A method for coordinated control of power medium and long-term transaction and grid frequency modulation demand

Abstract

This invention relates to the field of collaborative control processing technology, specifically a collaborative control method for medium- and long-term power trading and grid frequency regulation demand. The method includes the following steps: extracting eigenvalues ​​and eigenvectors of the system state matrix, screening low-damped electromechanical oscillation modes on tie lines, calculating oscillation frequencies and damping ratios, and constructing a feature library of key grid oscillation modes. This invention, through the extraction and analysis of eigenvalues ​​and eigenvectors of the system state matrix, screens insufficiently damped electromechanical oscillation modes on tie lines, identifies key areas and modes of grid oscillation risk, and effectively reduces the difficulty of identifying oscillation risks. Utilizing the coupling relationship between the power transmission path defined in the trading contract and the oscillation mode, it constructs a quantitative indicator reflecting the degree of risk in trading execution, revealing the frequency regulation risk that the trading plan may generate under renewable energy fluctuation scenarios, and enabling the early quantification of conflicts between trading execution and frequency regulation demand.

Description

Technical Field

[0001] This invention relates to the field of collaborative control processing technology, and in particular to a collaborative control method for medium- and long-term power transactions and grid frequency regulation needs. Background Technology

[0002] The field of collaborative control processing technology involves multiple independent control entities or systems. In order to achieve the overall goal of a complex system, under various control requirements, resource constraints, environmental changes, and dynamic influences between them, comprehensive control is achieved through coordination, cooperation, and collaborative decision-making to optimize the overall performance and ensure stable and reliable operation of the system.

[0003] Existing technologies, when dealing with complex coordination problems among multiple independent control entities, typically focus only on the simple cooperative relationships between the entities, neglecting the dynamic characteristics within each entity and the impact of changes in the external environment. This is particularly evident in the coupling problem between grid frequency regulation and power trading demand. The lack of sufficient consideration for the identification and dynamic quantification of electromechanical oscillation modes leads to ambiguity in risk positioning and one-sided risk assessment. Furthermore, potential risks between trading and frequency regulation cannot be quantified in a timely manner, resulting in frequent unexpected power fluctuations and frequency regulation capacity shortages during operation. Therefore, improvements are needed. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a coordinated control method for medium- and long-term power trading and grid frequency regulation needs.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for coordinated control of medium- and long-term power trading and grid frequency regulation demand, comprising the following steps:

[0006] The eigenvalues ​​and eigenvectors of the system state matrix are extracted, low-damped electromechanical oscillation modes on the tie line are screened, and the oscillation frequency and damping ratio are calculated to construct a feature library of key oscillation modes of the power grid.

[0007] Based on the power transmission path defined by the power grid key oscillation mode feature library and the transaction contract, the participation factor of the power transmission path on each electromechanical oscillation mode in the power grid key oscillation mode feature library is calculated, and a transaction path mode coupling matrix is ​​generated. The transaction path mode coupling matrix is ​​combined with the planned power curve ramp rate of the transaction contract and the predicted value of the new energy output fluctuation in the corresponding period to quantitatively calculate the transaction-frequency regulation conflict index.

[0008] Based on the key oscillation mode feature library of the power grid and the location node information of available frequency regulation resources in the power grid, the controllability index of each frequency regulation resource injection power on each electromechanical oscillation mode in the library is calculated to obtain the individual resource mode suppression efficiency value. The individual resource mode suppression efficiency values ​​of all available frequency regulation resources are summarized to form a view of the mode suppression capability of the entire network frequency regulation resources.

[0009] Based on the comparison between the transaction-frequency regulation conflict index and the preset safety threshold, power transmission paths exceeding the threshold are filtered out to form a set of transaction paths to be optimized. The network-wide frequency regulation resource mode suppression capability view is called to adjust the power transmission paths in the set of transaction paths to be optimized, and frequency regulation resources are preferentially allocated to the nodes with the strongest suppression capability to generate a power transaction and frequency regulation resource collaborative scheduling scheme.

[0010] Preferably, the steps for obtaining the key oscillation mode feature library of the power grid are as follows:

[0011] The system collects operating data of the power system, establishes a system state matrix, performs eigenvalue decomposition on the system state matrix, and extracts the eigenvalues ​​and corresponding eigenvectors of the system state matrix one by one to obtain the set of eigenvalues ​​and eigenvectors of the system state matrix.

[0012] Based on the set of eigenvalues ​​and eigenvectors of the system state matrix, the eigenvalues ​​and eigenvectors corresponding to the tie lines are extracted one by one. The damping values ​​of the tie line eigenvalues ​​are determined one by one. The tie line eigenvalues ​​and corresponding eigenvectors with damping values ​​less than a preset damping threshold are selected to obtain a set of low-damped electromechanical oscillation modes on the tie lines.

[0013] Based on the set of low-damped electromechanical oscillation modes on the interconnecting line, the oscillation frequency and damping ratio of each low-damped electromechanical oscillation mode are calculated, and the oscillation frequency and damping ratio of each mode are used to form a mode feature pair to obtain a key oscillation mode feature library of the power grid.

[0014] Preferably, the steps for obtaining the transaction path pattern coupling matrix are as follows:

[0015] Based on the power grid key oscillation mode feature library and transaction contracts, the node information covered in each power transmission path is extracted, and the time period positioning value of the path plan power curve is called at each node to establish the correspondence between power transmission paths and nodes, and generate a power transmission path and power grid node mapping table.

[0016] Based on the power transmission path and grid node mapping table, and combined with the feature vector of each oscillation mode in the key oscillation mode feature library of the power grid, the component values ​​corresponding to each node in the feature vector are extracted in sequence, multiplied by the power proportion coefficient of the node, and then normalized. The response contribution of each power transmission path to each oscillation mode is accumulated to obtain the trading path mode coupling matrix.

[0017] Preferably, the step of obtaining the transaction-frequency modulation conflict index is as follows:

[0018] Based on the transaction path mode coupling matrix, the transaction-frequency modulation conflict index of each power transmission path is calculated.

[0019] Preferably, the step of obtaining the single-resource mode suppression efficiency value is as follows:

[0020] Based on the key oscillation mode feature library of the power grid and the location node information of available frequency regulation resources, the node number of each frequency regulation resource is extracted and associated with the oscillation mode feature vector components corresponding to the nodes in the power grid topology. At the same time, the maximum active power regulation power and typical response time of each frequency regulation resource are read from the frequency regulation resource database to obtain the dynamic response capability set of frequency regulation resource nodes.

