A method for energy balance control of self-powered power supply of pole-mounted vacuum circuit breaker

Abstract

This invention discloses an energy balance control method for a self-powered pole-mounted vacuum circuit breaker, relating to the field of distributed control optimization technology. The method includes: constructing initial feature vectors for energy storage nodes and performing temperature compensation; using particle swarm optimization analysis to adjust amplitude and direction to generate a first distributed control vector set; combining real-time temperature adaptive correction to obtain a second distributed control vector set; issuing commands in parallel; each energy storage node independently recursively updating and generating a corrected feature vector; then using extended Kalman filtering to construct a low-temperature state transition matrix; recursively correcting the corrected feature vectors to obtain a corrected data matrix; and finally generating a distributed feature vector control matrix. This method is applied to energy balance control, solving the problem that distributed battery balance control technology in cold regions is prone to over-response due to increased differences in battery state, and that abnormal state change rates lead to insufficient control synchronization.

Description

Technical Field

[0001] This invention relates to the field of distributed control optimization technology, and more specifically, to a method for energy balance control of self-powered power supply of pole-mounted vacuum circuit breakers. Background Technology

[0002] In cold-region power distribution networks, pole-mounted vacuum circuit breakers typically rely on self-powered power supplies to provide continuous power to their control, communication, and protection units. These self-powered power supplies are often composed of battery packs consisting of multiple individual cells. At low temperatures, the chemical activity of the batteries decreases significantly, internal resistance increases, and the consistency between individual cells deteriorates, leading to uneven energy distribution during charging and discharging. Existing technologies typically use battery management systems to monitor the battery pack's state and perform basic balancing control, combined with temperature monitoring, load forecasting, and temperature-controlled heating, to comprehensively regulate the battery's operating environment and maintain the basic power supply stability of the pole-mounted vacuum circuit breaker under extremely cold conditions.

[0003] Although existing battery management systems can collect operating parameters such as voltage, current, and temperature in real time and dynamically monitor battery status, in cold regions, drastic temperature drops can significantly alter the internal resistance characteristics and capacity output of individual battery cells, further amplifying performance differences between different cells. Current energy balancing methods often adjust parameters based on models under normal temperature conditions, making it difficult to reflect state shifts caused by low temperatures in a timely manner. This leads to a discrepancy between the energy distribution strategy and the actual battery state, potentially causing overcharging or over-discharging of some individual cells and reducing the overall battery pack balancing effect.

[0004] To address the aforementioned issues, distributed battery balancing control technology has been proposed in other technological fields. For example, in electric vehicles and smart grid energy storage systems, by dividing the battery pack into multiple independent control units, each battery cell can independently adjust based on its own state, thereby reducing the lag and error accumulation caused by centralized control. In this type of technology, each battery cell can autonomously adjust its charging and discharging behavior according to its own voltage and state of charge, which improves the system's balancing capability under complex operating conditions to a certain extent. Therefore, theoretically, it can alleviate the problem of slow response of traditional centralized battery management systems in cold environments.

[0005] However, the aforementioned distributed battery equalization control technology still faces adaptation challenges when applied to low-temperature environments. On one hand, the difference between the battery's internal resistance and terminal voltage widens further at low temperatures, making distributed control prone to over-responding to local state changes during independent adjustment. This leads to frequent fluctuations in control commands and increases system complexity. On the other hand, battery charging and discharging efficiency decreases significantly in low-temperature environments, and the rate of battery state change differs markedly from that under normal operating conditions. This makes it difficult for traditional distributed equalization strategies to adjust synchronously and in a timely manner, potentially causing some individual battery cells to overcharge or over-discharge during local control, thus affecting the overall operational stability of the battery pack.

[0006] Therefore, existing energy balancing control methods for pole-mounted vacuum circuit breakers in cold regions still suffer from insufficient adaptability to changes in the nonlinear performance of batteries under low-temperature conditions, difficulty in achieving precise energy distribution when the state differences between individual battery cells increase, and problems of control strategy response lag and over-adjustment coexisting under dynamic temperature change conditions, which in turn affect the balancing effect of the battery pack and the long-term power supply reliability.

[0007] To address the above problems, this invention proposes a solution. Summary of the Invention

[0008] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a power balance control method for a pole-mounted vacuum circuit breaker. This method utilizes a distributed battery balance control method based on dynamic temperature correction and particle swarm optimization in low-temperature environments to address the problems of excessive response caused by widening differences in battery state in distributed battery balance control technology in cold regions, and insufficient control synchronization due to abnormal state change rates.

