state of charge of the energy storage system on the power purchase side in the same coordinate system, the rated power coordinate distance and the power extreme value coordinate distance with the rated operating power node and the operating power extreme value node on the power generation side are obtained respectively. Based on the value range relationship of the rated power coordinate distance and the power extreme value coordinate distance, the power coordination interval between the power generation side and the power purchase side is established. Based on the power of the energy storage system on the power purchase side, the rated operating power node and the extreme operating power node on the power generation side, the minimum power generation deviation is obtained. Then, based on the minimum power generation deviation and the corresponding real-time remaining power generation on the power generation side, the base power generation and matching power generation are allocated to perform steady-state clearing, and the dynamic settlement of electricity market transactions is completed.

[0007] Preferably, the step of collecting and aligning the data from the power generation system's operating status, the energy storage system's state of charge on the power purchase side, and environmental impact characteristics, and then obtaining the equivalent kinetic energy characteristics of the power generation unit based on the power generation system's operating status includes: Establish the start time for collecting electricity trading data; Based on the aforementioned start time, the generator units on the generating side are periodically collected and processed to obtain rotor speed, mechanical angular velocity, inertial constant, and operating power, and to establish the operating data characteristics of the generator units on the generating side. Based on the aforementioned start time, the energy storage unit on the power purchase side is periodically collected and processed to obtain the SOC and operating parameters respectively, and to establish the operating data characteristics of the energy storage unit on the power purchase side. The operating data characteristics of the power generation units and the operating data characteristics of the energy storage units on the power purchase side are periodically uniformly processed, and the equivalent kinetic energy characteristics of the power generation units are calculated based on the operating data characteristics of the power generation units. Among them, the environmental impact characteristic is the ambient temperature of the generator unit.

[0008] Furthermore, the process of collecting electricity consumption data from the purchasing side and performing fluctuation analysis to identify the base electricity consumption and the fluctuation inertia period includes: Collect the electricity demand and power demand from the electricity purchaser to establish electricity consumption data for the electricity purchaser; The basic electricity consumption is divided based on the median corresponding to the aforementioned electricity demand. The required power is divided into fluctuation inertia periods based on power changes.

[0009] Furthermore, the environmental impact feature data groups are clustered to extract cluster center features corresponding to each environmental impact feature data group, including: Data density clustering is performed based on the environmental impact feature data set to obtain nodes with different environmental parameter fluctuation characteristics and nodes with different power output responses. Based on the nodes with different environmental parameter fluctuation characteristics and nodes with different power output responses, obtain the corresponding node characteristic values ​​to determine the different center values ​​of environmental impact characteristics and power response respectively. Outlier removal is performed based on the outlier center values ​​of the environmental impact features and the outlier center values ​​of the power response to obtain the cluster center features corresponding to each environmental impact feature data group.

[0010] Furthermore, the correction process for the fluctuation inertia period based on the cluster center characteristics includes: Based on the mean standard deviation of the environmental impact feature dissimilar center values ​​and the power response dissimilar center values ​​corresponding to each cluster center feature, a cluster center verification threshold is established. Based on the comparison results of the number of cluster centers corresponding to each environmental impact characteristic data group with 1, and the comparison results of the deviation of each cluster center with the cluster center verification threshold, it is determined whether the environmental impact characteristic data should be re-trunculated and the fluctuation inertia period should be corrected. The cluster center verification threshold is obtained by averaging the standard deviations of the environmental impact feature dissimilar center values ​​and the power response dissimilar center values ​​corresponding to the features of each cluster center. When the number of cluster centers in any environmental impact feature data group is not 1, and the deviation of each cluster center in this environmental impact feature data group is greater than or equal to the cluster center verification threshold, the fluctuation inertia period is shortened based on the deviation of the cluster centers and the cluster center verification threshold, so as to re-truncate and process the environmental impact feature data. When the number of cluster centers corresponding to each environmental impact feature data group is 1, and the deviation of each cluster center is less than the cluster center verification threshold, the correction is considered complete, and the corrected fluctuation inertia period is output.

[0011] Furthermore, the power kinetic energy curve is truncated based on the corrected fluctuation inertia period to obtain the basic operating power node and the extreme operating power node on the generation side, including: Based on the ratio of the equivalent kinetic energy characteristics of the generator-side unit to the center value of the power response, the threshold for intercepting the power kinetic energy curve is determined, and the power kinetic energy curve of the generator-side is intercepted based on the corrected fluctuation inertia period to obtain several power kinetic energy curve intervals. Based on the power kinetic energy curve interception threshold, the power kinetic energy curve intervals are filtered and deleted, and abnormal power points in each power kinetic energy curve interval that are greater than the power kinetic energy curve interception threshold are removed. The median power value corresponding to each power kinetic energy curve interval is collected to determine the basic operating power node on the power generation side, and the extreme power value corresponding to each power kinetic energy curve interval is collected to determine the extreme operating power node.

[0012] Furthermore, establishing a power coordination interval between the generation side and the power purchase side includes: With the power generation side power value as the X-axis and the state of charge of the energy storage system on the power purchase side as the Y-axis, a unified power coordinate system is established. Based on the unified power coordinate system, coordinate mapping processing is performed on the basic operating power node, the extreme operating power node and the corresponding state of charge of the energy storage system on the power generation side, so as to obtain the rated power coordinate distance and the extreme power coordinate distance respectively. Based on the rated power coordinate distance and power extreme value coordinate distance corresponding to the fluctuation inertia period, the distance value range is determined, and based on the product of the power response dissimilar center value corresponding to the cluster center and the power kinetic energy curve intercept threshold, the reasonable value range threshold of the relative distance is determined. When the rated power coordinate distance and the power extreme value coordinate distance are both within the distance range, and the relative distance between the rated power coordinate distance and the power extreme value coordinate distance is within the reasonable range threshold of the relative distance, the power supply range of the generation side and the power demand range of the purchase side are determined based on the corresponding basic operating power node and the state of charge of the energy storage system on the power purchase side. The power supply interval on the power generation side and the power demand interval on the power purchase side are intersected to obtain the overlapping interval and establish a power coordination interval.