[0021] Based on the set of dynamic response capabilities of the frequency modulation resource nodes, calculate the individual resource mode suppression efficiency value of each frequency modulation resource for each key oscillation mode.

[0022] Preferably, the steps for obtaining the whole network frequency modulation resource mode suppression capability view are as follows:

[0023] Based on the individual resource mode suppression performance values ​​of all frequency modulation resources, a two-dimensional matrix is ​​constructed with frequency modulation resources as rows and electromechanical oscillation modes as columns. The individual resource mode suppression performance values ​​of each resource in each mode are filled into the matrix, and the matrix is ​​extracted as a view of the mode suppression capability of the entire network frequency modulation resources.

[0024] Preferably, the step of obtaining the set of transaction paths to be optimized is as follows:

[0025] Based on the transaction-frequency conflict index, the transaction-frequency conflict index value corresponding to each power transmission path is extracted one by one and compared with the preset safety threshold. If the transaction-frequency conflict index value is greater than the preset safety threshold, the corresponding power transmission path is included in the set of transaction paths to be optimized.

[0026] Preferably, the steps for obtaining the power trading and frequency regulation resource coordinated scheduling scheme are as follows:

[0027] Based on the set of transaction paths to be optimized, the network-wide frequency modulation resource mode suppression capability view is invoked to extract the key oscillation modes associated with each power transmission path in the set of transaction paths to be optimized. Based on the key oscillation modes, the individual resource mode suppression efficiency values ​​of all frequency modulation resource nodes under the corresponding modes are retrieved one by one from the network-wide frequency modulation resource mode suppression capability view. The node with the largest individual resource mode suppression efficiency value is selected as the target frequency modulation node to form a target frequency modulation node list.

[0028] Based on the target frequency regulation node list, the frequency regulation resource configuration scheme of the power transmission path in the set of trading paths to be optimized is adjusted one by one, the active power injection position of the frequency regulation resource is adjusted to the target frequency regulation node of the corresponding path, the set of trading paths to be optimized is optimized, and a power trading and frequency regulation resource collaborative scheduling scheme is generated.

[0029] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0030] This invention extracts and analyzes the eigenvalues ​​and eigenvectors of the system state matrix to screen for insufficiently damped electromechanical oscillation modes on tie lines, identifying key areas and modes of grid oscillation risk and effectively reducing the difficulty of identifying oscillation risk. Utilizing the coupling relationship between the power transmission path defined in the trading contract and the oscillation mode, it constructs a quantitative indicator reflecting the degree of risk in trading execution, revealing the frequency regulation risk that the trading plan may generate under renewable energy fluctuation scenarios. This allows for the early quantification of conflicts between trading execution and frequency regulation demand, thereby avoiding stability risks caused by power fluctuations. For the node location and dynamic response characteristics of frequency regulation resources, it evaluates the mode suppression effectiveness of each frequency regulation resource, forming a global resource allocation capability analysis and providing precise basis for the rational deployment of frequency regulation resources. In the coordinated scheduling of frequency regulation resources and trading risks, it accurately screens the power transmission paths that need optimization based on the risk level indicator and prioritizes resource allocation to key nodes with significant suppression effects according to the principle of optimal efficiency ranking across the entire network. This promotes the matching of power trading execution with grid frequency regulation resources, achieving a dual improvement in grid operation stability and trading economic benefits. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the steps of the present invention. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0033] Please see Figure 1 This invention provides a technical solution, a method for coordinated control of medium- and long-term power trading and grid frequency regulation demand, comprising the following steps:

[0034] The eigenvalues ​​and eigenvectors of the system state matrix are extracted, low-damped electromechanical oscillation modes on the tie line are screened, and the oscillation frequency and damping ratio are calculated to construct a feature library of key oscillation modes of the power grid.

[0035] Based on the power transmission path defined by the power grid key oscillation mode feature library and the transaction contract, the participation factor of the power transmission path on each electromechanical oscillation mode in the power grid key oscillation mode feature library is calculated, and the transaction path mode coupling matrix is ​​generated. The transaction path mode coupling matrix is ​​combined with the planned power curve ramp rate of the transaction contract and the predicted value of the new energy output fluctuation in the corresponding period to quantitatively calculate the transaction-frequency regulation conflict index.

[0036] Based on the feature library of key oscillation modes of the power grid and the location node information of available frequency regulation resources in the power grid, the controllability index of the injected power of each frequency regulation resource on each electromechanical oscillation mode in the library is calculated to obtain the mode suppression efficiency value of individual resources. The mode suppression efficiency values ​​of all available frequency regulation resources are summarized to form a view of the mode suppression capability of frequency regulation resources in the entire network.

[0037] Based on the comparison between the trading-frequency regulation conflict index and the preset safety threshold, power transmission paths exceeding the threshold are screened to form a set of trading paths to be optimized. The network-wide frequency regulation resource mode suppression capability view is called to adjust the power transmission paths in the set of trading paths to be optimized, and frequency regulation resources are preferentially allocated to the nodes with the strongest suppression capability to generate a power trading and frequency regulation resource collaborative scheduling scheme.

[0038] The steps for obtaining the feature library of key oscillation modes in power grids are as follows:

[0039] The system collects operating data of the power system, establishes a system state matrix, performs eigenvalue decomposition on the system state matrix, and extracts the eigenvalues ​​and corresponding eigenvectors of the system state matrix one by one to obtain the set of eigenvalues ​​and eigenvectors of the system state matrix.

[0040] Based on the set of eigenvalues ​​and eigenvectors of the system state matrix, the eigenvalues ​​and eigenvectors corresponding to the tie lines are extracted one by one. The damping values ​​of the tie line eigenvalues ​​are determined one by one. The tie line eigenvalues ​​and corresponding eigenvectors with damping values ​​less than the preset damping threshold are selected to obtain a set of low-damped electromechanical oscillation modes on the tie lines.

[0041] Based on the set of low-damped electromechanical oscillation modes on the interconnection line, the oscillation frequency and damping ratio of each low-damped electromechanical oscillation mode are calculated. The oscillation frequency and damping ratio of each mode are then used to form a mode feature pair, thus obtaining a feature library of key oscillation modes of the power grid.

