Power distribution network bearing capacity evaluation system based on dynamic correction

Abstract

The invention relates to the technical field of power distribution network evaluation, and discloses a power distribution network bearing capacity evaluation system based on dynamic correction. The system comprises a dynamic data acquisition module, a multi-dimensional state space construction module, a security domain analysis module, a partition coupling degree calculation module and a bearing capacity evaluation engine module. The dynamic data acquisition module acquires a power injection quantity sequence, a voltage deviation ratio sequence and uncontrollable parameter fluctuation data of each partition node of the power distribution network; a multi-dimensional state space construction module performs dimension raising mapping on the sequence to generate a linearized power flow state space model containing a power-voltage Jacobian matrix; the security domain analysis module corrects the boundary of the model according to uncontrollable parameter fluctuation and generates a dynamic security operation constraint set; the partition coupling degree calculation module quantifies an electrical independence index by means of a spectrum radius; and the bearing capacity evaluation engine constructs a chance constraint optimization model, outputs the photovoltaic maximum accessible capacity of each partition and a safety guarantee supply control strategy set, and improves the evaluation accuracy and practicability.

Description

Technical Field

[0001] This invention relates to the field of distribution network assessment technology, specifically a distribution network carrying capacity assessment system based on dynamic correction. Background Technology

[0002] As the penetration rate of new energy sources such as distributed photovoltaics in the distribution network continues to increase, the operating status of the distribution network exhibits significant volatility and uncertainty, and traditional methods for assessing the carrying capacity of the distribution network are gradually revealing many limitations.

[0003] At the data acquisition level, traditional methods rely on static data acquisition at fixed intervals, which makes it difficult to capture the dynamic changes of key operating parameters such as power injection and voltage deviation rate of each node in the distribution network in real time. Especially in scenarios such as fluctuations in new energy output and sudden load changes, static data cannot reflect the real operating status of the distribution network in a timely manner, resulting in the loss of an accurate data basis for subsequent carrying capacity assessment.

[0004] Traditional methods for constructing state-space models often employ simplified linearized models that fail to fully consider the complex mapping relationship between node power injection and voltage deviation rate, resulting in a lack of comprehensive characterization of the power flow state of the distribution network. This simplified model ignores numerous influencing factors, leading to significant discrepancies between the constructed state-space model and the actual operation of the distribution network, thus failing to provide accurate model support for capacity assessment.

[0005] In terms of security domain analysis, traditional methods typically determine the security domain of a distribution network based on fixed operating parameters and static constraints, failing to effectively incorporate uncontrollable parameter fluctuations. However, in actual operation, uncontrollable factors such as fluctuations in renewable energy output due to changes in wind speed and solar intensity, as well as load fluctuations caused by changes in user electricity consumption behavior, constantly alter the safe operating boundaries of the distribution network. Static security domain analysis methods struggle to adapt to these dynamic changes, easily leading to conservative assessment results or the existence of potential safety hazards.

[0006] In the calculation of zonal coupling, traditional methods mostly ignore the electrical interconnections between different zones of the distribution network and fail to effectively quantify the electrical independence between zones. The different zones of the distribution network do not operate completely independently; the coupling relationships such as power exchange and voltage impact between zones directly affect the carrying capacity assessment results of each zone. The lack of accurate calculation of this coupling degree will lead to discrepancies between the carrying capacity assessment at the zone level and the actual situation, thus affecting the reasonable determination of photovoltaic grid connection capacity.

[0007] Regarding capacity assessment engines, traditional assessment models mostly employ deterministic optimization methods, failing to fully integrate dynamic safety operation constraints and electrical independence indicators. This makes it difficult to maximize the photovoltaic access capacity of each zone while ensuring the safe operation of the distribution network. Furthermore, traditional models cannot output targeted sets of safety and supply control strategies, making it difficult to guide the actual operation and regulation of the distribution network and failing to meet the refined and dynamic needs of distribution networks for capacity assessment under high renewable energy penetration rates. Summary of the Invention

[0008] The purpose of this invention is to provide a power distribution network carrying capacity assessment system based on dynamic correction, so as to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides a power distribution network carrying capacity assessment system based on dynamic correction, the system comprising: The dynamic data acquisition module is used to acquire the real-time operating parameter set of each zone of the distribution network. The real-time operating parameter set includes the node power injection sequence, voltage deviation rate sequence and uncontrollable parameter fluctuation data. A multidimensional state space construction module is used to perform a complete up-dimensional mapping process on the node power injection sequence and the voltage deviation rate sequence to generate a linearized power flow state space model of the distribution network. The linearized power flow state space model includes a power-voltage Jacobian matrix and a simplified mapping table of control variables. The safety domain analysis module is used to perform boundary correction processing on the linearized power flow state space model based on the uncontrollable parameter fluctuation data, and generate a dynamic safety operation constraint set, which includes partition power limit curves and node voltage safety thresholds. The partition coupling degree calculation module is used to calculate the electrical independence index between each partition based on the power-voltage Jacobian matrix. The electrical independence index is quantified by the spectral radius of the power sensitivity matrix between partitions. The carrying capacity assessment engine is used to integrate the dynamic safe operation constraint set and electrical independence index to construct an opportunity constraint optimization model at the zonal level, and output the maximum grid-connectable photovoltaic capacity and safe supply control strategy set for each zonal.

[0010] Preferably, the multidimensional state space construction module includes: The nonlinear power flow analysis unit is used to input the node power injection sequence into a preset nonlinear power flow equation to generate an initial voltage distribution field and a power loss gradient field. The dimension-upgrading mapping unit is used to perform Taylor series expansion on the initial voltage distribution field and extract second-order and above terms to form a fully dimension-upgrading state space basis. The control variable compression unit is used to dynamically update the simplified control variable mapping table according to the power loss gradient field. The simplified control variable mapping table records the dimensionality reduction association rules of active and reactive control variables. The Jacobian matrix generation unit is used to calculate the sparsification correction coefficients of the power-voltage Jacobian matrix based on the fully upgraded state space basis and the reduced-dimensional association rule.

[0011] Preferably, the security domain analysis module includes: The parameter disturbance simulation unit is used to generate multiple sets of random disturbance scenarios based on the uncontrollable parameter fluctuation data. Each set of disturbance scenarios includes the distributed photovoltaic power output fluctuation and load mutation increment. The boundary search unit is used to successively load the random disturbance scenario into the linearized power flow state space model and search for the feasible boundary points of the power injection space using the golden section method. The threshold fitting unit is used to perform least-squares fitting between the feasible domain boundary points and the preset node voltage safety benchmark to generate the slope adjustment amount of the partition power limit curve and the offset compensation value of the node voltage safety threshold.