[0013] Furthermore, the step of obtaining the minimum power generation deviation based on the approximation algorithm for the power of the energy storage system on the power purchase side, the rated operating power node on the power generation side, and the extreme operating power node includes: The average value of the upper and lower limit power values ​​of the power coordination interval is divided into the midpoint power value of the interval, and the power deviation calculation benchmark is calculated based on the midpoint power value of the interval and the power demand value corresponding to the basic electricity consumption on the power purchase side. The real-time power of the energy storage system on the power purchase side is calculated based on the operating parameters corresponding to the operating data characteristics of the energy storage unit on the power purchase side, and the power values ​​corresponding to the rated operating power nodes and the extreme operating power nodes on the power generation side are collected respectively. Based on the real-time power of the energy storage system on the power purchase side, the power value corresponding to the rated operating power node on the power generation side, and the power value corresponding to the extreme operating power node on the power generation side, the parameter dimensions are uniformly processed under a unified power coordinate system. Based on the least squares approximation algorithm and the power deviation calculation benchmark, the power deviation values ​​between the real-time power of the energy storage system on the power purchase side and the rated operating power node and the extreme operating power node on the power generation side are calculated respectively to establish a power deviation dataset. Based on the threshold of the power kinetic energy curve, outlier deletion processing is performed on the power deviation dataset, and the minimum power deviation data corresponding to the power deviation dataset is obtained as the minimum power generation deviation value.

[0014] Furthermore, the steady-state clearing based on the minimum power generation deviation and the corresponding real-time remaining power generation allocation base power and matching power generation on the power generation side includes: The difference between the operating power in the power generation side unit operation data characteristics and the power corresponding to the basic operating power node is used to determine the real-time remaining power generation on the power generation side. The power allocation coefficient is determined based on the minimum value of the power generation deviation. Based on the real-time remaining power generation on the power generation side and the power allocation coefficient, basic power generation allocation is performed, and the matching power generation is cleared in a steady state. The power distribution factor ranges from 0.5 to 1.

[0015] Furthermore, the basic power generation allocation process based on the real-time remaining power generation capacity and power allocation coefficient on the power generation side, and the matching power generation capacity for steady-state clearing, includes: Based on the power allocation coefficient and the real-time remaining power generation on the power generation side, the basic power generation and the matching power generation are allocated according to the power demand value corresponding to the basic electricity consumption on the power purchase side. When the basic power generation is greater than the real-time remaining power generation on the power generation side, the real-time remaining power generation on the power generation side is used as the basic power generation and the matching power generation is used as the remaining supplementary power. Determine whether the sum of the base power generation and the remaining supplementary power is less than or equal to the real-time remaining power generation on the power generation side, and whether the difference between the base power generation and the remaining supplementary power is less than or equal to the minimum power generation deviation. If both are true, then steady-state clearing processing is performed based on the basic power generation and the remaining supplementary power. Otherwise, adjust the power allocation coefficient and return to the basic power generation allocation process.

[0016] Compared with the closest existing technology, the present invention has the following advantages: This invention constructs a dynamic response power kinetic energy curve through multi-view structured light scanning data acquisition and alignment processing. It also introduces cluster analysis and density clustering algorithms to intelligently extract environmental impact characteristics and remove outliers, effectively overcoming the influence of the external environment on the settlement model. Based on the data-driven dynamic settlement method, it not only solves the problems of inconsistent and asynchronous data dimensions, but also reduces the need for manual intervention through intelligent algorithms, providing reliable technical support for the optimization and refined operation of the power market.

[0017] Furthermore, by unifying the start time of data collection and performing periodic unified processing, this invention achieves spatiotemporal alignment of parameters on the power generation side, power purchase side, and environment side, thus solving the problem of inconsistent and asynchronous data dimensions.

[0018] Furthermore, this invention introduces cluster analysis and density clustering algorithms to extract the distinct center values ​​of environmental impact characteristics and power response, thereby achieving dynamic correction of the fluctuation inertia period and improving the accuracy of period division.

[0019] Furthermore, this invention establishes a unified power coordinate system and constructs a power coordination interval between the power generation side and the power purchase side by determining the Euclidean distance and the value range, thereby improving the accuracy of supply and demand matching. At the same time, it obtains the minimum value of power generation deviation based on the least squares approximation algorithm and dynamically allocates the basic power generation and the matching power generation accordingly to achieve steady-state clearing optimization. Attached Figure Description

[0020] Figure 1 This is a flowchart of a dynamic settlement method for electricity market transactions based on data acquisition, according to an embodiment of the present invention. Figure 2 This is a logical flowchart of the dynamic settlement method for electricity market transactions based on data acquisition, according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the construction of the power coordinate system and power coordination interval in the dynamic settlement method for electricity market transactions based on data acquisition, according to an embodiment of the present invention. Figure 4This is a flowchart of the dynamic power allocation and steady-state clearing process of the dynamic settlement method for electricity market transactions based on data acquisition, according to an embodiment of the present invention. Detailed Implementation

[0021] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] To facilitate understanding of this invention, it is explained as follows: the power generation side referred to in this invention refers to the power generation side that includes wind power generation, and the power purchase side energy storage unit refers to the user side energy storage unit that includes an energy storage inverter, a battery pack, and a charge and discharge management system. Example

[0024] This invention provides a dynamic settlement method for electricity market transactions based on data acquisition, such as... Figure 1 As shown, it includes: S1. After collecting the operating status of the power generation system, the state of charge of the energy storage system on the power purchase side, and the environmental impact characteristics, and performing data alignment processing, the equivalent kinetic energy characteristics of the power generation unit are obtained based on the operating status of the power generation system, and the power kinetic energy curve of the power generation side is established. S2. Collect electricity consumption data from the electricity purchase side and perform fluctuation analysis to identify the basic electricity consumption and fluctuation inertia period respectively, and extract and process the environmental impact characteristic data based on the fluctuation inertia period to determine several environmental impact characteristic data groups. S3. Cluster the environmental impact feature data group to extract the cluster center features corresponding to each environmental impact feature data group, and correct the fluctuation inertia period based on the cluster center features. S4. The power kinetic energy curve is truncated based on the corrected fluctuation inertia period to obtain the basic operating power node and the extreme operating power node on the power generation side. S5. Based on the state of charge of the energy storage system on the power purchase side in the same coordinate system, obtain the rated power coordinate distance and the power extreme value coordinate distance with the rated operating power node and the operating power extreme value node on the power generation side, respectively, and establish the power coordination interval between the power generation side and the power purchase side based on the value range relationship of the rated power coordinate distance and the power extreme value coordinate distance. S6. Based on the approximation algorithm, obtain the minimum power generation deviation of the power storage system on the power purchase side, the rated operating power node and the extreme value node of the operating power on the power generation side, and perform steady-state clearing based on the minimum power generation deviation and the corresponding real-time remaining power generation on the power generation side, and complete the dynamic settlement of electricity market transactions.