[0042] Specifically, power system operation data is collected through phasor measurement units distributed throughout the power grid at various generator nodes and key substations. These units collect and upload rotor angles, rotor angular velocities, and voltage and current phasor data for each generator at a sampling frequency of 50 times per second. Subsequently, based on the system equilibrium point determined by the Newton-Raphson power flow calculation, and combined with the classical third-order generator model and network equations, a linearized state-space model describing the dynamic behavior of the power system is established. , Represents the state vector The derivative with respect to time represents the rate of change of the state variable. It is the system's state vector. It is the input vector, representing the small perturbation experienced by the system. It is the system state matrix. It is the input matrix, where the state vector is... Includes rotor angle deviations of each generator and rotor angular velocity deviation System state matrix The elements are filled in based on the synchronous torque coefficient and damping torque coefficient under the current operating conditions of the system. These coefficients are calculated from the network admittance matrix and generator parameters, and are used to establish the system state matrix. Numerical computation is performed using the QR decomposition algorithm to achieve eigenvalue decomposition. This decomposition process transforms the matrix... It is represented as its eigenvalue diagonal matrix With the right eigenvector matrix The product form, i.e. Then, traverse the eigenvalue diagonal matrix one by one. Extract the feature value of each complex number from the diagonal elements. , Indicates the first A number of complex eigenvalues, It is the real part of the eigenvalue, which determines the damping (decay rate) of this oscillation mode. It is the imaginary part of the eigenvalue, representing the angular frequency of this oscillation mode. It is the imaginary unit, satisfying Meanwhile, from the right eigenvector matrix Extract the corresponding column vector as the right eigenvector corresponding to that eigenvalue. All extracted eigenvalues ​​and their corresponding eigenvectors are organized and stored in the form of data pairs to obtain the set of eigenvalues ​​and eigenvectors of the system state matrix.

[0043] Based on the set of eigenvalues ​​and eigenvectors of the system state matrix, the electromechanical oscillation modes closely related to the tie lines are first identified by calculating the participation factor. Specifically, for each oscillation mode in the set... Calculate all generator state variables, especially the rotor angular velocity. Participation factors in this model The participating factors are mediated through the right eigenvector components. With left eigenvector components The product of the two is obtained, where the left eigenvector matrix is ​​the inverse of the right eigenvector matrix. When the participation factor of an oscillation mode exhibits large values ​​in generator sets in different regions, for example, the sum of the participation factors of the generator set in region A is greater than 0.4, and the sum of the participation factors of the generator set in region B connected by the tie line is also greater than 0.4, then the mode is identified as an inter-regional oscillation mode on the tie line, and its corresponding eigenvalues ​​and eigenvectors are extracted. Next, for these selected tie line oscillation modes, their damping levels are determined one by one, and the damping ratio of each mode is calculated. The calculation process is based on its eigenvalues. , Indicates the first in the power system The complex eigenvalues ​​corresponding to each electromechanical oscillation mode It is an eigenvalue The real part of the oscillation represents the damping of this oscillation mode, determining whether the oscillation is decaying, continuous, or divergent. It is an eigenvalue The imaginary part of represents the angular frequency of this oscillation mode. It is an imaginary unit, subscript It is an index used to distinguish multiple different oscillation modes existing in the system, for example Representing the first pattern, This represents the second pattern, and so on, through the formula. To complete, among which It is the first Damping ratio for each mode, It is the feature value corresponding to this pattern. The real part, It is the feature value corresponding to this pattern. The imaginary part of the damping ratio is then calculated, and the calculated damping ratio is compared with a dynamically set preset damping threshold. This threshold is not fixed but is determined by analyzing the damping ratio data of all identified tie-line oscillation modes in the power grid over the past month. Specifically, the setting method is as follows: first, calculate the arithmetic mean and standard deviation of all damping ratio sample data within this month; then, set the preset damping threshold to the average minus 1.5 times the standard deviation. For example, if the average damping ratio within a month is 0.06 and the standard deviation is 0.015, then the preset damping threshold is [value missing]. All tie-line oscillation modes with a damping ratio less than 0.0375 are identified as low-damped modes, and their eigenvalues ​​and corresponding eigenvectors are selected. The eigenvalues ​​and eigenvectors of all selected low-damped modes are then summarized to obtain a set of low-damped electromechanical oscillation modes on the tie line.

[0044] Based on the set of low-damped electromechanical oscillation modes on the interconnecting line, each low-damped electromechanical oscillation mode in the set is traversed, and the complex eigenvalues ​​corresponding to each mode are analyzed. Perform detailed parameter calculations, specifically, extract the imaginary part of the feature values. Through formula The oscillation frequency of this mode was calculated. It is the first The oscillation frequency of each mode, measured in Hz, characterizes how quickly the system's power oscillates in that mode. For example, if... If the value is 3.14 rad / s, then the oscillation frequency is 0.5 Hz. Simultaneously, using the real part of the eigenvalue... and the virtual part Through formula Calculate the damping ratio of this mode. This damping ratio directly reflects the rate of oscillation decay and is a dimensionless parameter. For example, if the eigenvalue of a mode is... Its damping ratio is 0.05, or 5%. Then, the calculated oscillation frequency and damping ratio, these two key parameters, are combined into a mode characteristic pair, for example... This feature fully describes the core dynamic characteristics of a low-damped oscillation mode. Finally, a structured record containing a mode identifier, associated tie-line information, original eigenvalues, corresponding eigenvectors, calculated oscillation frequency, and damping ratio is created for each low-damped mode in the set. All these records are then aggregated to form a systematic database that can be queried and accessed at any time, resulting in a feature library of key oscillation modes in the power grid.

[0045] The steps to obtain the transaction path pattern coupling matrix are as follows:

[0046] Based on the feature library of key oscillation modes of the power grid and the transaction contract, the node information covered in each power transmission path is extracted, and the time period positioning value of the planned power curve of the path is called at each node to establish the correspondence between the power transmission path and the node, and generate a mapping table between the power transmission path and the power grid node.

[0047] Based on the power transmission path and grid node mapping table, and combined with the feature vector of each oscillation mode in the key oscillation mode feature library of the power grid, the component values ​​corresponding to each node in the feature vector are extracted in sequence, multiplied by the power proportion coefficient of the node, and then normalized. The response contribution of each power transmission path to each oscillation mode is accumulated to obtain the trading path mode coupling matrix.