[0012] Preferably, the partition coupling calculation module includes: A sensitivity matrix construction unit is used to extract the mutual admittance parameters of cross-partition node pairs from the power-voltage Jacobian matrix and construct a block diagonal subarray of the inter-partition power sensitivity matrix; The spectral analysis unit is used to calculate the maximum eigenvalue modulus of the block diagonal subarray, compare the maximum eigenvalue modulus with a preset electrical decoupling reference value, and output the inter-interval coupling strength level. An independent partition identification unit is used to mark the corresponding partition as an electrically independent partition and generate a partition division topology diagram when the coupling strength level is lower than the dynamic decoupling threshold.

[0013] Preferably, the load-bearing capacity assessment engine includes: An uncertainty modeling unit is used to convert the uncontrollable parameter fluctuation data into a probability density function of photovoltaic power output and a Markov transition matrix of load change; The segmented analysis unit is used to divide the distribution network into multiple capacity assessment segments according to the partitioned power limit curve, with each segment corresponding to a different level of simplification of the linearized power flow equations. An optimization unit is used to embed the probability density function and Markov transition matrix into the opportunity-constrained optimization model, and to solve for the maximum grid-connectable photovoltaic capacity of each partition using the branch and bound method. The strategy generation unit is used to output a differentiated control instruction sequence for each electrical independent zone based on the superposition result of the safety supply control strategy set and the partition topology diagram.

[0014] Preferably, the system further includes: a real-time correction module, used to periodically collect deviation data between the actual photovoltaic access capacity and the theoretical maximum access capacity of the distribution network, and generate weight correction coefficients for the opportunity constraint optimization model; The real-time correction module includes: The deviation analysis unit is used to calculate the root mean square error between the actual photovoltaic grid connection rate and the predicted grid connection rate, and adjust the confidence interval width of the probability density function according to the root mean square error. The model iteration unit is used to feed back the weight correction coefficients to the piecewise analysis unit, triggering a dynamic reorganization of the simplified level of the linearized power flow equations.

[0015] Preferably, the system further includes: a topology adaptive module, used to recalculate the coupling relationship between the electrical independence index and the dynamic safe operation constraint set when detecting a topology change event in the distribution network; The topology adaptive module includes: The event response unit is used to identify topology change flags triggered by adding a new distributed power supply access point or line reconfiguration operations. The fast reconfiguration unit is used to locally update the cross-admittance parameters related to the changed region in the power-voltage Jacobian matrix and synchronously adjust the node affiliation relationships of the partitioned topology diagram.

[0016] Preferably, the system further includes: a multi-timescale coordination module, used to perform hierarchical synchronization of the execution cycle of the dynamic data acquisition module and the carrying capacity assessment engine according to the differences in the update frequency of the distribution network operating status; The multi-timescale coordination module includes: The fast and slow channel separation unit is used to divide the node power injection sequence into millisecond-level sampling channels and minute-level sampling channels; The clock alignment unit is used to establish a timestamp mapping relationship between different sampling channels in the opportunity constraint optimization model to ensure the timeliness consistency of the dynamic safe operation constraint set.

[0017] Preferably, the system further includes: a data-driven compensation module, used to activate a non-complete dimensionality-upgrading data-driven compensation mechanism when the fitting residual of the linearized power flow state space model is detected to exceed a threshold; The data-driven compensation module includes: The residual monitoring unit is used to compare the absolute deviation between the actual measured voltage value and the model predicted voltage value. The compensation activation unit is used to embed the power-voltage mapping relationship in the historical operating data into the Jacobian matrix generation unit in the form of a lookup table when the absolute deviation continues to exceed the residual tolerance.

[0018] Preferably, the system further includes: a dynamic weight adjustment module, used to adjust the weight allocation of each partition in the chance constraint optimization model in real time according to changes in the operating conditions of the distribution network; The dynamic weight adjustment module includes: The operating condition identification unit is used to detect sudden load changes or abnormal output events of distributed power sources in the current power distribution network. The weight optimization unit is used to dynamically calculate the weight coefficients of each partition in the opportunity constraint optimization model based on the severity of the event, wherein the weight coefficients of the critical load concentration area are positively correlated with voltage sensitivity.

[0019] Compared with the prior art, the beneficial effects of the present invention are: The dynamic data acquisition module can acquire real-time power injection sequences, voltage deviation rate sequences, and uncontrollable parameter fluctuation data for each zone of the distribution network, breaking the limitations of traditional static data acquisition. Real-time data acquisition can promptly capture dynamic changes during distribution network operation. Whether it's random fluctuations in renewable energy output or sudden changes in load, these can be quickly sensed and recorded, providing comprehensive and accurate data input for subsequent state space construction, security domain analysis, and carrying capacity assessment. This ensures that the entire assessment process is always based on the actual operating state of the distribution network.

[0020] The multidimensional state space construction module performs a complete up-dimensional mapping between the node power injection sequence and the voltage deviation rate sequence, generating a linearized power flow state space model that includes a power-voltage Jacobian matrix and a simplified mapping table of control variables. This complete up-dimensional mapping fully explores the complex relationship between node power injection and voltage deviation rate. Compared to traditional simplified models, the constructed linearized power flow state space model can more comprehensively and accurately depict the power flow operation state of the distribution network. The power-voltage Jacobian matrix clearly reflects the sensitive relationship between power changes and voltage changes, while the simplified mapping table of control variables facilitates subsequent model calculations and analyses. This makes the state space model both highly accurate and practical, laying a reliable model foundation for subsequent security domain analysis and carrying capacity assessment.

[0021] The safety domain analysis module performs boundary correction processing on the linearized power flow state space model based on uncontrollable parameter fluctuation data, generating a dynamic set of safety operation constraints that includes partitioned power limit curves and node voltage safety thresholds. This module fully considers the uncertainties in distribution network operation, adjusting the safety operation boundaries in real time through uncontrollable parameter fluctuation data, thus overcoming the shortcomings of traditional static safety domain analysis, which cannot adapt to dynamic parameter changes. The dynamic set of safety operation constraints can be updated in real time with the fluctuation of uncontrollable parameters, ensuring that the safety domain of the distribution network always matches the actual operating environment. This avoids underestimating the carrying capacity due to overly conservative static constraints, and also prevents safety risks caused by neglecting parameter fluctuations, ensuring that the distribution network remains in a safe and controllable state under various operating scenarios.

[0022] The partition coupling calculation module calculates the electrical independence index between partitions based on the power-voltage Jacobian matrix and quantifies it using the spectral radius of the power sensitivity matrix between partitions, effectively solving the problem of traditional methods neglecting partition coupling relationships. The electrical independence index accurately reflects the degree of electrical correlation between partitions, and the quantification method using the spectral radius provides a clear numerical representation of this correlation. By accurately understanding the electrical independence between partitions, the mutual influence between partitions can be fully considered when assessing partition carrying capacity, avoiding assessment biases caused by ignoring coupling relationships. This makes the carrying capacity assessment results of each partition more consistent with the actual operating characteristics of the distribution network, providing an important basis for the rational allocation of subsequent photovoltaic access capacity.