[0025] In this embodiment, a dynamic settlement method for electricity market transactions based on data acquisition is proposed. Based on the PTP time synchronization protocol, it ensures that the sampling clock deviation of all sensor nodes in the acquisition network is controlled within 1 microsecond, and performs periodic sampling processing on the generator-side units based on the start time to establish the operating data characteristics of the generator-side units, specifically including timestamps, speed vectors, angular velocity scalars, inertial constants, and real-time operating power.

[0026] The energy storage unit on the power purchase side is synchronously collected and processed. The state of charge of the energy storage unit is extracted once every 200 milliseconds. The collected parameters include the DC side voltage of the energy storage inverter, the AC side current, the real-time temperature of the battery cluster, and the allowable charging and discharging power as the operating data characteristics of the energy storage unit on the power purchase side. The periodic unified processing is performed based on linear interpolation algorithm and Lagrange polynomial fitting, and all data streams are resampled to a unified 100-millisecond reference period.

[0027] S1 specifically includes: S1-1. Establish the start time for collecting electricity trading data; Specifically, historical electricity trading data is typically backtested periodically to dynamically match the output and demand on the generation and purchasing sides. Preferably, the starting point for electricity trading data collection is set to any point between 3 months and 1 year ago, allowing for analysis of data from the historical collection start point to the current time. Optionally, the starting point for electricity trading data collection can also be set to any point in the present, with data accumulation for a period of time, such as 3 months to 1 year, as long as sufficient data is obtained for data feature learning; this is not limited.

[0028] S1-2. Based on the starting time, periodically collect and process data from the generator-side unit to obtain rotor speed, mechanical angular velocity, inertial constant and operating power, and establish the operating data characteristics of the generator-side unit. Specifically, the rotor speed is the angular velocity of the wind turbine rotor, the mechanical angular velocity is the actual mechanical angular velocity of the generator rotor, the inertial constant is the ratio of the kinetic energy stored in the rotor to the rated capacity at the rated speed, and the operating power is the active power output value of the generator set at the current moment.

[0029] S1-3. Based on the starting time, periodically collect data from the energy storage unit on the power purchase side to obtain the SOC and operating parameters, and establish the operating data characteristics of the energy storage unit on the power purchase side. Specifically, SOC stands for State of Charge, and its operating parameters include the DC-side voltage, AC-side current, real-time temperature of the battery cluster, and allowable charging and discharging power of the energy storage inverter on the power purchase side. These parameters can be obtained through conventional technical means.

[0030] S1-4. Perform periodic unified processing on the operating data characteristics of the generator-side units and the operating data characteristics of the energy storage units on the power purchase side, and calculate the equivalent kinetic energy characteristics of the generator-side units based on the operating data characteristics of the generator-side units. Among them, the environmental impact characteristic is the ambient temperature of the generator unit.

[0031] Specifically, the equivalent kinetic energy characteristic of the generator unit is calculated based on the rotor speed, mechanical angular velocity, and inertial constant in the generator unit's operating data characteristics. It is used to characterize the equivalent kinetic energy level stored in the rotating components of the generator unit under the current operating state. The calculation formula is as follows:

[0032] Where E is the equivalent kinetic energy characteristic of the generator unit, H is the inertial constant, P is the rated power, ω is the mechanical angular velocity, and ω rated The rated mechanical angular velocity is the rated angular velocity of the generator rotor.

[0033] Specifically, based on the equivalent kinetic energy characteristics of the generator-side units calculated in time series, a power kinetic energy curve for the generator-side is generated according to the time series.

[0034] S2 specifically includes: S2-1. Collect the electricity demand and power demand of the electricity purchaser to establish electricity consumption data of the electricity purchaser. S2-2. Based on the median corresponding to the required electricity consumption, the basic electricity consumption is divided; S2-3. Divide the required power into fluctuation inertia periods according to the same power change.

[0035] Specifically, the cumulative daily electricity consumption of the purchasing side within the historical period is collected as the demand electricity consumption, and the corresponding average power is collected as the demand power to establish the electricity consumption data of the purchasing side. After sorting all the collected demand electricity consumption data according to their numerical values, the median is taken. The electricity value corresponding to the median is divided into basic electricity consumption, which is used to characterize the rigid electricity demand of the purchasing side under normal operating conditions. The time series of demand power is analyzed point by point to identify the stages of power increase, decrease and stability. The time interval with the same power change trend is divided into fluctuation inertia period. Each period is used to characterize a complete fluctuation process of the electricity consumption behavior of the purchasing side, including the fluctuation start time, duration and change amplitude.

[0036] In one specific embodiment, the cumulative daily electricity consumption (e.g., the electricity consumption on a certain day is 1200kWh) of the electricity purchase side over a historical one-year (365-day) period is collected as the demand electricity consumption, and the average daily power (e.g., the average power from 0:00 to 23:59 is 50kW) is collected as the demand power. The two are associated and stored to form the electricity consumption data of the electricity purchase side. Sort all electricity demand data (365 values ​​in total) from smallest to largest, and take the 183rd value (e.g., 1100kWh) as the median, which is used to define the basic electricity demand. A point-by-point slope analysis of the power demand time series is performed: if the power demand remains stable at 30kW±2kW from day 1 to day 45, it is considered a stable phase; if the power demand continues to rise at an average rate of 5kW / day to 80kW from day 46 to day 110, it is considered an upward phase; if the power demand fluctuates between 78-82kW from day 111 to day 185, it is considered a stable phase; if the power demand continues to fall at an average rate of 4kW / day to 45kW from day 186 to day 283, it is considered a downward phase; if the power demand rises to 70kW and then falls again from day 284 to day 365, the time points corresponding to 70kW are used to segment the time series, corresponding to the upward and downward phases respectively. Time intervals with continuous and identical trends are divided into independent fluctuation inertia cycles. For example, a sustained upward interval from 6:00 to 10:00 is divided into a complete upward fluctuation inertia cycle with a duration of 4 hours, and a sustained downward interval from 14:00 to 18:00 is divided into a downward fluctuation inertia cycle with a duration of 4 hours. The stable phase can be further subdivided based on its duration. Specifically, in the stable phase, the difference between the maximum and minimum demand power within the time period is set to be less than or equal to 20% of the average demand power within that time period. A minimum duration of 5 days is also set for each fluctuation inertia cycle. Fluctuation inertia cycles with a duration of less than 5 days are merged with the shorter duration of their adjacent fluctuation inertia cycles to eliminate data abrupt changes.