[0048] Specifically, based on the key oscillation mode feature library of the power grid and transaction contracts, the system first automatically parses the archived digitized medium- and long-term transaction contracts, reading the structured data fields in each contract, including the trading parties, the transaction volume, the execution period, and the power transmission path explicitly defined as a series of transmission lines and substation nodes. For example, for a transaction contract from a wind farm cluster in Province A to a load center in Province B, its power transmission path is defined as "wind farm booster station - line L1 - substation S1 - line L2 - substation S2 - load center step-down station in Province B". Based on this, the system extracts the sequence of all power grid nodes covered by the path, i.e., {booster station, S1, S2, step-down station}. Subsequently, the system accesses the path planned power curve associated with the contract. This curve has a time resolution of 15 minutes and specifies in detail the power transmission path throughout the entire contract execution period. When analyzing the planned power transmission value at each time point, the system will, for a specific future scheduling period, such as the next 24 hours, retrieve the planned power values ​​at 15-minute intervals from the planned power curve as the time period positioning values. For example, the positioning value is 500MW for the period from 14:00 to 14:15. Then, the system will establish a detailed mapping record for each power transmission path. This record will associate the unique identifier of the path with the list of nodes it covers, and mark the role of each node in the list in power transmission, such as injection node, receiving end node, or intermediate hub node. At the same time, it will attach the time period positioning values ​​for all time points within the corresponding scheduling period. Finally, the mapping records of all paths will be summarized to establish the correspondence between power transmission paths and nodes, and generate a power transmission path and grid node mapping table.

[0049] Based on the power transmission path and grid node mapping table, the system performs coupling analysis on each power transmission path and each oscillation mode in the grid key oscillation mode feature library. Specifically, the system iterates through each power transmission path in the mapping table and, for that path, it iterates again through all low-damping electromechanical oscillation modes in the grid key oscillation mode feature library. For the current combination of power transmission path and oscillation mode, the system first obtains a list of all nodes included in the path from the mapping table, and then retrieves the feature vector corresponding to the current oscillation mode from the feature library. This feature vector is a long vector containing the state variable components of all generators in the grid. The system accurately extracts the rotor angular velocity component values ​​related to each generator node on the path. Subsequently, the system calculates the power proportion coefficient for each generator node on the path. This coefficient is defined as the proportion of the power injected or absorbed by the node to the total traded power of the path. For example, in a path with a total traded power of 800MW, if node G1 injects 500MW and node G2 injects 300MW, then the power proportion coefficient of G1 is... G2 is The coefficients of the receiving nodes are negative. The system multiplies the extracted feature vector components by the power proportion coefficient of the corresponding node and algebraically sums the calculation results of all nodes on the path to obtain a preliminary response contribution value. To eliminate the influence of differences in dimensions and amplitudes between different oscillation modes, the system normalizes the response contribution values ​​of all power transmission paths for the same oscillation mode. The maximum value normalization method is used, dividing the contribution value of each path by the maximum absolute value of the contribution values ​​of all paths to that mode, so that the normalized value falls within the range of... Finally, within the interval, the normalized response contributions of all power transmission paths to all key oscillation modes are accumulated and summarized to obtain the trading path mode coupling matrix.

[0050] The steps to obtain the transaction-frequency conflict index are as follows:

[0051] Based on the trading path pattern coupling matrix, the trading-FM conflict index for each power transmission path is calculated using the following formula:

[0052] ;

[0053] in, For the first The transaction-frequency interference index of power transmission paths The first in the transaction path pattern coupling matrix The power transmission path for the first Participation factors of individual electromechanical oscillation modes For the first The planned power curve ramp rate of each power transmission path during the execution period. For the first The predicted amplitude of renewable energy output fluctuations for each power transmission path during the execution period. To establish a uniformly defined benchmark fluctuation range value, For the first Damping ratio of each electromechanical oscillation mode For the first The angular frequency of each electromechanical oscillation mode This represents the total number of key oscillation modes.

[0054] Specifically, the formula: The advantage of the formula is that it reflects market behavior (as the planned ramp-up rate of the transaction path). The uncertainty of the physical system (manifested as fluctuations in new energy output) and the uncertainty of the physical system. The inherent dynamic vulnerability of the power grid (manifested as the damping ratio of specific oscillation modes) and the inherent dynamic vulnerability of the power grid. and frequency This is tightly coupled by introducing participation factors for oscillation patterns through trading paths. The formula can identify which trading paths have the greatest impact on which weak oscillation patterns, achieving targeted risk assessment. Furthermore, it uses an exponential function. To address the impact of new energy fluctuations, it can nonlinearly amplify the risks brought about by high volatility, which is highly consistent with the actual situation of power grid safe operation, that is, the impact of large power fluctuations on system stability is far greater than the linear superposition of small fluctuations.

[0055] The first in the transaction path pattern coupling matrix The power transmission path for the first The participation factor for each electromechanical oscillation mode, obtained directly from the trading path mode coupling matrix generated in the previous step, quantifies the extent to which a specific power trading flow will excite or suppress a particular electromechanical oscillation mode in the power grid. Its value falls between [value missing]. Between these values, a larger absolute value indicates a stronger coupling, and the positive or negative sign indicates the phase relationship. During the calculation, the system will number the power transmission path to be evaluated. and the oscillation mode number being analyzed Locating the first in the transaction path pattern coupling matrix line, number The element values ​​of the column are read. For example, for the power transmission path numbered T005 and the inter-regional oscillation mode numbered Mode2, the corresponding participation factor value is found to be 0.68 from the matrix. This indicates that the power fluctuation of the trading path is highly positively correlated with the oscillation pattern of the oscillation mode, which will significantly stimulate the mode.

[0056] For the first The ramp rate of the planned power curve for each power transmission path during the execution period reflects the rate of change of the planned power of the transaction. It is calculated from the planned power curve of the path associated with the transaction contract. The system locks the upcoming scheduling period, such as the next 15 minutes, reads the planned power values ​​at the start and end of that period, and calculates the ramp rate by dividing the difference between the two by the time span. The calculation formula is as follows: ,in and The planned power at the end and beginning of the time period, respectively. For example, if a trading path has a planned power of 400MW at 15:00 and a planned power of 450MW at 15:15, then the ramp rate for that time period is... , to be consistent with the angular frequency in the formula The units (rad / s) should remain consistent; they will be converted to the same unit here. ;

[0057] For the first The parameter represents the predicted fluctuation range of renewable energy output for each power transmission path during the execution period. This parameter quantifies the uncertainty of renewable energy power generation (such as wind power and photovoltaic) output related to the trading path. The data comes from the grid's renewable energy power prediction system, which provides each renewable energy power station with predicted output values ​​for each future period and a confidence interval. Defined as half the width of the confidence interval, this represents the maximum possible deviation of the predicted power from the mean. For example, if a trading path is primarily powered by a wind farm, and the prediction system gives a 95% confidence interval of 180MW to 240MW for the period 15:00-15:15, then the width of this interval is 60MW, corresponding to the predicted power output fluctuation range. That is, 30MW;