[0023] The carrying capacity assessment engine integrates dynamic safety operation constraints and electrical independence indicators to construct a zone-level opportunity-constrained optimization model, outputting the maximum grid-connectable photovoltaic (PV) capacity and a set of safety supply control strategies for each zone. This opportunity-constrained optimization model seeks the optimal PV grid connection capacity scheme while considering various uncertainties, ensuring the safe operation of the distribution network and maximizing the PV acceptance potential of each zone. Compared to traditional deterministic optimization models, this model is better suited to the operational characteristics of distribution networks with high renewable energy penetration, and the output maximum grid-connectable PV capacity is more practical and reasonable. Simultaneously, the safety supply control strategy set provides specific operational guidance for the operation and regulation of the distribution network, enabling it to maintain stable operation through effective control strategies after a large number of PV installations, ensuring the safety and reliability of power supply, and further improving the distribution network's capacity to accept renewable energy and its operational management level. Attached Figure Description

[0024] Figure 1 This is a timing diagram of the power distribution network carrying capacity assessment system based on dynamic correction described in this invention. Figure 2 A flowchart illustrating the process of building a multidimensional state space module; Figure 3 Data flow diagram of key parameters for multiple stages of distribution network carrying capacity assessment; Figure 4 A flowchart illustrating the operation of the load-bearing capacity assessment engine. Detailed Implementation

[0025] 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, and 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.

[0026] Please see Figure 1 This invention provides a power distribution network carrying capacity assessment system based on dynamic correction. The system includes: a dynamic data acquisition module that acquires a set of real-time operating parameters for each zone of the power distribution network, including node power injection sequences, voltage deviation rate sequences, and uncontrollable parameter fluctuation data; a multi-dimensional state space construction module that performs a complete up-dimensional mapping process on the node power injection sequences and voltage deviation rate sequences to generate a linearized power flow state space model of the power distribution network, which includes a power-voltage Jacobian matrix and a simplified mapping table of control variables; a security domain analysis module that performs boundary correction processing on the linearized power flow state space model based on the uncontrollable parameter fluctuation data to generate a set of dynamic safe operation constraints, including zone power limit curves and node voltage safety thresholds; a zone coupling degree calculation module that calculates the electrical independence index between each zone based on the power-voltage Jacobian matrix, which is quantified by the spectral radius of the power sensitivity matrix between zones; and a carrying capacity assessment engine that integrates the dynamic safe operation constraint set and the electrical independence index to construct a zone-level opportunity constraint optimization model, outputting the maximum grid-connectable photovoltaic capacity and a set of safe power supply control strategies for each zone. The system achieves dynamic assessment and correction of the power distribution network's carrying capacity through modular design, ensuring that the assessment results adapt to real-time operating conditions.

[0027] Example 1: See Figure 2The operation of the multidimensional state space construction module begins with the processing of the node power injection sequence provided by the dynamic data acquisition module. The nonlinear power flow analysis unit imports this sequence into a preset nonlinear power flow equation for solution. This nonlinear power flow equation adopts a power balance equation in polar coordinates. The solution process uses an improved Newton-Raphson algorithm to enhance convergence. During algorithm initialization, the rated voltage of the distribution network is used as the initial estimate of the voltage of each node. During the iterative calculation, the norm of the power imbalance is monitored in real time. When the norm is less than the preset convergence accuracy threshold, the calculation is considered to have converged, thereby generating an initial voltage distribution field covering all nodes in the network. This distribution field contains the voltage amplitude and phase angle information of each node. At the same time, the unit calculates the total power loss in the network based on the converged power flow solution, and further derives the power loss gradient field through the adjoint network method or direct partial differential method. This gradient field quantifies the sensitivity of the small changes in the injected power of each node to the total system loss. The upgraded mapping unit then performs in-depth processing on the obtained initial voltage distribution field. Its core operation is to perform Taylor series expansion. This expansion takes the current system operating point, i.e., the initial voltage distribution field, as the basis point. It not only retains the first-order terms that traditional linearization relies on, but also systematically extracts second-order, third-order, and even higher-order terms. These higher-order terms together constitute the fully upgraded state space basis. The order of the Taylor series expansion is not fixed, but is adaptively adjusted according to the degree of deviation between the current operating point and the rated state. When the voltage deviation rate sequence shows that the system operating state is far from the rated point, a higher-order expansion will be automatically adopted to more accurately capture nonlinear characteristics. The construction of the fully upgraded state space basis is essentially the formation of a tensor structure containing the higher-order partial derivatives of voltage with respect to power.

[0028] The control variable compression unit dynamically maintains the simplified control variable mapping table in parallel. This mapping table is updated based on the power loss gradient field output by the nonlinear power flow analysis unit. An online feature extraction algorithm is integrated within the unit, continuously analyzing the intrinsic correlation between active and reactive power control variables. The power loss gradient field reveals the impact patterns of different control variable combinations on system losses. Based on these patterns, the unit uses principal component analysis to reduce the dimensionality of the original high-dimensional control variable space, identifying the dominant control direction that has the greatest impact on the system state. Dimensionality reduction correlation rules are recorded in the simplified control variable mapping table. These rules explicitly define how to map the original active and reactive power control variable sets to a new variable space with significantly reduced dimensionality, such as aggregating the associated control variables of adjacent nodes, thereby reducing the number of decision variables in the optimization problem and improving subsequent computational efficiency. The Jacobian matrix generation unit, as the final output of this module, is responsible for synthesizing the power-voltage Jacobian matrix. Its generation process deeply integrates the fully upgraded state-space basis provided by the upgraded mapping unit and the dimension-reduced correlation rules provided by the control variable compression unit. The generation process is not a simple calculation of the traditional power flow Jacobian matrix, but rather the formation of a complete Jacobian structure containing high-order information on the fully upgraded state-space basis. The dimension-reduced correlation rules are used to compress and approximate this structure, focusing on calculating those elements that still play a dominant role in the dimension-reduced space. For a large number of elements in the matrix that are close to zero, the unit applies sparsity techniques to process them, and calculates sparsity correction coefficients to selectively ignore these small interactions, generating a relatively sparse power-voltage Jacobian matrix that can reflect the nonlinear characteristics of the system and is easy to calculate numerically.