[0037] Understandably, after merging the oscillation inertia cycles, if adjacent oscillation inertia cycles are of the same type, such as both being in a stable phase, both being in an upward phase, or both being in a downward phase, then oscillation inertia cycles of the same type will be further merged until adjacent oscillation inertia cycles are of different types.

[0038] Subsequently, the start time, duration, and power change amplitude (e.g., +50kW for the rising cycle and -35kW for the falling cycle) of each fluctuating inertia cycle are recorded for subsequent processing of environmental characteristic data. This is used to obtain environmental impact characteristic data within each fluctuating inertia cycle as a set of environmental impact characteristic data groups, and to determine each set of environmental impact characteristic data corresponding to the fluctuating inertia cycle.

[0039] S3 specifically includes: Data density clustering is performed based on the environmental impact feature data set to obtain nodes with different environmental parameter fluctuation characteristics and nodes with different power output responses. Based on the nodes with different environmental parameter fluctuation characteristics and nodes with different power output responses, obtain the corresponding node characteristic values ​​to determine the different center values ​​of environmental impact characteristics and power response respectively. Outlier removal is performed based on the outlier center values ​​of the environmental impact features and the outlier center values ​​of the power response to obtain the cluster center features corresponding to each environmental impact feature data group.

[0040] Specifically, the correction process for the fluctuation inertia period based on the cluster center characteristics includes: S3-1. Establish a cluster center verification threshold based on the mean standard deviation of the environmental impact feature dissimilar center values ​​and the power response dissimilar center values ​​corresponding to the features of each cluster center. S3-2. Based on the comparison results of the number of cluster centers corresponding to each environmental impact characteristic data group with 1, and the comparison results of the deviation of each cluster center with the cluster center verification threshold, determine whether to re-truncate the environmental impact characteristic data and correct the fluctuation inertia period. Wherein, the cluster center verification threshold is the average standard deviation of the environmental impact feature dissimilar center value and the power response dissimilar center value corresponding to each cluster center feature; S3-3. When the number of cluster centers in any environmental impact feature data group is not 1, and the deviation of each cluster center in this environmental impact feature data group is greater than or equal to the cluster center verification threshold, the fluctuation inertia period is shortened based on the deviation of the cluster centers and the cluster center verification threshold, so as to re-truncate the environmental impact feature data. S3-4. When the number of cluster centers corresponding to each environmental impact characteristic data group is 1, and the deviation of each cluster center is less than the cluster center verification threshold, the correction is completed, and the corrected fluctuation inertia period is output.

[0041] Specifically, the density-based clustering analysis algorithm can calculate the Euclidean distance between data points in a multidimensional feature space for each environmental impact feature data group. By setting the neighborhood radius and minimum number of members, it can identify nodes with different environmental parameter fluctuation characteristics, representing special moments such as sudden temperature changes or abnormal sensor sampling. At the same time, it can map the generator power in the same time period to the feature space, extract nodes with different power output responses, and identify the mapping weight between environmental fluctuations and power response by performing spatiotemporal correlation processing on these two types of nodes. In addition, it can obtain the center values ​​of different environmental impact characteristics and different power response values.

[0042] Specifically, for each environmental impact characteristic data group, a density clustering analysis algorithm is used to calculate the Euclidean distance between each environmental impact characteristic data point within the group and the data points within other groups. The maximum Euclidean distance is determined, and the radius corresponding to the minimum Euclidean distance covering 90% of the data in the group is set as the compliance radius. Data points exceeding the compliance radius are recorded as nodes with dissimilar environmental parameter fluctuation characteristics. After removing these nodes, the cluster centers for each environmental impact characteristic data group are determined, and their corresponding center values ​​are recorded as the environmental impact characteristic dissimilar center values.

[0043] Simultaneously, the generator power on the power generation side is divided into data groups based on the fluctuation inertia period to determine several generator power characteristic groups. Here, generator power refers to the average daily operating power. The data processing method within each group is the same as that for the environmental impact characteristic data group. In this way, several generator power exceeding the compliance radius are selected and recorded as power output response dissimilar nodes. The data in the environmental impact characteristic data group corresponding to the power output response dissimilar nodes are synchronously deleted. At the same time, the data value corresponding to the cluster center obtained within the group after deleting the power output response dissimilar nodes is recorded as the power response dissimilar center value.

[0044] like Figure 2 As shown, based on the environmental impact feature dissimilar center values ​​(e.g., temperature fluctuation anomaly center value 23.5℃) and power response dissimilar center values ​​(e.g., power output anomaly fluctuation center value 15.2MW) corresponding to the features of each cluster center, the standard deviation of the data within each group is calculated. The ratio of the smaller standard deviation to the corresponding center value is used to determine the cluster center validation threshold. Specifically, the ratio of the smaller standard deviation to the corresponding center value is recorded as the threshold coefficient, and the cluster center validation threshold is twice the product of the corresponding center value and the threshold coefficient. The number of cluster centers within each environmental impact feature data group is counted, and the average deviation between any cluster center and other cluster centers within the group is calculated. The average deviation is then compared with the cluster center validation threshold. When the number of cluster centers in any environmental impact feature data group is not 1, and the deviation of each cluster center is greater than or equal to the cluster center verification threshold, it is determined that the current fluctuation inertia period is too long, resulting in a multi-center discrete correspondence between the environment and power response. The fluctuation inertia period is shortened proportionally according to the ratio of the deviation of the cluster center to the verification threshold (for example, when the deviation is 1.5 times the verification threshold, the period is shortened to 2 / 3 of the original, thus dividing one fluctuation inertia period into two fluctuation inertia periods). The environmental impact feature data is then re-processed based on the shortened fluctuation inertia period. When the number of cluster centers for all environmental impact feature data groups is 1, or the deviation of each cluster center is less than the cluster center verification threshold, the environment and power response within each data group have a unique and stable coupling center, the division accuracy of the fluctuation inertia period meets the requirements, the correction is completed, and the corrected fluctuation inertia period is output for subsequent processing.