[0058] The benchmark fluctuation amplitude value is a standardized parameter uniform across the entire grid, set by the power grid operation department based on long-term historical data statistical analysis. Its purpose is to provide a benchmark for measuring the severity of power output fluctuations from renewable energy sources. Specifically, the method involves collecting power output fluctuation amplitude data from all renewable energy power plants across the entire grid on a 15-minute scale over the past year, forming a dataset, calculating the probability distribution of this dataset, and using the 95th percentile of the distribution as the benchmark. For example, based on statistical analysis, if 95% of the new energy power output fluctuations across the entire network are below 25MW, then set... This value represents the level of fluctuation that the power grid can generally tolerate under normal operating conditions;

[0059] For the first The damping ratio of each electromechanical oscillation mode is one of the core indicators describing the dynamic characteristics of the power grid. It is directly extracted from the key oscillation mode feature library generated in the first step of this application. For each key oscillation mode in the feature library, its corresponding damping ratio has been pre-calculated and stored through small-disturbance stability analysis. The damping ratio is a dimensionless value that reflects the decay rate of the oscillation. The smaller the value, the less easily the oscillation decays, and the worse the system stability. Generally, a damping ratio less than 0.05 is considered weakly damped. For example, the damping ratio of the oscillation mode with the number Mode2 found in the feature library is [insert value here]. ;

[0060] For the first The angular frequency of each electromechanical oscillation mode is also directly extracted from the key oscillation mode feature library of the power grid. It describes the speed of the oscillation, measured in rad / s, and together with the damping ratio, constitutes complete information describing the dynamic characteristics of the oscillation mode. In the feature library, the angular frequency of each mode is also pre-calculated and stored. For example, querying the feature library yields the oscillation mode numbered Mode2, whose angular frequency is... This corresponds to an oscillation frequency of Low-frequency oscillations;

[0061] The total number of key oscillation modes represents the total number of the most unstable oscillation modes in the power grid considered when calculating the conflict degree. This value is determined by the size of the key oscillation mode feature library of the power grid, that is, the total number of low-damping mode entries contained in the feature library.

[0062] Calculation process:

[0063] To calculate the first Trading-Frequency Collision Index for Power Transmission Paths For example, and set the total number of key oscillation modes. ,

[0064] The values ​​obtained from the steps of obtaining each parameter are as follows:

[0065] ;

[0066] ;

[0067] ;

[0068] And obtain the parameters corresponding to each mode from the trading path mode coupling matrix and the feature library of key power grid oscillation modes:

[0069] For the pattern : , , ;

[0070] For the pattern : , , ;

[0071] For the pattern : , , ;

[0072] First, the calculation formula does not follow the pattern. The changing molecular part:

[0073]

[0074] Next, the conflict contribution of this path to each mode is calculated separately:

[0075] For the pattern :

[0076]

[0077] For the pattern :

[0078]

[0079] For the pattern :

[0080]

[0081] Finally, the conflict contributions of each mode are summed to obtain the final transaction-frequency modulation conflict index:

[0082] ;

[0083] This result indicates that the first The trading-frequency conflict index for this power transmission path is 1.3556 MW. This is a risk indicator in terms of power, and this value will be used to compare with the system's set safety threshold. For example, if the safety threshold is set to 60 MW, then a result of 1.3556 means that the trading path has a low safety risk under the current predicted operating conditions. Although its absolute risk value is not high, the sub-item calculation results still show that the potential risk of this path mainly comes from its strong coupling with mode 2 (contributing 1.1873), because mode 2 itself has extremely weak damping. And this path has a high degree of involvement in it. ).

[0084] The steps to obtain the suppression efficiency value of the single resource mode are as follows:

[0085] Based on the key oscillation mode feature library of the power grid and the location node information of available frequency regulation resources, the node number of each frequency regulation resource is extracted and associated with the oscillation mode feature vector components corresponding to the nodes in the power grid topology. At the same time, the maximum active power regulation power and typical response time of each frequency regulation resource are read from the frequency regulation resource database to obtain the dynamic response capability set of frequency regulation resource nodes.

[0086] Based on the dynamic response capability set of frequency modulation resource nodes, the individual resource mode suppression effectiveness value of each frequency modulation resource for each key oscillation mode is calculated. The calculation formula is as follows:

[0087] ;

[0088] in, For the first The frequency modulation resource for the first Individual resource mode suppression efficiency value of each electromechanical oscillation mode For the first The eigenvector corresponding to the eigenmode of the electromechanical oscillation mode The component values ​​of each frequency modulation resource node. For the first The maximum active power regulation of each frequency regulation resource For the first Damping ratio of each electromechanical oscillation mode For the first The angular frequency of each electromechanical oscillation mode For the first Typical response time of a frequency modulation resource.

[0089] Specifically, based on the key oscillation mode feature library of the power grid and the location node information of available frequency regulation resources, the system first accesses the list of available frequency regulation resources across the entire network. This list is updated in real time and includes all units in the power grid with frequency regulation capabilities, such as hydro-thermal power units equipped with automatic generation control functions, pumped storage power stations, and grid-connected electrochemical energy storage systems. It also records the unique identifier of each resource and its electrical node number at the point of grid connection. Subsequently, for each frequency regulation resource in the list, the system extracts its node number and uses it as an index to search within the pre-established power grid topology model. Centered on this node, it correlates the corresponding component values ​​within the feature vector of each key oscillation mode in the key oscillation mode feature library. Specifically, for frequency regulation resources of generator type, the system extracts the component of its rotor angular velocity in the feature vector; while for resources such as energy storage connected to the grid through power electronic interfaces, the system extracts the component of the rotor angular velocity in the feature vector. The system extracts the components of the injected active power or grid connection point voltage phase angle in the eigenvector. The magnitude of this component directly reflects the observability and controllability of the node for a specific oscillation mode. At the same time, the system also accesses the frequency regulation resource database in parallel. This is a static library that records the technical parameters of various frequency regulation resources in detail. Based on the unique identifier of the frequency regulation resource being processed, two key performance indicators are accurately read from it: maximum active power regulation, which is the maximum upward or downward regulation power that the resource can continuously provide under technical constraints, and typical response time, which is the time required for the resource to reach 90% of the command value from receiving the scheduling command. Finally, the node number of each frequency regulation resource, all associated mode eigenvector components, maximum active power regulation, and typical response time are integrated into a structured data record. The records of all resources are then aggregated to obtain the dynamic response capability set of the frequency regulation resource nodes.