[0029] The security domain analysis module then performs boundary correction on the linearized power flow state space model. The parameter disturbance simulation unit then begins to work, reading uncontrollable parameter fluctuation data. This data typically includes the historical fluctuation statistics of distributed photovoltaic power output and the probability distribution of load changes. This unit uses the Monte Carlo simulation method to generate a large number of representative random disturbance scenarios. Each disturbance scenario contains a random photovoltaic power output fluctuation sequence and a potentially related load mutation increment sequence. The generation of the scenarios considers the temporal correlation and the spatial correlation between disturbances at different nodes to ensure that the simulated disturbance scenarios can cover various uncertainty modes that may occur in actual operation. The boundary search unit bears the critical task of determining the safe operating boundary of the system. It loads a large number of random disturbance scenarios generated by the parameter disturbance simulation unit into the constructed linearized power flow state space model. For each loaded disturbance scenario, the unit adopts the golden section method, a single-variable optimization search algorithm, to perform an efficient one-dimensional search in the power injection space. The search direction is usually along the direction from the current operating point to the power injection limit. The golden section method accurately locates the boundary point of the feasible region by continuously narrowing the search interval. This boundary point is defined as the limit value of power injection that the system can withstand under the condition of satisfying all node voltage constraints and line thermal stability constraints. This process is repeated in all key power injection directions to outline the boundary contour of the safe operating domain of the system.

[0030] The threshold fitting unit then performs mathematical processing on a large number of feasible domain boundary points obtained from the search. These boundary points are scattered in a high-dimensional power-voltage space. The unit associates these boundary points with a preset node voltage safety benchmark, which is usually specified by the operating procedures and represents the upper and lower limits that each node voltage must maintain. The fitting process uses the least squares method, aiming to find a set of linear relationship parameters such that the power limit curve determined by this relationship can optimally fit all the searched boundary points. By minimizing the fitting error, the unit calculates the slope adjustment amount required for the partition power limit curve. This adjustment amount reflects the actual impact of the allowed power injection changes in the partition on the critical node voltage under the current network structure and operating state. At the same time, the unit also calculates the offset compensation value of the node voltage safety threshold. This compensation value is used to fine-tune the theoretical voltage safety upper limit to absorb the errors caused by model linearization and the effects of system uncertainties. Finally, it outputs dynamically updated partition power limit curves and node voltage safety thresholds, which constitute the core content of the dynamic safe operation constraint set and provide accurate boundary conditions for subsequent load-bearing capacity assessment.

[0031] See Figure 3In the integrated visualization of multi-dimensional state space construction and security domain analysis of distribution networks, the technical implementation relies on the collaborative processing of dynamic data acquisition and multi-dimensional state mapping. In specific operations, subgraph (a) is based on the time series curve of power injection of node AC, and uses blue, orange and green broken lines to show the quantification process of non-stationary operation characteristics, where the power injection sequence serves as the input dimension for state space construction; subgraph (b) uses dashed lines to present voltage deviation rate fluctuations, and quantifies the heterogeneity of stability margin by calculating the spatiotemporal variance of voltage deviation between nodes, supporting the dynamic correction of the security domain boundary; subgraph (c) uses a bar-line composite chart to map the partitioned power limit to the voltage safety threshold, where the safety threshold range is set to 1.06-0.94 pu, and the boundary slope is optimized by least squares fitting. During the quantization process, the power injection sequence and voltage deviation rate data are mapped to generate a linearized power flow state space model through a fully upgraded dimension. The sparsity correction coefficient of the Jacobian matrix is ​​dynamically calculated based on the node coupling degree. The safe domain boundary search uses the golden section method to locate the feasible region in the power injection space. The fitting error between the boundary point and the voltage reference is evaluated by the root mean square index.

[0032] Example 2: See Figure 4 The startup of the partition coupling degree calculation module relies on the power-voltage Jacobian matrix generated by the multi-dimensional state space construction module. The sensitivity matrix construction unit analyzes this matrix, and its core task is to extract parameters that reflect the electrical interaction strength between different distribution network partitions from this global matrix. The unit pre-sets a preliminary partitioning scheme for the distribution network, which is usually initially defined based on the physical topology of the power grid (such as feeders and substation power supply range). The extraction process focuses on the boundary nodes connecting different partitions, specifically identifying and extracting the mutual admittance parameters from the complete Jacobian matrix that have row indices belonging to one partition and column indices belonging to another partition. These parameters quantitatively describe the direct impact of a change in the power of a node in one partition on the voltage of a node in another partition. The extracted cross-admittance parameters are used to construct the block diagonal subarray of the power sensitivity matrix between partitions. This subarray is not a matrix that only contains the main diagonal blocks in the traditional sense, but specifically refers to a macroscopic model that regards the entire network as a series of partition subsystems, where each sub-block corresponds to the sensitivity relationship of a partition itself. The block diagonal subarray constructed by the module specifically includes these sub-blocks located in "off-diagonal" positions that represent the coupling relationship between partitions, thus forming a simplified model that can clearly show the interaction relationship between partitions.

[0033] The spectral analysis unit then performs an in-depth analysis of the block diagonal subarrays of the constructed interval power sensitivity matrix. This unit calculates the maximum eigenvalue modulus of the subarray using numerical methods such as the QR algorithm to ensure stability and accuracy. The magnitude of the maximum eigenvalue modulus directly reflects the comprehensive strength of coupling effects across all intervals and is a global quantitative indicator. This modulus value is compared with a preset electrical decoupling benchmark value, which is a threshold value set based on a large amount of historical operating data, network structure characteristics, and stability requirements. The comparison result is quantified into an interval coupling strength level, such as "high coupling," "medium coupling," and "low coupling." If the maximum eigenvalue modulus far exceeds the benchmark value, it is determined that the system intervals are tightly coupled and have significant mutual influence. If the modulus value is close to or lower than the benchmark value, it indicates that the electrical independence between intervals is relatively strong. The independent partition identification unit performs partition identification operations based on the coupling strength level output by the spectral analysis unit. It has a dynamic decoupling threshold, which may be fine-tuned according to the overall operational risk level of the system. When the coupling strength level of a partition with all other associated partitions is lower than this dynamic decoupling threshold after calculation, the unit marks this partition as an electrically independent partition. After marking, the unit generates a partition topology diagram, which graphically and clearly shows the range of all marked electrically independent partitions, their connection relationships, and the location of boundary nodes. This topology diagram is an important basis for subsequent autonomous partition control and independent load-bearing capacity assessment.

[0034] The load capacity assessment engine is the core component that ultimately outputs the assessment results. The uncertainty modeling unit performs probabilistic processing on the uncontrollable parameter fluctuation data of the input system. It fits the historical data of distributed photovoltaic power output fluctuations into a continuous probability density function, usually using a Gaussian mixture model to better characterize its randomness and intermittency. For load changes, this unit uses a discrete-state Markov chain for modeling. By analyzing the state transition patterns of historical load sequences, it constructs a Markov transition matrix for load changes, which describes the probability of the load transitioning from the current state to various possible future states. Based on the partitioned power limit curves provided by the security domain analysis module, the segmented analysis unit divides the entire distribution network's operating range into multiple continuous capacity assessment segments, such as light-load, normal-load, and heavy-load segments. Each segment corresponds to a different level of simplification of the linearized power flow equations. In the light-load segment, the system nonlinearity is relatively weak, so a highly simplified linear model can be used to improve computational speed. In the heavy-load segment, the system is close to its stability limit and nonlinearity is significant, so a refined model that retains more higher-order terms is used to ensure assessment accuracy. This segmented processing method achieves an adaptive balance between computational efficiency and model accuracy.