[0009] To achieve the above objectives, the present invention provides the following technical solution: A method for energy balancing control of a pole-mounted vacuum circuit breaker self-powered power supply includes: constructing an initial feature vector for each energy storage node in the pole-mounted vacuum circuit breaker self-powered system, and compensating the initial feature vector based on a temperature compensation algorithm to generate a feature compensation vector; using the feature compensation vector, analyzing the adjustment amplitude and direction identifier of each energy storage node during the energy distribution process through a particle swarm optimization algorithm, and generating a first distributed control vector set in vector form; real-time acquisition of the target ambient temperature and the internal temperature of the energy storage node, constructing a temperature change gradient matrix, and performing matrix transformation and adaptive correction on the first distributed control vector set to obtain a second distributed control vector set; and based on... In the distributed numerical recursive model, the second distributed control vector set is distributed in parallel to each energy storage node. Each energy storage node independently performs feature vector recursive update and collects the corrected feature vector of each energy storage node after the recursive update. The corrected feature vector also includes an optional low-temperature state transition matrix constructed based on extended Kalman filtering under low-temperature conditions, and the corrected feature vectors of all energy storage nodes are recursively corrected to obtain the feature vector correction data matrix. A distributed feature vector control matrix is ​​generated based on the feature vector correction data matrix and the second distributed control vector set, and the distributed feature vector control matrix is ​​applied to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker.

[0010] This invention also includes a system for energy balance control of a pole-mounted vacuum circuit breaker self-powered power supply, comprising a feature vector compensation module, a control vector set generation module, a control vector set correction module, a distributed recursive update module, a Kalman filter correction module, and a feature control matrix generation module. The feature vector compensation module constructs an initial feature vector for each energy storage node within the pole-mounted vacuum circuit breaker self-powered system and compensates the initial feature vector based on a temperature compensation algorithm to generate a feature compensation vector. The control vector set generation module uses the feature compensation vector to analyze the adjustment amplitude and direction identifier of each energy storage node during energy distribution using a particle swarm optimization algorithm, and generates a first distributed control vector set in vector form. The control vector set correction module collects the target ambient temperature and the internal temperature of the energy storage node in real time, constructs a temperature change gradient matrix, and corrects the first distributed control vector set. The distributed control vector set undergoes matrix transformation and adaptive correction to obtain a second distributed control vector set. A distributed recursive update module distributes the second distributed control vector set to each energy storage node in parallel based on a distributed numerical recursive model. Each energy storage node independently performs eigenvector recursive updates and collects the corrected eigenvectors of each energy storage node after the recursive update. A Kalman filter correction module is used to construct a low-temperature state transition matrix under low-temperature conditions based on an optional extended Kalman filter, and recursively corrects the corrected eigenvectors of all energy storage nodes to obtain an eigenvector correction data matrix. A eigenvector control matrix generation module generates a distributed eigenvector control matrix based on the eigenvector correction data matrix and the second distributed control vector set, and applies the distributed eigenvector control matrix to the energy balance control of the pole-mounted vacuum circuit breaker's self-powered power supply.

[0011] The present invention also includes an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a pole-mounted vacuum circuit breaker self-powered power supply energy balancing control method.

[0012] The present invention also includes a computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for energy balance control of a self-powered power supply for a pole-mounted vacuum circuit breaker.

[0013] The technical effects and advantages of the energy balance control method for self-powered pole-mounted vacuum circuit breakers of the present invention are as follows: 1. This invention, by analogy with distributed battery balancing control technology, enables individual battery cells to independently adjust according to their own state, thereby ensuring the balance of all batteries within the battery pack. Especially in cold regions, where low temperatures significantly impact battery performance, traditional centralized battery management methods often fail to effectively address the problems caused by performance differences. The distributed control method dynamically adjusts the charging and discharging current of each battery cell, compensating for performance differences based on real-time temperature changes. This avoids overcharging or over-discharging caused by temperature differences at low temperatures, ensuring the stable operation of the entire battery pack and thus improving system reliability and efficiency.

[0014] 2. This invention employs the Extended Kalman Filter (EKF) algorithm to estimate the charge / discharge state of individual battery cells in real time under low-temperature conditions, including key parameters such as voltage, current, and temperature, as well as state variables such as internal resistance and chemical reaction rate. Through real-time state estimation of battery performance, EKF can promptly adjust for performance fluctuations caused by low temperatures, dynamically optimizing the battery equalization strategy. Compared to traditional methods, EKF significantly improves the accuracy and response speed of the battery management system under low-temperature conditions, reduces equalization errors caused by temperature changes, and ensures that the battery pack can still operate efficiently and safely in extremely cold environments. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the energy balance control method for a self-powered power supply of a pole-mounted vacuum circuit breaker according to the present invention. Detailed Implementation

[0016] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0017] Example 1, Figure 1 The present invention provides an energy balance control method for a self-powered power supply of a pole-mounted vacuum circuit breaker, comprising the following steps: S1. Construct the initial feature vector of each energy storage node in the self-powered system of the pole-mounted vacuum circuit breaker, and compensate the initial feature vector based on the temperature compensation algorithm to generate the feature compensation vector.