[0090] formula: The advantage of the formula lies in the introduction of eigenvector components. The formula can quantify the control of the electrical location of frequency regulation resources on the target oscillation mode, solving the location problem of "using the best resources where they are most needed," and the maximum active power regulation. It directly reflects the "power" of resources, while the denominator contains... The time delay effect in the control system was simulated, showing that for higher frequency oscillations, the suppression effect is more severely discounted due to the resource response time lag. Finally, the damping ratio of the mode itself was... By placing it in the denominator, the evaluation value of the inhibitory effectiveness is amplified for those modes that are inherently extremely fragile and have very little damping, thus giving them higher priority when prioritizing resources.

[0091] For the first The eigenvector corresponding to the eigenmode of the electromechanical oscillation mode The component value of the frequency regulation resource node, which is obtained directly from the characteristic library of key oscillation modes of the power grid, characterizes the frequency regulation resource node's frequency regulation component value. In the oscillation mode, the first The oscillation amplitude and phase of a certain state variable (such as generator speed or bus voltage phase angle) at the node where a frequency regulation resource is located are key physical quantities for measuring the controllability of the mode at that location. The larger the absolute value, the more significant the effect of injecting or absorbing power at that location on suppressing the mode oscillation. During calculation, the system uses the currently evaluated frequency regulation resource number. Its grid-connected nodes and target oscillation mode number The first one stored in the feature library Extracting the resource from the feature vectors of each pattern. The complex component corresponding to the grid-connected node is determined, and its modulus is taken. For example, for the energy storage power station located at node 28 (resource number B05) and the No. 2 key oscillation mode in the system (mode number k=2), the component corresponding to node 28 in the feature vector of this mode is found from the feature library. ,but The calculated value is This value will be substituted into the formula for calculation;

[0092] For the first The maximum active power regulation capacity of a frequency regulation resource, measured in megawatts (MW), is derived from the frequency regulation resource database. This data is a parameter determined by the power system dispatching agency when verifying the capacity of grid-connected frequency regulation resources. It is typically recorded in the resource's grid connection agreement or technical documentation. It defines the maximum frequency regulation power capacity that the resource can stably provide. For conventional hydroelectric and thermal power units, it is usually half of their AGC (Automatic Generation Control) regulation range; for battery energy storage systems, it is their rated charging and discharging power. This parameter is obtained by querying the frequency regulation resource database and matching the resource number. This is achieved through corresponding records. For example, for a battery energy storage power station numbered B05, by consulting its technical documents, it is confirmed that its rated power is 20MW and it has symmetrical charging and discharging capabilities. Therefore, its maximum active power regulation capacity is... That is, 20 MW;

[0093] For the first The damping ratio of each electromechanical oscillation mode is a dimensionless value, directly retrieved from the key oscillation mode feature library of the power grid. This value was calculated during the previous small-disturbance stability analysis of the entire network and is a core parameter describing the inherent dynamic characteristics of the power grid. A smaller value indicates a slower decay of the oscillation mode and a greater threat to system security. This parameter was already used in the previous step of calculating the "trading-frequency regulation conflict index," so the same data is directly used here to ensure consistency. For example, for the most threatening key oscillation mode, number 2, its damping ratio is retrieved from the feature library as follows: ;

[0094] For the first The angular frequency of each electromechanical oscillation mode, measured in rad / s, is also directly extracted from the key oscillation mode feature library of the power grid. Together with the damping ratio, it defines the dynamic characteristics of an oscillation mode, representing the natural frequency of the oscillation. This value is consistent with the value used in the previous step's calculation of the "trading-frequency regulation conflict index." For example, for key oscillation mode 2, its angular frequency read from the feature library is... This frequency falls within the typical inter-regional low-frequency oscillation range of the power grid;

[0095] For the first The typical response time of a frequency regulation resource, measured in seconds (s), is obtained from the frequency regulation resource database. It quantifies the response speed of the resource and is defined as the time it takes for the resource to receive a power regulation command and for the actual output power to reach 90% of the command value. This value is calibrated through on-site dynamic performance testing of the resource. Different types of resources exhibit vastly different response times. Battery energy storage systems, due to the rapid control capabilities of their power electronic interfaces, typically have sub-second response times. In contrast, conventional hydroelectric and thermal power units, due to mechanical inertia and combustion process delays, have response times ranging from tens of seconds to several minutes. For example, for battery energy storage power station numbered B05, its typical response time, as determined by its grid connection test report, is calibrated as follows: ;

[0096] Calculation process:

[0097] Calculate the energy storage power station (resource) with the calculation number B05. ) for key oscillation pattern No. 2 (pattern) The single-resource mode suppression efficiency value For example,

[0098] The values ​​obtained from the steps of obtaining each parameter are as follows:

[0099] ;

[0100] ;

[0101] ;

[0102] ;

[0103] ;

[0104] First, the numerator of the calculation formula:

[0105] ;

[0106] Next, calculate the denominator of the formula:

[0107]

[0108] Finally, dividing the numerator by the denominator yields the final single-resource mode suppression efficiency value:

[0109] ;

[0110] The results show that the energy storage power station numbered B05 has a single-resource mode suppression efficiency of 54.78 for the No. 2 key oscillation mode. The value itself represents the relative magnitude of the equivalent damping contribution that the resource can provide to the target mode under ideal conditions. The higher this value, the more effective the resource is in suppressing the oscillation of this specific mode. By calculating this efficiency value for all available frequency regulation resources in the entire network for this mode, their advantages and disadvantages can be directly ranked.

[0111] The steps to obtain the network-wide frequency modulation resource mode suppression capability view are as follows:

[0112] Based on the individual resource mode suppression performance values ​​of all frequency modulation resources, a two-dimensional matrix is ​​constructed with frequency modulation resources as rows and electromechanical oscillation modes as columns. The individual resource mode suppression performance values ​​of each resource in each mode are filled into the matrix, and the matrix is ​​extracted as a view of the mode suppression capability of the entire network frequency modulation resources.

[0113] Specifically, based on the individual resource mode suppression effectiveness values ​​of all frequency regulation resources, the system then executes a matrix construction process. This process first determines the dimensions of the matrix. The number of rows in the matrix equals the total number of available frequency regulation resources in the entire network, and the row index is the unique identifier for each resource, such as "BESS-01" or "Hydro-G5". The number of columns in the matrix equals the total number of key oscillation modes in the power grid key oscillation mode feature library, and the column index is the unique identifier for each mode, such as "Mode1", "Mode2", or "Mode3". Subsequently, the system initializes an empty matrix with the above dimensions and begins to fill it with data. This is done by traversing the set of all individual resource mode suppression effectiveness values ​​calculated in the previous step, and for each effectiveness value... The system parses out its corresponding resource number. and pattern number Then locate the first in the matrix. line and number Intersecting cells in the column and setting performance values This process continues until all calculated performance values ​​are accurately filled into the corresponding positions in the matrix. The resulting two-dimensional matrix has each element precisely quantifying the suppression capability of a specific frequency modulation resource for a specific oscillation mode. This matrix is ​​extracted as a complete data structure and named the "Network-wide Frequency Modulation Resource Mode Suppression Capability View".