[0035] The optimization unit is responsible for solving the core opportunity-constrained optimization model. This model takes the maximum grid-connectable photovoltaic capacity as the optimization objective and requires, in probabilistic form (i.e., opportunity constraints), that the probability of events such as node voltage exceeding limits and line overload occurring is lower than a given confidence level. The unit embeds the probability density function and Markov transition matrix generated by the uncertainty modeling unit into the constraints, enabling the model to fully consider the inherent uncertainties of photovoltaic output and load demand. Solving such complex optimization problems with probabilistic constraints usually adopts the branch and bound method. This method gradually narrows the search range by continuously dividing the feasible region into smaller subsets (branches) and calculating the upper and lower bounds of the objective function of each subset (bounds), and finally finds the global optimal solution or near-optimal solution that satisfies the maximum grid-connectable photovoltaic capacity of each partition under all opportunity constraints. Finally, the strategy generation unit overlays the calculation results of the optimization solution unit with the partition topology diagram generated by the partition coupling degree calculation module. Based on the electrical independence degree of each partition, the maximum grid-connectable photovoltaic capacity, and the specific equipment status within the partition (such as the presence or absence of reactive power compensation devices, load importance, etc.), it generates differentiated control command sequences for each electrically independent partition. For partitions with high electrical independence and strong internal resource regulation capabilities, the control strategy may focus more on utilizing local reactive power resources for voltage support. For partitions with still high coupling degree or concentrated critical loads, the control strategy will include more conservative load shedding and photovoltaic output restriction measures to ensure power supply. These command sequences constitute a set of specific executable safety power supply control strategies.

[0036] Example 3: The real-time correction module enables the entire evaluation system to continuously evolve. It periodically collects actual operating data from the distribution network monitoring system. Its core function is to obtain the actual photovoltaic (PV) grid connection capacity of each zone and the theoretical maximum grid-connectable capacity calculated by the carrying capacity assessment engine. The system compares these two sets of data to generate a deviation data sequence, reflecting the difference between the model prediction and the actual grid absorption capacity. The deviation data is fed into an adaptive filter for smoothing to eliminate short-term disturbances in measurement noise. Subsequently, based on the processed deviation sequence, correction coefficients are generated to adjust the weights of the internal parameters of the opportunity constraint optimization model. These coefficients serve as a feedback signal to narrow the gap between the model and reality. The deviation analysis unit within this module is specifically responsible for handling data uncertainty. It calculates the root mean square error (RMSE) between the actual PV grid absorption rate and the model-predicted absorption rate within a statistical period. The actual PV grid absorption rate is calculated from the actual power generation and load power measured at the grid connection point, while the predicted absorption rate comes from the result of the previous optimization solution. The formula for calculating the RMS error is:

[0037] Where: N represents the total number of sampling points within an evaluation period, and k is the sampling point index. It is the actual photovoltaic absorption rate at the k-th sampling time. This is the predicted absorption rate at the corresponding time; this error value The magnitude of the value directly determines the direction and width of the confidence interval of the probability density function in the uncertainty modeling unit. When the error is large, the unit will appropriately widen the confidence interval of the probability density function to acknowledge the existence of uncertainties that are not fully recognized in the model. Conversely, it will narrow the interval to obtain a more accurate optimization solution.

[0038] The model iteration unit is responsible for converting the correction coefficients into model structure updates. It receives the weight correction coefficients from the deviation analysis unit. These coefficients are multi-dimensional vectors, with different components corresponding to the importance weights of different partitions or different constraints in the chance-constrained optimization model. This unit feeds these coefficients back to the segmented analysis unit inside the carrying capacity assessment engine, triggering a dynamic reorganization process at the simplified level of the distribution network linearized power flow equations. The reorganization logic depends on the change pattern of the weight coefficients. If the weight correction coefficient of a certain partition increases significantly, it indicates that the model error in that area has a greater impact on the overall assessment. The segmented analysis unit will divide the assessment segment corresponding to that partition into finer segments and use a higher-order linearized model within that segment to improve local accuracy. Conversely, it may merge segments or use a simpler model to improve the overall calculation speed.

[0039] The topology adaptive module ensures the system can respond quickly to changes in the distribution network's physical structure without restarting the entire modeling process. It continuously monitors signals from the power grid dispatch center or distribution automation system to detect topology change events, typically including new distributed generation points, network reconfiguration due to line switch changes, or equipment maintenance shutdowns. Once such an event is detected, the module immediately initiates a recalculation process, focusing on whether the coupling relationship between electrical independence indices and the dynamic safety operation constraint set has fundamentally changed after the change. The event response unit within this module acts as a trigger, identifying topology change indicators by parsing communication messages sent by the power grid energy management system or monitoring remote signaling changes of specific switches. New distributed generation points trigger the "source addition" flag, while line reconfiguration operations trigger the "connection relationship change" flag. Each flag includes a timestamp and an indication of the affected area. The rapid reconfiguration unit begins operation after a flag is triggered. Its design principle is to perform local updates rather than global reconfiguration to maximize efficiency. Based on the event-affected area identifier, the unit locates the areas in the power-voltage Jacobian matrix that need modification. These areas typically involve only a few nodes directly connected to the changed node. The unit uses pre-calculated changes in the node admittance matrix and matrix perturbation theory to quickly deduce the incremental changes in the corresponding mutual admittance parameters in the power-voltage Jacobian matrix, updating only these parameters to avoid recalculating the entire massive matrix. While updating electrical parameters, the unit simultaneously adjusts the partitioned topology diagram generated by the partitioned coupling degree calculation module, re-determining the membership relationships of nodes around the changed point. For example, the closing of a tie switch may merge two previously independent electrical partitions into a new partition. The unit will quickly refresh the partitioned topology diagram based on the updated coupling degree calculation, thereby ensuring the real-time accuracy of the network model.

[0040] The collaborative work of the real-time correction module and the topology adaptation module constitutes the system's self-correction mechanism. The real-time correction module focuses on soft adjustments to model parameters in response to slow drift or changes in uncertainty patterns of operating parameters, which is a continuous fine-tuning process. The topology adaptation module, on the other hand, performs hard corrections to the model structure in response to hard changes in the physical connection relationship of the network. Together, these two mechanisms ensure that the carrying capacity assessment system can maintain the reliability and practicality of its assessment results in a dynamically changing real power grid environment. Fine-tuning prevents the model from gradually deviating from reality, while hard correction prevents the model from failing instantly due to structural abrupt changes.