[0045] In one specific embodiment, after completing the density clustering analysis of the environmental impact feature data groups, it is set that a total of 3 environmental impact feature data groups have been acquired. Cluster analysis of the first dataset yielded two cluster centers with distinct environmental impact characteristics of 23.5℃ and 24.1℃, and distinct power response values ​​of 15.2MW and 15.8MW, respectively. The second dataset yielded one cluster center with a distinct environmental impact characteristic of 22.8℃ and a distinct power response value of 14.5MW. The third dataset yielded three cluster centers with distinct environmental impact characteristics of 23.1℃, 23.9℃, and 24.5℃, and distinct power response values ​​of 14.8MW, 15.4MW, and 16.1MW, respectively. The standard deviations of the environmental impact characteristic dissimilar center values ​​and their within-group data, and the standard deviations of the power response dissimilar center values ​​and their within-group data are calculated based on all cluster centers, and the average of the standard deviations of each group is taken as the cluster center validation threshold. The data groups were subjected to singleness verification. The number of cluster centers in the first data group was 2 (not 1), and the deviation of each cluster center within it (approximately 0.85 based on Euclidean distance) was greater than the cluster center verification threshold (0.50). Therefore, it was determined that the current fluctuation inertia period was too long, resulting in a multi-center discrete phenomenon in the correspondence between the environment and power response. The fluctuation inertia period was shortened from the original 4 hours to 2.35 hours, and the environmental impact characteristic data was re-extracted for cluster analysis. The second data group has 1 cluster center and 0 deviation, which is less than the cluster center verification threshold. Therefore, the current data group is deemed to meet the requirements. The third data group has 3 cluster centers, and the maximum deviation of each cluster center (1.05) is greater than the cluster center verification threshold (0.50). The fluctuation inertia period is shortened from 4 hours to 2 hours and the data is re-processed. If the number of cluster centers in each data group is 1, or the deviation of each cluster center is less than the cluster center verification threshold, then the correction is considered complete, and the corrected fluctuation inertia period is output for subsequent processing of the power kinetic energy curve.

[0046] S4 specifically includes: S4-1. Based on the ratio of the equivalent kinetic energy characteristics of the generator-side unit to the center value of the power response, the power kinetic energy curve is divided into threshold values, and the power kinetic energy curve of the generator-side is processed based on the fluctuation inertia period to obtain several power kinetic energy curve intervals. S4-2. Based on the power kinetic energy curve interception threshold, the power kinetic energy curve intervals are filtered and deleted from each power kinetic energy curve interval if the abnormal power points are greater than the power kinetic energy curve interception threshold. S4-3. Collect the median power value corresponding to the several power kinetic energy curve intervals to divide them into basic operating power nodes on the power generation side, and collect the extreme power values ​​corresponding to the several power kinetic energy curve intervals to divide them into operating power extreme value nodes.

[0047] Specifically, the power kinetic energy curve is segmented using the fluctuation inertia period as the sampling step size to obtain several local power kinetic energy curve intervals. Within each interval, data cleaning is performed based on the power kinetic energy curve truncation threshold to remove outliers where the power fluctuation slope exceeds the threshold. The median of the remaining power value within each interval is collected and defined as the baseline operating power node on the generation side. Simultaneously, the maximum and minimum power values ​​within each interval are extracted and designated as operating power extreme value nodes. The power kinetic energy curve truncation threshold corresponding to each power kinetic energy curve interval is determined based on the slope of the environmental impact characteristic interval curve formed by the time sequence of data in the environmental impact characteristic data group corresponding to the power kinetic energy curve interval.

[0048] In one specific embodiment, the environmental impact characteristic data of each dimension in the environmental impact characteristic data group corresponding to a certain power kinetic energy curve interval are respectively formed into environmental impact characteristic interval curves according to time sequence. The data change rate corresponding to each collection point in the environmental impact characteristic interval curve is calculated. The data change rate is divided by the data value of the corresponding collection point to obtain the quantization ratio. Under each dimension, the intersection of the range formed between the maximum and minimum values ​​of the quantization ratio calculated based on the environmental impact characteristic interval curve under each dimension is obtained to obtain the intersection range of the quantization ratio. The two endpoints of the intersection range are determined and denoted as the power kinetic energy curve interception threshold.

[0049] Furthermore, within any power-kinetic energy curve interval, the rate of change of data at each data acquisition point is calculated, and the quantization ratio of each acquisition point is calculated based on the corresponding data value. This ratio is then compared with the data interval containing the corresponding power-kinetic energy curve cutoff threshold. If the value exceeds the interval containing the power-kinetic energy curve cutoff threshold, it is recorded as an abnormal power point and deleted.

[0050] After deleting outliers, the remaining data in each power kinetic energy curve interval is analyzed. The median of the data value in the interval is taken as the basic operating power node of the power generation side within the current fluctuation inertia period. The minimum and maximum values ​​of the data value in the interval are taken as the lower limit extreme value and upper limit extreme value of the operating power extreme value node, respectively, so as to obtain the set of basic operating power nodes and operating power extreme value nodes for each interval of the power generation side.

[0051] S5 specifically includes: S5-1. Establish a unified power coordinate system with the power generation side power value as the X-axis and the state of charge of the energy storage system on the power purchase side as the Y-axis. Based on the unified power coordinate system, perform coordinate mapping processing on the basic operating power node, the extreme operating power node and the corresponding state of charge of the energy storage system on the power generation side, respectively. Calculate the Euclidean distance between the basic operating power node, the extreme operating power node and the corresponding state of charge of the energy storage system on the power purchase side in the unified power coordinate system. Record the Euclidean distance corresponding to the basic operating power node as the rated power coordinate distance, and record the Euclidean distance corresponding to the extreme operating power node as the power extreme coordinate distance. S5-2. Based on the rated power coordinate distance and power extreme value coordinate distance corresponding to the fluctuation inertia period, the distance value range is divided into intervals, and based on the product of the power response dissimilar center value corresponding to the cluster center and the threshold of the power kinetic energy curve, the relative distance reasonable value range threshold is divided. S5-3. When the rated power coordinate distance and the power extreme value coordinate distance are both within the distance range, and the relative distance between the rated power coordinate distance and the power extreme value coordinate distance is within the reasonable range threshold of the relative distance, the generation side power supply range and the power purchase side power demand range are divided based on the corresponding basic operating power node and the charge state of the energy storage system on the power purchase side. S5-4. Perform intersection calculations on the power supply interval on the power generation side and the power demand interval on the power purchase side to obtain overlapping intervals and establish power coordination intervals.