[0114] The steps to obtain the set of transaction paths to be optimized are as follows:

[0115] Based on the transaction-frequency conflict index, the transaction-frequency conflict index value corresponding to each power transmission path is extracted one by one and compared with the preset safety threshold. If the transaction-frequency conflict index value is greater than the preset safety threshold, the corresponding power transmission path is included in the set of transaction paths to be optimized.

[0116] Specifically, based on the trading-frequency regulation conflict index, the system automatically traverses the list of all planned power transmission paths. For each path in the list, it extracts the trading-frequency regulation conflict index value calculated in the previous steps and compares this value with a dynamically set preset safety threshold. This preset safety threshold is not a fixed empirical value, but is dynamically generated based on historical data statistics and the current power grid operating status. Its setting process consists of two steps: the first step is to establish a baseline threshold. By collecting the trading-frequency regulation conflict index for all trading periods over the past three months, a historical risk database is formed. The arithmetic mean and standard deviation of all index values ​​in this database are calculated, and the baseline threshold is set as the average plus twice the standard deviation. This statistic can cover approximately 97.5% of historical normal operating conditions. For example, if the historical average is 45 and the standard deviation is 8, then the baseline threshold is... The second step is to introduce an operational status adjustment coefficient. Based on the overall safety margin of the current power grid, the grid dispatcher divides the operational status into three levels: "Normal," "Watch," and "Alarm," corresponding to adjustment coefficients of 1.0, 0.8, and 0.6, respectively. The preset safety threshold is ultimately equal to the baseline threshold multiplied by the current operational status adjustment coefficient. If the current power grid is in the "Watch" state, the preset safety threshold is adjusted to... In the comparison process, if the transaction-frequency conflict index of a certain power transmission path is 81.312, it is determined that the value is greater than the currently effective preset security threshold of 48.8. The system then records the unique identifier of the power transmission path in a temporary list. After this process is completed for all power transmission paths, all path identifiers in the temporary list are summarized to form a set of transaction paths to be optimized.

[0117] The steps to obtain the power trading and frequency regulation resource coordinated dispatch scheme are as follows:

[0118] Based on the set of transaction paths to be optimized, the entire network frequency modulation resource mode suppression capability view is called, and the key oscillation modes associated with each power transmission path in the set of transaction paths to be optimized are extracted one by one. Based on the key oscillation modes, the individual resource mode suppression efficiency values ​​of all frequency modulation resource nodes under the corresponding modes are retrieved one by one from the entire network frequency modulation resource mode suppression capability view. The node with the largest individual resource mode suppression efficiency value is selected as the target frequency modulation node, and a target frequency modulation node list is formed.

[0119] Based on the target frequency regulation node list, the frequency regulation resource configuration scheme of the power transmission path in the set of trading paths to be optimized is adjusted one by one. The active power injection position of the frequency regulation resource is adjusted to the target frequency regulation node of the corresponding path, the set of trading paths to be optimized is optimized, and a power trading and frequency regulation resource collaborative scheduling scheme is generated.

[0120] Specifically, based on the set of transaction paths to be optimized, the system calls the network-wide frequency modulation resource mode suppression capability view and iteratively processes each high-risk power transmission path in the set. First, the system needs to identify the key oscillation mode that causes the risk to exceed the limit for each path. Specifically, the system backtracks the calculation process of the transaction-frequency modulation conflict index for that path, that is, summing the terms that constitute the total index. A step-by-step analysis was conducted to calculate each key oscillation mode. The system calculates the contribution value of each mode to the overall index and identifies the mode with the largest contribution value. For example, if mode 2 contributes 87% of the overall index, then mode 2 is identified as the key oscillation mode associated with that path. After determining the key oscillation mode, the system uses the identifier of this mode as a query index to search within the two-dimensional matrix of the network-wide frequency modulation resource mode suppression capability view. This search operation focuses on a specific column in the matrix representing this mode. This column contains the individual resource mode suppression effectiveness values ​​of all available frequency modulation resources in the network for this mode. The system then reads all values ​​in this column and... The system performs sorting or direct maximum value lookup operations. For example, in the column corresponding to Mode 2, the system finds that the energy storage power station numbered B05 has an efficiency value of 54.78, which is the highest among all resources. Therefore, the node connected to the B05 energy storage power station, i.e., node 28, is selected as the target frequency regulation node for the current optimization path. The system establishes a temporary key-value pair mapping between the identifier of this path and node 28. This process is repeated for all paths in the set of transaction paths to be optimized. Finally, all the established key-value pair mappings are summarized to form a target frequency regulation node list.

[0121] Based on the target frequency regulation node list, the system begins to adjust the existing frequency regulation resource allocation schemes for each power transmission path recorded in the set of transaction paths to be optimized. This adjustment process is based on the mapping relationship between the paths and the optimal frequency regulation nodes provided by the target frequency regulation node list. Specifically, the system first reads the original scheduling plan of a path to be optimized. This plan may contain a default frequency regulation resource source or one determined through a conventional market clearing mechanism, such as a conventional generator unit at the path's starting point. Subsequently, based on the target frequency regulation node list, the system finds the target frequency regulation node corresponding to the path, such as node 28, and the optimal frequency regulation resource on that node, namely the B05 energy storage power station. The system then generates a new scheduling instruction that explicitly changes the frequency regulation originally allocated to the conventional generator unit. The task was reassigned to the B05 energy storage power station, and the active power injection location was adjusted from the original node to node 28. This new instruction will cover the frequency regulation configuration part of the original plan. After this series of operations, all paths in the original set of trading paths to be optimized will have a precisely optimized frequency regulation resource configuration scheme. The system integrates these optimized schemes with the scheduling plans of other normal trading paths in the power grid that have not been adjusted, and finally constructs and outputs a complete and structured collaborative scheduling scheme. This scheme details the execution plan of all power transactions in the next scheduling cycle, and specifies the optimal frequency regulation resources and their active power injection locations for those trading paths with high stability risks, optimizes the set of trading paths to be optimized, and generates a collaborative scheduling scheme for power transactions and frequency regulation resources.