[0041] Example 4: The implementation of the multi-timescale coordination module aims to solve the fusion problem caused by inconsistent update frequencies of data from different sources in the distribution network. The core function of this module is to perform hierarchical synchronization of the execution cycles of the dynamic data acquisition module and the carrying capacity assessment engine. In a specific example of a distribution network application in an urban industrial park, the data sources include high-precision synchronous phasor measurement units (PMUs) installed at key nodes and widely deployed distribution automation terminals (FTUs). The node voltage and current phasor data provided by the PMUs update at a rate of 100 frames per second, forming a millisecond-level sampling channel, while the data such as line power and switch status collected by the FTUs typically update every 2 to 5 minutes, forming a minute-level sampling channel. The fast and slow channel separation unit classifies the input node power injection sequence in real time using the timestamp and source device identifier in the data packet header. The millisecond-level channel data is marked as high priority and directly stored in a high-speed circular buffer for real-time analysis, while the minute-level channel data is marked as regular priority and sent to a regular database for persistent storage. This separation mechanism effectively avoids the problem of low-speed data blocking the high-speed data processing flow. The clock alignment unit is crucial for ensuring data timeliness consistency. Internally, it maintains a high-precision system logical clock. This unit assigns a uniform logical timestamp to every arriving data packet. For example, at 14:05:30.800, it might simultaneously receive millisecond-level data packets collected by the PMU at 14:05:30.800 and minute-level data packets collected by the FTU at 14:05:25.000 (the FTU's collection cycle is 5 minutes). The alignment unit does not discard delayed low-speed data. Instead, it employs a predictive interpolation algorithm based on the smoothness of grid state changes. It utilizes the trend of recent high-speed data to interpolate low-speed data forward on the logical time axis, estimating its approximate value at the latest logical timestamp. This allows data from different channels to be fused and calculated under the same time base, providing the capacity assessment engine with a time-consistent snapshot of the dynamic safety operation constraint set.

[0042] The data-driven compensation module serves as a safety net when the accuracy of the theoretical model declines. Its implementation can be illustrated through a specific scenario: When the industrial park's power distribution network encounters a rapidly moving cloud cover on a summer afternoon, the distributed photovoltaic output experiences severe fluctuations. This fluctuation pattern may exceed the typical scenario library built by the linearized power flow state space model based on historical data. The residual monitoring unit continuously operates, comparing the actual voltage values ​​measured by the PMU at key nodes with the voltage values ​​predicted by the model based on the current node's power injection, calculating the absolute deviation between the two. This unit sets a dynamically adjusted residual tolerance threshold for each node, which is typically correlated with the historical standard deviation of the node's voltage fluctuations. The residual tolerance refers to the maximum permissible absolute deviation between the actual measured voltage values ​​at each node of the distribution network and the voltage values ​​predicted by the linearized power flow state space model. When this absolute deviation continuously exceeds the residual tolerance, it indicates that the theoretical calculation results of the model deviate too much from the actual operating state of the power grid, requiring the activation of the data-driven compensation mechanism to correct the model error and ensure the accuracy of subsequent capacity assessment results. The residual tolerance is calculated based on historical operating data of node voltage. The actual measured voltage data of the target node over the past 90 days (the statistical period can be adjusted according to the stability of the power grid operation; it can be extended to 180 days for a stable power grid and shortened to 30 days for a power grid with frequent fluctuations) is extracted at each hour. (t is the time index, unit: h), with a total of M data points, the historical predicted voltage data of the node over the past 90 days is calculated by substituting the corresponding node power injection sequence back into the current linearized power flow state-space model. Calculate the historical voltage deviation sequence of this node. And calculate the standard deviation of the sequence. The residual tolerance is set at 1.5 times the standard deviation of the historical voltage deviation sequence. This value covers the voltage deviation fluctuation range during normal grid operation and effectively identifies model errors exceeding normal fluctuations, avoiding false triggering of the compensation mechanism. At the beginning of each month, the residual monitoring unit of the data-driven compensation module automatically repeats the above calculation steps to update the residual tolerance of each node. If a node triggers the data-driven compensation mechanism more than 3 times in the previous month (indicating that the node's operating status fluctuates greatly and the original residual tolerance may be too small), the residual tolerance calculation coefficient for that node in the following month is adjusted from 1.5 to 1.8; if a node does not trigger the compensation mechanism in the previous month (indicating that the node's operating status is stable and the original residual tolerance can be appropriately tightened), the calculation coefficient is adjusted from 1.5 to 1.2, achieving dynamic adaptation of the residual tolerance to the actual operating characteristics of the grid.

[0043] Refer to Table 1, which describes the residual monitoring of three key nodes at several consecutive sampling times.

[0044] Table 1: Monitoring of Nodal Voltage Prediction Residuals Node number Timestamp Measured voltage (pu) Model predicted voltage (pu) absolute deviation Residual tolerance (pu) Is it beyond the limit? N101 T1 1.032 1.028 0.004 0.010 no N101 T2 1.045 1.029 0.016 0.010 yes N205 T1 0.981 0.985 0.004 0.015 no N205 T2 0.978 0.986 0.008 0.015 no N307 T1 1.058 1.051 0.007 0.012 no N307 T2 1.062 1.050 0.012 0.012 yes The compensation activation unit makes decisions based on the output of the residual monitoring unit. When it detects that, as with nodes N101 and N307 at time T2, the absolute deviation not only exceeds their individual tolerance but also continues to occur within a preset time window, the unit determines that the pure theoretical model can no longer accurately track the actual state of the system. At this time, the module activates a non-completely upgraded data-driven compensation mechanism. The so-called "non-completely upgraded" means that this mechanism does not attempt to reconstruct a complex new model, but instead quickly retrieves historical scenarios similar to the current operating point from the historical operating database. These historical scenarios contain the correspondence between the actual power injection and voltage measurement values ​​of each node under similar total load and similar total photovoltaic output levels. The compensation mechanism embeds the "power-voltage" mapping relationship formed by these historical data points into the calculation process of the Jacobian matrix generation unit in a lookup table manner. When the Jacobian matrix elements calculated by the theoretical model are used for state prediction, they are weighted and averaged with the results obtained from the data-driven lookup table. The weights are dynamically adjusted according to the current residual size. The larger the residual, the higher the weight of the data-driven part, thereby using historical experience to quickly correct the instantaneous model error.

[0045] The collaborative work of the multi-timescale coordination module and the data-driven compensation module reflects the system's strategy for coping with complex operating conditions. The former solves the problem of data consistency over time, providing high-quality, synchronous input to the evaluation engine; the latter acts as a "safety net" for model accuracy. When the theoretical model's accuracy deteriorates due to deviations from the norm at system operating points or encountering unmodeled dynamics, data-driven experience is introduced to maintain the reliability of the overall evaluation results. This design enables the distribution network carrying capacity evaluation system to not only rely on accurate physical models but also possess the ability to learn from historical data and adaptively adjust, enhancing its robustness in real-world uncertain environments.