[0052] In this embodiment, a dynamic settlement method for electricity market transactions based on data acquisition is described, such as... Figure 3As shown, constructing a two-dimensional unified power coordinate system specifically includes: in the current coordinate system, using the power generation value as the X-axis and the state of charge (SOC) of the energy storage system on the purchasing side as the Y-axis, performing dimensionless processing; normalizing the power generation power to the rated capacity of the unit; and using the SOC of the energy storage system as a percentage value between 0 and 1, mapping the basic operating power node and the extreme operating power node of the power generation side to the current coordinate system to form feature points. Simultaneously, mapping the real-time collected SOC of the energy storage system on the purchasing side to target points; and obtaining the rated power coordinate distance and the extreme power coordinate distance by calculating the Euclidean distance between the feature points and the target points, based on the fluctuation inertia. The historical distance distribution within the period is used to define the distance value range. Simultaneously, the power response dissimilarity center value corresponding to the cluster center is multiplied by the threshold value of the power kinetic energy curve to obtain the reasonable value range threshold for relative distance. The system then performs a logical judgment: when both the rated power coordinate distance and the power extreme value coordinate distance fall within the preset value range, and the relative deviation between them is less than the reasonable value range threshold, it indicates that the current power generation and purchase sides have a strong foundation for coupling and coordination. Based on the functional mapping relationship between the basic operating power node and the current state of charge, the power supply range on the generation side and the power demand range on the purchase side are divided. The intersection of these two ranges is calculated, and their overlapping part is extracted to establish a power coordination range. The current range is defined as the range within which power balance can be achieved within the current settlement period.

[0053] S6 specifically includes: S6-1. The average value of the upper and lower limit power values ​​of the power coordination interval is divided into the midpoint power value of the interval, and the power demand value corresponding to the basic electricity consumption on the power purchase side is used to calculate and obtain the power deviation calculation benchmark. S6-2. Calculate and obtain the real-time power of the energy storage system on the power purchase side based on the operating parameters corresponding to the operating data characteristics of the energy storage unit on the power purchase side, and collect the power values ​​corresponding to the rated operating power nodes and the extreme operating power nodes on the power generation side respectively. S6-3. Based on the real-time power of the energy storage system on the power purchase side, the power value corresponding to the rated operating power node on the power generation side, and the power value corresponding to the extreme operating power node on the power generation side, the parameter dimensions are uniformly processed under a unified power coordinate system. Based on the least squares approximation algorithm and the power deviation calculation benchmark, the power deviation values ​​between the real-time power of the energy storage system on the power purchase side and the rated operating power node and the extreme operating power node on the power generation side are calculated respectively to establish a power deviation dataset. S6-4. Based on the threshold of the power kinetic energy curve, perform outlier deletion processing on the power deviation dataset, and obtain the minimum power deviation data corresponding to the power deviation dataset as the minimum power generation deviation value.

[0054] S6-5. Based on the difference between the operating power in the characteristics of the generator unit's operating data and the power corresponding to the basic operating power node, determine the real-time remaining generating power on the generator side. S6-6. Determine the power allocation coefficient based on the minimum value of the power generation deviation; S6-7. Based on the real-time remaining power generation on the power generation side and the power allocation coefficient, perform basic power generation allocation processing, and match the power generation for steady-state clearing. The power distribution factor ranges from 0.5 to 1.

[0055] Specifically, based on the minimum deviation of power generation, an inverse proportional function is used to obtain the power allocation coefficient, which is calculated as follows:

[0056] Where η is the power distribution coefficient, its value ranges from 0.5 to 1, and ΔP min The minimum deviation of power generation is defined as α, and β are preset sensitivity adjustment operators.

[0057] In one specific embodiment, the upper and lower limit power values ​​of the power coordination interval are set to 45 MW and 68 MW, respectively, and the corresponding average value is taken as the midpoint power value of the interval, which is 56.5 MW. The larger of the current midpoint power value and the power demand value of 52 MW corresponding to the basic electricity consumption on the power purchase side (i.e., 56.5 MW) is used as the power deviation calculation benchmark. Based on the operating parameters (e.g., AC side current of energy storage inverter 120 A, DC side voltage 700 V) in the operating data characteristics of the energy storage unit on the power purchase side, the real-time power of the energy storage system is calculated to be 0.084 MW. At the same time, the power value corresponding to the rated operating power node on the power generation side is 55 MW, and the upper limit value is 68 MW and the lower limit value is 48 MW corresponding to the extreme value node of the operating power. Based on a unified power coordinate system, the above parameters are normalized and dimension-unified. Based on the least squares approximation algorithm, with a power deviation calculation benchmark of 56.5 MW as the target, the deviation values ​​between each power value and the benchmark are calculated. The resulting power deviation dataset includes the rated power node deviation of 1.5 MW, the upper limit of the extreme value node deviation of 11.5 MW, the lower limit of the extreme value node deviation of 8.5 MW, and the real-time power deviation of the energy storage system of 56.416 MW. Based on the threshold of the power kinetic energy curve (e.g., the power change boundary corresponding to 15 s is converted to 5 MW), outlier deletion processing is performed on the power deviation dataset. Data points with deviations greater than 5 MW are removed, and the minimum deviation value of 1.5 MW is retained as the minimum value of power generation deviation. Based on the difference between the current operating power of 58 MW and the basic operating power node of 55 MW in the characteristics of the generator unit operation data, the real-time remaining generating power on the generator side is calculated to be 3 MW. Based on the minimum power generation deviation of 1.5 MW, and with sensitivity adjustment operators α=2 and β=0.5, the power allocation coefficient η=2 / (1+2.117)=0.642, with the value range limited to 0.5 to 1, is obtained as 0.642. Based on the real-time remaining power generation of 3 MW on the generation side and the power allocation coefficient of 0.642, the basic power generation is allocated, resulting in a basic power generation of 3 MW×0.642=1.926MW and a matching power generation of 3MW-1.926MW=1.074MW. The sum of the basic power generation and the matching power generation is 3MW, which is not greater than the real-time remaining power generation on the generation side (which is equal). The difference between the two is 1.926MW-1.074MW=0.852MW, which is less than the minimum power generation deviation of 1.5MW, satisfying the dual constraint conditions. Therefore, based on the basic power generation of 1.926MW and the matching power generation of 1.074MW, steady-state clearing is performed to complete the dynamic settlement of electricity market transactions.