[0122] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for coordinated control of medium- and long-term power trading and grid frequency regulation demand, characterized in that, Includes the following steps: The eigenvalues ​​and eigenvectors of the system state matrix are extracted, low-damped electromechanical oscillation modes on the tie line are screened, and the oscillation frequency and damping ratio are calculated to construct a feature library of key oscillation modes of the power grid. Based on the power transmission path defined by the power grid key oscillation mode feature library and the transaction contract, the participation factor of the power transmission path on each electromechanical oscillation mode in the power grid key oscillation mode feature library is calculated, and a transaction path mode coupling matrix is ​​generated. The transaction path mode coupling matrix is ​​combined with the planned power curve ramp rate of the transaction contract and the predicted value of the new energy output fluctuation in the corresponding period to quantitatively calculate the transaction-frequency regulation conflict index. Based on the key oscillation mode feature library of the power grid and the location node information of available frequency regulation resources in the power grid, the controllability index of each frequency regulation resource injection power on each electromechanical oscillation mode in the library is calculated to obtain the individual resource mode suppression efficiency value. The individual resource mode suppression efficiency values ​​of all available frequency regulation resources are summarized to form a view of the mode suppression capability of the entire network frequency regulation resources. Based on the comparison between the transaction-frequency regulation conflict index and the preset safety threshold, power transmission paths exceeding the threshold are screened to form a set of transaction paths to be optimized. The network-wide frequency regulation resource mode suppression capability view is called to adjust the power transmission paths in the set of transaction paths to be optimized, and frequency regulation resources are preferentially allocated to the nodes with the strongest suppression capability to generate a power transaction and frequency regulation resource collaborative scheduling scheme. The steps for obtaining the suppression effectiveness value of the single resource mode are as follows: Based on the key oscillation mode feature library of the power grid and the location node information of available frequency regulation resources, the node number of each frequency regulation resource is extracted and associated with the oscillation mode feature vector components corresponding to the nodes in the power grid topology. At the same time, the maximum active power regulation power and typical response time of each frequency regulation resource are read from the frequency regulation resource database to obtain the dynamic response capability set of frequency regulation resource nodes. Based on the set of dynamic response capabilities of the frequency modulation resource nodes, calculate the individual resource mode suppression efficiency value of each frequency modulation resource for each key oscillation mode; The steps for obtaining the network-wide frequency modulation resource mode suppression capability view are as follows: Based on the individual resource mode suppression performance values ​​of all frequency modulation resources, a two-dimensional matrix is ​​constructed with frequency modulation resources as rows and electromechanical oscillation modes as columns. The individual resource mode suppression performance values ​​of each resource in each mode are filled into the matrix, and the matrix is ​​extracted as a view of the frequency modulation resource mode suppression capability of the entire network. The steps for obtaining the power trading and frequency regulation resource coordinated scheduling scheme are as follows: Based on the set of transaction paths to be optimized, the network-wide frequency modulation resource mode suppression capability view is invoked to extract the key oscillation modes associated with each power transmission path in the set of transaction paths to be optimized. Based on the key oscillation modes, the individual resource mode suppression efficiency values ​​of all frequency modulation resource nodes under the corresponding modes are retrieved one by one from the network-wide frequency modulation resource mode suppression capability view. The node with the largest individual resource mode suppression efficiency value is selected as the target frequency modulation node to form a target frequency modulation node list. Based on the target frequency regulation node list, the frequency regulation resource configuration scheme of the power transmission path in the set of trading paths to be optimized is adjusted one by one, the active power injection position of the frequency regulation resource is adjusted to the target frequency regulation node of the corresponding path, the set of trading paths to be optimized is optimized, and a power trading and frequency regulation resource collaborative scheduling scheme is generated.

2. The coordinated control method for medium- and long-term power trading and grid frequency regulation demand according to claim 1, characterized in that, The steps for obtaining the key oscillation mode feature library of the power grid are as follows: The system collects operating data of the power system, establishes a system state matrix, performs eigenvalue decomposition on the system state matrix, and extracts the eigenvalues ​​and corresponding eigenvectors of the system state matrix one by one to obtain the set of eigenvalues ​​and eigenvectors of the system state matrix. Based on the set of eigenvalues ​​and eigenvectors of the system state matrix, the eigenvalues ​​and eigenvectors corresponding to the tie lines are extracted one by one. The damping values ​​of the tie line eigenvalues ​​are determined one by one. The tie line eigenvalues ​​and corresponding eigenvectors with damping values ​​less than a preset damping threshold are selected to obtain a set of low-damped electromechanical oscillation modes on the tie lines. Based on the set of low-damped electromechanical oscillation modes on the interconnecting line, the oscillation frequency and damping ratio of each low-damped electromechanical oscillation mode are calculated, and the oscillation frequency and damping ratio of each mode are used to form a mode feature pair to obtain a key oscillation mode feature library of the power grid.

3. The coordinated control method for medium- and long-term power trading and grid frequency regulation demand according to claim 1, characterized in that, The steps for obtaining the transaction path pattern coupling matrix are as follows: Based on the power grid key oscillation mode feature library and transaction contracts, the node information covered in each power transmission path is extracted, and the time period positioning value of the path plan power curve is called at each node to establish the correspondence between power transmission paths and nodes, and generate a power transmission path and power grid node mapping table. Based on the power transmission path and grid node mapping table, and combined with the feature vector of each oscillation mode in the key oscillation mode feature library of the power grid, the component values ​​corresponding to each node in the feature vector are extracted in sequence, multiplied by the power proportion coefficient of the node, and then normalized. The response contribution of each power transmission path to each oscillation mode is accumulated to obtain the trading path mode coupling matrix.

4. The coordinated control method for medium- and long-term power trading and grid frequency regulation demand according to claim 1, characterized in that, The steps for obtaining the transaction-frequency modulation conflict index are as follows: Based on the transaction path mode coupling matrix, the transaction-frequency modulation conflict index of each power transmission path is calculated.

5. The method for coordinated control of medium- and long-term power trading and grid frequency regulation demand according to claim 1, characterized in that, The steps for obtaining the set of transaction paths to be optimized are as follows: Based on the transaction-frequency conflict index, the transaction-frequency conflict index value corresponding to each power transmission path is extracted one by one and compared with the preset safety threshold. If the transaction-frequency conflict index value is greater than the preset safety threshold, the corresponding power transmission path is included in the set of transaction paths to be optimized.

Images