[0046] Example 5: The function of the dynamic weight adjustment module is to realize adaptive resource allocation in the process of distribution network carrying capacity assessment. Its core lies in dynamically adjusting the weight coefficients of each partition in the opportunity constraint optimization model according to the real-time changes in the power grid operating conditions. The implementation process of the optimization model begins with the real-time perception of operating conditions and the quantitative assessment of event severity. The operating condition identification unit in the dynamic weight adjustment module continuously monitors the real-time data stream of the distribution network. When events such as a sudden increase in power demand in a critical load concentration area or an abnormal drop in the output of distributed power sources are detected, the unit will immediately extract features and classify the events. The assessment of event severity is not a single-dimensional judgment, but a comprehensive consideration of multiple factors such as the proportion of power deficit to the load base of the partition, the time gradient of power change, and the socio-economic importance level of the affected load, forming a quantitative severity index. The weight optimization unit dynamically adjusts the weight coefficients of each partition in the chance-constrained optimization model based on the calculated severity index and the node voltage sensitivity parameters obtained from the power-voltage Jacobian matrix. The adjustment process follows two principles: prioritizing voltage stability and ensuring critical loads. For partitions with high voltage sensitivity and important loads, their weight coefficients will be significantly increased. This increase is directly reflected in the reconstruction of the objective function of the optimization model. During the calculation process, the optimization solver will assign a larger violation penalty factor to the constraints of high-weight partitions, making the solution process naturally tilted towards prioritizing the safe operation conditions of these partitions. The optimization unit performs calculations within the model embedded with dynamic weights. Its core objective is to solve a mathematical programming problem that aims to maximize the grid's grid-accessible photovoltaic capacity while satisfying a series of weighted security constraints. Because the introduction of weight coefficients alters the original objective function, the solver needs to use an iterative algorithm to gradually approach the optimal solution. In each iteration, the solver verifies the voltage constraints, line capacity constraints, and other conditions of each partition. However, for high-weight partitions, the tolerance for violations of these constraints is set extremely low, even zero, while for lower-weight partitions, a certain degree of constraint relaxation is allowed in extreme cases. This differentiated processing mechanism ensures that, under conditions of limited global computational resources, the power supply security and power quality of the most critical areas are always prioritized. The model does not generate a fixed photovoltaic (PV) grid connection capacity value after the solution is completed. Instead, it generates a set of optimal solutions corresponding to the current weight allocation. The weight optimization unit analyzes the overall operating status of the system under this solution set. If it finds that the weight adjustment leads to excessively low resource utilization in some non-critical areas or excessively severe degradation of the overall optimization objective function value, the unit will initiate a fine-tuning procedure for the weight coefficients. By observing the response of the overall system performance through small-step, tentative weight changes, the unit finds a balance point that can effectively ensure the safety of critical areas while taking into account the overall absorption capacity. Finally, it outputs a maximum PV grid connection capacity scheme that matches the dynamic weights and corresponding safety control strategies.

[0047] Consider a power distribution network partition example that includes a residential area, a commercial center, and an important hospital. On a sunny weekday afternoon, the load is relatively stable. The initial weights of each partition in the optimization model may be set based on its peak load ratio, with the commercial area having a higher weight and the residential area a lower weight. However, as evening approaches, the residential load rises sharply, and at the same time, a dark cloud obscures the photovoltaic power station, causing a sharp drop in output. This simultaneous change in load and power supply constitutes a typical sudden change in operating conditions. The operating condition identification unit continuously monitors such events. It analyzes real-time data streams from the data acquisition system. This unit has multiple filters and comparators to detect load mutation events. The principle is to calculate the instantaneous rate of change of the total load of each zone. When the rate of change exceeds a threshold calculated based on historical data (e.g., a change exceeding 15% of the average load of the zone per minute), a load mutation flag is triggered. The flag information includes the mutation zone identifier, the amount of change, and the direction (increase or decrease). For distributed power generation output anomalies, the detection logic is similar, but it focuses on the rate and magnitude of the output decline. When the photovoltaic output drops by more than 30% of its maximum output within a few minutes, a power anomaly flag is triggered. The unit can distinguish between planned output adjustments and unexpected abnormal fluctuations.

[0048] Upon receiving the event flag from the operating condition identification unit, the weight optimization unit immediately initiates calculations. This unit embeds a weight allocation algorithm that assesses the severity of the event. Severity is a comprehensive indicator determined by the event type (load surge or power failure), the scope of impact (involving several zones), the size of the power deficit (in megawatts), and the speed at which the event occurs. For example, if the load in the zone where the hospital is located increases significantly due to an emergency medical mission, and a major distributed photovoltaic line supplying power to the zone trips due to a fault, this event is considered to be of high severity because it involves critical loads and has a large power deficit. The calculation of weighting coefficients follows the principle of prioritizing voltage sensitivity. The unit extracts the voltage sensitivity parameter of each partition's central node or critical node to power injection from the power-voltage Jacobian matrix. This parameter quantifies the impact of partition power balance changes on voltage stability. For critical load concentration areas such as hospitals, the voltage stability requirements are extremely high, so their voltage sensitivity values ​​are relatively large. The weighting optimization unit makes the weighting coefficient of this partition positively correlated with the voltage sensitivity, specifically through a variant of the weighted least squares method. When constructing the objective function of the optimization model, the cost of violating constraints such as voltage deviation and power limit exceedance in high-weight partitions is set much higher than that in low-weight partitions. This makes the solver prioritize ensuring the power supply security and power quality of high-weight partitions during global optimization, even if it means more strictly limiting the photovoltaic access capacity of other non-critical partitions or taking more aggressive load shedding measures.

[0049] The implementation of dynamic weight adjustment is a closed-loop feedback process. The weight coefficients are not set once and then remain unchanged. The module continuously monitors the system response after the weight adjustment. If the voltage stability of the critical zone improves after the adjustment but the overall photovoltaic absorption capacity decreases too much, the weight optimization unit will fine-tune the weight coefficients to find a balance point. Conversely, if the overall system operation remains unstable, the weight of the critical zone will be further increased. This dynamic adjustment ensures that when the distribution network responds to emergencies, limited control resources (such as reactive power compensation devices, energy storage systems, and interruptible loads) can be prioritized to ensure the power supply reliability of the most critical areas. This achieves the best balance between safety and economy at the global level, making the carrying capacity assessment result not only a static number but also a dynamic decision-making basis that can adapt to changes in the grid state.