[0058] S6-7 specifically includes: S6-7-1. Based on the power allocation coefficient and the real-time remaining power generation on the power generation side, the basic power generation and the matching power generation are allocated according to the power demand value corresponding to the basic electricity consumption on the power purchase side. When the basic power generation is greater than the real-time remaining power generation on the power generation side, the real-time remaining power generation on the power generation side is used as the basic power generation and the matching power generation is used as the remaining supplementary power. S6-7-2. Determine whether the sum of the basic power generation and the remaining supplementary power is less than or equal to the real-time remaining power generation on the power generation side, and whether the difference between the basic power generation and the remaining supplementary power is less than or equal to the minimum power generation deviation. If both are true, then steady-state clearing processing is performed based on the basic power generation and the remaining supplementary power. Otherwise, adjust the power allocation coefficient and return to the basic power generation allocation process.

[0059] Specifically, such as Figure 4 As shown, the allocation logic between base power generation and matching power generation includes: Based on the basic electricity demand on the purchasing side, the basic generating capacity is pre-allocated. If the calculated basic generating capacity demand is greater than the real-time remaining generating capacity on the generating side, all remaining generating capacity is divided into basic generating capacity, and the shortfall is treated as surplus supplementary power, which is borne by the matching generating capacity (i.e., the energy storage regulation part). Initiate a closed-loop verification procedure: determine whether the sum of the base generating power and the remaining supplementary power is within the safe allowable range of the unit's real-time remaining generating power, and simultaneously verify whether the difference between the two is not greater than the minimum deviation of generating power. If the above dual constraints are met, the settlement scheme is determined to meet the steady-state clearing requirements, and a formal settlement instruction is issued to complete the dynamic clearing of the electricity market transaction. If not, the system automatically reverts to a previous state, fine-tuning the power allocation coefficient. Preferably, in a single fine-tuning, the power allocation coefficient is reduced to 90% of its original value, and the allocation process is re-entered.

[0060] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0062] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, the instruction means being implemented in the process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1The steps of the function specified in one or more boxes.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A dynamic settlement method for electricity market transactions based on data acquisition, characterized in that, include: After collecting and aligning the data on the operating status of the generation-side system, the state of charge of the energy storage system on the power purchase side, and the environmental impact characteristics, the equivalent kinetic energy characteristics of the generation-side unit are obtained based on the operating status of the generation-side system, and the power kinetic energy curve of the generation side is established. Electricity consumption data from the electricity purchase side is collected and processed for fluctuation analysis to identify the basic electricity consumption and the fluctuation inertia period. Based on the fluctuation inertia period, environmental impact characteristic data is extracted and processed to determine several environmental impact characteristic data groups. The environmental impact feature data groups are clustered to extract the cluster center features corresponding to each environmental impact feature data group, and the fluctuation inertia period is corrected based on the cluster center features. The power kinetic energy curve is truncated based on the corrected fluctuation inertia period to obtain the basic operating power node and the extreme operating power node on the power generation side. Based on the state of charge of the energy storage system on the power purchase side in the same coordinate system, the rated power coordinate distance and the power extreme value coordinate distance with the rated operating power node and the operating power extreme value node on the power generation side are obtained respectively. Based on the value range relationship of the rated power coordinate distance and the power extreme value coordinate distance, the power coordination interval between the power generation side and the power purchase side is established. Based on the power of the energy storage system on the power purchase side, the rated operating power node and the extreme operating power node on the power generation side, the minimum power generation deviation is obtained. Then, based on the minimum power generation deviation and the corresponding real-time remaining power generation on the power generation side, the base power generation and matching power generation are allocated to perform steady-state clearing, and the dynamic settlement of electricity market transactions is completed.

2. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 1, characterized in that, After data alignment processing of the separately collected operating status of the generation-side system, the state of charge of the energy storage system on the power purchase side, and environmental impact characteristics, the equivalent kinetic energy characteristics of the generation-side unit are obtained based on the operating status of the generation-side system, including: Establish the start time for collecting electricity trading data; Based on the aforementioned start time, the generator units on the generating side are periodically collected and processed to obtain rotor speed, mechanical angular velocity, inertial constant, and operating power, and to establish the operating data characteristics of the generator units on the generating side. Based on the aforementioned start time, periodic data acquisition and processing are performed on the energy storage unit on the power purchase side to obtain the SOC and operating parameters respectively, and to establish the operating data characteristics of the energy storage unit on the power purchase side. The operating data characteristics of the power generation units and the operating data characteristics of the energy storage units on the power purchase side are periodically uniformly processed, and the equivalent kinetic energy characteristics of the power generation units are calculated based on the operating data characteristics of the power generation units. Among them, the environmental impact characteristic is the ambient temperature of the generator unit.

3. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 2, characterized in that, The process of collecting electricity consumption data from the purchasing side and performing fluctuation analysis to identify the base electricity consumption and the fluctuation inertia period includes: Collect the electricity demand and power demand from the electricity purchaser to establish electricity consumption data for the electricity purchaser; The basic electricity consumption is divided based on the median corresponding to the aforementioned electricity demand. The oscillation inertia period is divided according to the power change of the required power.

4. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 3, characterized in that, Clustering the environmental impact feature data groups to extract cluster center features for each environmental impact feature data group includes: Data density clustering is performed based on the environmental impact feature data set to obtain nodes with different environmental parameter fluctuation characteristics and nodes with different power output responses. Based on the nodes with different environmental parameter fluctuation characteristics and nodes with different power output responses, obtain the corresponding node characteristic values ​​to determine the different center values ​​of environmental impact characteristics and power response respectively. Outlier removal is performed based on the outlier center values ​​of the environmental impact features and the outlier center values ​​of the power response to obtain the cluster center features corresponding to each environmental impact feature data group.

5. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 4, characterized in that, The correction process for the oscillation inertia period based on the cluster center characteristics includes: Based on the mean standard deviation of the environmental impact feature dissimilar center values ​​and the power response dissimilar center values ​​corresponding to each cluster center feature, a cluster center verification threshold is established. Based on the comparison results of the number of cluster centers corresponding to each environmental impact characteristic data group with 1, and the comparison results of the deviation of each cluster center with the cluster center verification threshold, it is determined whether the environmental impact characteristic data should be re-trunculated and the fluctuation inertia period should be corrected. The cluster center verification threshold is obtained by averaging the standard deviations of the environmental impact feature dissimilar center values ​​and the power response dissimilar center values ​​corresponding to the features of each cluster center. When the number of cluster centers in any environmental impact feature data group is not 1, and the deviation of each cluster center in this environmental impact feature data group is greater than or equal to the cluster center verification threshold, the fluctuation inertia period is shortened based on the deviation of the cluster centers and the cluster center verification threshold, so as to re-truncate and process the environmental impact feature data. When the number of cluster centers corresponding to each environmental impact feature data group is 1, and the deviation of each cluster center is less than the cluster center verification threshold, the correction is considered complete, and the corrected fluctuation inertia period is output.

6. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 5, characterized in that, The power kinetic energy curve is truncated based on the corrected fluctuation inertia period to obtain the basic operating power node and the extreme operating power node on the power generation side, including: Based on the ratio of the equivalent kinetic energy characteristics of the generator-side unit to the center value of the power response, the threshold for intercepting the power kinetic energy curve is determined, and the power kinetic energy curve of the generator-side is intercepted based on the corrected fluctuation inertia period to obtain several power kinetic energy curve intervals. Based on the power kinetic energy curve interception threshold, the power kinetic energy curve intervals are filtered and deleted, and abnormal power points in each power kinetic energy curve interval that are greater than the power kinetic energy curve interception threshold are removed. The median power value corresponding to each power kinetic energy curve interval is collected to determine the basic operating power node on the power generation side, and the extreme power value corresponding to each power kinetic energy curve interval is collected to determine the extreme operating power node.

7. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 6, characterized in that, The establishment of a power coordination interval between the generation side and the power purchase side includes: A unified power coordinate system is established with the power generation side power value as the X-axis and the state of charge of the energy storage system on the power purchase side as the Y-axis. Based on the unified power coordinate system, coordinate mapping processing is performed on the basic operating power node, the extreme operating power node and the corresponding state of charge of the energy storage system on the power generation side to obtain the rated power coordinate distance and the extreme power coordinate distance respectively. Based on the rated power coordinate distance and power extreme value coordinate distance corresponding to the fluctuation inertia period, the distance range is determined, and based on the product of the power response dissimilar center value corresponding to the cluster center and the threshold of the power kinetic energy curve, it is divided into a reasonable range threshold of relative distance. When the rated power coordinate distance and the power extreme value coordinate distance are both within the distance range, and the relative distance between the rated power coordinate distance and the power extreme value coordinate distance is within the reasonable range threshold of the relative distance, the power supply range of the generation side and the power demand range of the purchase side are determined based on the corresponding basic operating power node and the state of charge of the energy storage system on the power purchase side. The power supply interval on the power generation side and the power demand interval on the power purchase side are intersected to obtain the overlapping interval and establish a power coordination interval.

8. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 7, characterized in that, The method of obtaining the minimum power generation deviation based on the approximation algorithm for the power of the energy storage system on the power purchase side, the rated operating power node on the power generation side, and the extreme operating power node includes: The average value of the upper and lower limit power values ​​of the power coordination interval is divided into the midpoint power value of the interval, and the power deviation calculation benchmark is obtained based on the midpoint power value of the interval and the power demand value corresponding to the basic electricity consumption on the power purchase side. The real-time power of the energy storage system on the power purchase side is calculated based on the operating parameters corresponding to the operating data characteristics of the energy storage unit on the power purchase side, and the power values ​​corresponding to the rated operating power nodes and the extreme operating power nodes on the power generation side are collected respectively. Based on the real-time power of the energy storage system on the power purchase side, the power value corresponding to the rated operating power node on the power generation side, and the power value corresponding to the extreme operating power node on the power generation side, the parameter dimensions are uniformly processed under a unified power coordinate system. Based on the least squares approximation algorithm and the power deviation calculation benchmark, the power deviation values ​​between the real-time power of the energy storage system on the power purchase side and the rated operating power node and the extreme operating power node on the power generation side are calculated respectively to establish a power deviation dataset. Based on the threshold of the power kinetic energy curve, outliers are removed from the power deviation dataset, and the minimum power deviation data corresponding to the power deviation dataset is obtained as the minimum power generation deviation.

9. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 8, characterized in that, Steady-state clearing based on the minimum power generation deviation and the corresponding real-time remaining power generation allocation base power and matching power generation on the power generation side includes: The real-time remaining power generation capacity of the generator side is determined based on the difference between the operating power and the power corresponding to the basic operating power node in the generator side unit operation data characteristics. The power allocation coefficient is determined based on the minimum value of the power generation deviation. Based on the real-time remaining power generation on the power generation side and the power allocation coefficient, the basic power generation allocation is performed, and the matching power generation is cleared in a steady state. The power distribution factor ranges from 0.5 to 1.

10. The dynamic settlement method for electricity market transactions based on data acquisition as described in claim 9, characterized in that, Based on the real-time remaining generating power and power allocation coefficient on the generating side, the basic generating power allocation process is performed, and the matching generating power is used for steady-state clearing, including: Based on the power allocation coefficient and the real-time remaining power generation on the power generation side, the basic power generation and the matching power generation are allocated according to the power demand value corresponding to the basic electricity consumption on the power purchase side. When the basic power generation is greater than the real-time remaining power generation on the power generation side, the real-time remaining power generation on the power generation side is used as the basic power generation and the matching power generation is used as the remaining supplementary power. Determine whether the sum of the base power generation and the remaining supplementary power is less than or equal to the real-time remaining power generation on the power generation side, and whether the difference between the base power generation and the remaining supplementary power is less than or equal to the minimum power generation deviation. If both are true, then steady-state clearing processing is performed based on the basic power generation and the remaining supplementary power. Otherwise, adjust the power allocation coefficient and return to the basic power generation allocation process.

Images