[0050] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0051] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A power distribution network carrying capacity assessment system based on dynamic correction, characterized in that, include: The dynamic data acquisition module is used to acquire the real-time operating parameter set of each zone of the distribution network. The real-time operating parameter set includes the node power injection sequence, voltage deviation rate sequence and uncontrollable parameter fluctuation data. A multidimensional state space construction module is used to perform a complete up-dimensional mapping process on the node power injection sequence and the voltage deviation rate sequence to generate a linearized power flow state space model of the distribution network. The linearized power flow state space model includes a power-voltage Jacobian matrix and a simplified mapping table of control variables. The safety domain analysis module is used to perform boundary correction processing on the linearized power flow state space model based on the uncontrollable parameter fluctuation data, and generate a dynamic safety operation constraint set, which includes partition power limit curves and node voltage safety thresholds. The partition coupling degree calculation module is used to calculate the electrical independence index between each partition based on the power-voltage Jacobian matrix. The electrical independence index is quantified by the spectral radius of the power sensitivity matrix between partitions. The carrying capacity assessment engine is used to integrate the dynamic safe operation constraint set and electrical independence index to construct an opportunity constraint optimization model at the zonal level, and output the maximum grid-connectable photovoltaic capacity and safe supply control strategy set for each zonal.

2. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 1, characterized in that, The multidimensional state space construction module includes: The nonlinear power flow analysis unit is used to input the node power injection sequence into a preset nonlinear power flow equation to generate an initial voltage distribution field and a power loss gradient field. The dimension-upgrading mapping unit is used to perform Taylor series expansion on the initial voltage distribution field and extract second-order and above terms to form a fully dimension-upgrading state space basis. The control variable compression unit is used to dynamically update the simplified control variable mapping table according to the power loss gradient field. The simplified control variable mapping table records the dimensionality reduction association rules of active and reactive control variables. The Jacobian matrix generation unit is used to calculate the sparsification correction coefficients of the power-voltage Jacobian matrix based on the fully upgraded state space basis and the reduced-dimensional association rule.

3. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 2, characterized in that, The security domain analysis module includes: The parameter disturbance simulation unit is used to generate multiple sets of random disturbance scenarios based on the uncontrollable parameter fluctuation data. Each set of disturbance scenarios includes the distributed photovoltaic power output fluctuation and load mutation increment. The boundary search unit is used to successively load the random disturbance scenario into the linearized power flow state space model and search for the feasible boundary points of the power injection space using the golden section method. The threshold fitting unit is used to perform least-squares fitting between the feasible domain boundary points and the preset node voltage safety benchmark to generate the slope adjustment amount of the partition power limit curve and the offset compensation value of the node voltage safety threshold.

4. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 3, characterized in that, The partition coupling degree calculation module includes: A sensitivity matrix construction unit is used to extract the mutual admittance parameters of cross-partition node pairs from the power-voltage Jacobian matrix and construct a block diagonal subarray of the inter-partition power sensitivity matrix; The spectral analysis unit is used to calculate the maximum eigenvalue modulus of the block diagonal subarray, compare the maximum eigenvalue modulus with a preset electrical decoupling reference value, and output the inter-interval coupling strength level. An independent partition identification unit is used to mark the corresponding partition as an electrically independent partition and generate a partition division topology diagram when the coupling strength level is lower than the dynamic decoupling threshold.

5. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 4, characterized in that, The load-bearing capacity assessment engine includes: An uncertainty modeling unit is used to convert the uncontrollable parameter fluctuation data into a probability density function of photovoltaic power output and a Markov transition matrix of load change; The segmented analysis unit is used to divide the distribution network into multiple capacity assessment segments according to the partitioned power limit curve, with each segment corresponding to a different level of simplification of the linearized power flow equations. An optimization unit is used to embed the probability density function and Markov transition matrix into the opportunity-constrained optimization model, and to solve for the maximum grid-connectable photovoltaic capacity of each partition using the branch and bound method. The strategy generation unit is used to output a differentiated control instruction sequence for each electrical independent zone based on the superposition result of the safety supply control strategy set and the partition topology diagram.

6. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 5, characterized in that, Also includes: The real-time correction module is used to periodically collect the deviation data between the actual photovoltaic access capacity and the theoretical maximum access capacity of the distribution network, and generate the weight correction coefficients of the opportunity constraint optimization model. The real-time correction module includes: The deviation analysis unit is used to calculate the root mean square error between the actual photovoltaic grid connection rate and the predicted grid connection rate, and adjust the confidence interval width of the probability density function according to the root mean square error. The model iteration unit is used to feed back the weight correction coefficients to the piecewise analysis unit, triggering a dynamic reorganization of the simplified level of the linearized power flow equations.

7. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 6, characterized in that, It also includes: a topology adaptive module, used to recalculate the coupling relationship between the electrical independence index and the dynamic safe operation constraint set when detecting topology change events in the distribution network; The topology adaptive module includes: The event response unit is used to identify topology change flags triggered by adding a new distributed power supply access point or line reconfiguration operations. The fast reconfiguration unit is used to locally update the cross-admittance parameters related to the changed region in the power-voltage Jacobian matrix and synchronously adjust the node affiliation relationships of the partitioned topology diagram.

8. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 7, characterized in that, Also includes: A multi-timescale coordination module is used to perform hierarchical synchronization of the execution cycle of the dynamic data acquisition module and the carrying capacity assessment engine based on the differences in the update frequency of the distribution network operation status. The multi-timescale coordination module includes: The fast and slow channel separation unit is used to divide the node power injection sequence into millisecond-level sampling channels and minute-level sampling channels; The clock alignment unit is used to establish a timestamp mapping relationship between different sampling channels in the opportunity constraint optimization model to ensure the timeliness consistency of the dynamic safe operation constraint set.

9. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 8, characterized in that, Also includes: The data-driven compensation module is used to activate the incomplete dimensionality-upgrading data-driven compensation mechanism when the fitting residual of the linearized power flow state space model is detected to exceed a threshold. The data-driven compensation module includes: The residual monitoring unit is used to compare the absolute deviation between the actual measured voltage value and the model predicted voltage value. The compensation activation unit is used to embed the power-voltage mapping relationship in the historical operating data into the Jacobian matrix generation unit in the form of a lookup table when the absolute deviation continues to exceed the residual tolerance.

10. The power distribution network carrying capacity assessment system based on dynamic correction according to claim 1, characterized in that, Also includes: The dynamic weight adjustment module is used to adjust the weight allocation of each partition in the opportunity constraint optimization model in real time according to the changes in the operating conditions of the distribution network. The dynamic weight adjustment module includes: The operating condition identification unit is used to detect sudden load changes or abnormal output events of distributed power sources in the current power distribution network. The weight optimization unit is used to dynamically calculate the weight coefficients of each partition in the opportunity constraint optimization model based on the severity of the event, wherein the weight coefficients of the critical load concentration area are positively correlated with voltage sensitivity.

Images