[0018] In this embodiment, the initial feature vector of each energy storage node in the self-powered system of the pole-mounted vacuum circuit breaker is constructed, and the initial feature vector is compensated based on the temperature compensation algorithm to generate a feature compensation vector. Specifically, the following steps are taken: the terminal voltage sequence, current sequence, and internal temperature sequence of each energy storage node are collected within a preset sampling period and aligned by timestamps to form an initial feature vector; the initial feature vector is segmented into sliding windows, and a window feature vector group of average voltage, average current, and average temperature is constructed in each window; based on the average temperature in each window feature vector group, the average voltage in each time window is corrected for temperature offset, and the average current is corrected for multiplier attenuation, generating a compensated two-dimensional feature sub-vector; the equivalent internal resistance value in each window is analyzed based on the compensated two-dimensional feature sub-vector, and the equivalent internal resistance value is jointly mapped with the average temperature to generate an internal resistance-temperature coupling feature value; the compensated two-dimensional feature sub-vector and the internal resistance-temperature coupling feature value are concatenated to form a feature compensation sub-vector under each window; the feature compensation sub-vectors of all windows are stacked in chronological order to generate a feature compensation vector.

[0019] In this embodiment, the preset sampling period is a fixed time interval artificially set according to the load change frequency of the self-powered power supply of the pole-mounted vacuum circuit breaker and the response hysteresis characteristics of the battery in low temperature environment. This is used to limit the acquisition rhythm of voltage, current and temperature data, so that the acquired data can cover the slow change process at low temperature, and avoid data redundancy and fluctuation amplification caused by excessive sampling.

[0020] In this embodiment, the terminal voltage sequence is a set of values ​​formed by the change of the voltage at both ends of a battery cell over time, which is continuously collected within a preset sampling period. This sequence can reflect the attenuation trend of the battery's output capability under low temperature conditions, such as the change process of voltage continuously declining or recovering hysterically under the same load conditions.

[0021] In this embodiment, the current sequence is a set of charging and discharging current changes collected at the corresponding timestamp. This sequence is used to reflect the output or absorption capacity of the battery at different working stages and can reflect the current response delay or amplitude reduction caused by the slowdown of electrochemical reaction at low temperatures.

[0022] In this embodiment, the internal temperature sequence is a time series data characterizing the actual thermal state inside the battery. This temperature is formed by the combined effect of the internal reaction heat of the battery and the external low temperature environment, and can more accurately reflect the working conditions of the battery than the ambient temperature.

[0023] In this embodiment, timestamp alignment involves synchronizing the three types of sequences—voltage, current, and temperature—according to a unified time reference, ensuring that data at the same time point correspond one-to-one, avoiding misjudgments of the state due to sampling time deviations, and thus guaranteeing consistency among physical quantities in subsequent analysis processes.

[0024] In this embodiment, the sliding window segmentation process divides the continuous time series into multiple time windows with a fixed time length, and generates a continuous window set by sliding with a set step size. This method can extract local stable features in the context of slow change at low temperature, while preserving the overall change trend.

[0025] In this embodiment, the window state feature group is a combination of the average voltage, average current and average temperature calculated in each time window. This combination is used to characterize the average operating state of the battery during that time period, such as the case where the voltage is generally low but fluctuates little at low temperatures.

[0026] In this embodiment, temperature offset correction is a process of compensating and adjusting the average voltage value based on the difference between the average temperature within the window and the reference temperature. Essentially, it eliminates the effect of low temperature on the voltage measurement value, making the voltage characteristics closer to the performance of the battery under standard temperature conditions.

[0027] In this embodiment, the rate decay correction is a process of adjusting the average current value based on the average temperature value. This is used to reflect the actual situation of the battery's charge and discharge capacity decreasing at low temperatures. For example, the current output capacity decreases under the same control conditions at low temperatures. This correction yields an effective current characteristic that is more in line with reality.

[0028] In this embodiment, the equivalent internal resistance value is a parameter obtained based on the correspondence between the corrected voltage characteristic value and the current characteristic value. It is used to characterize the internal impedance level of the battery. This value can reflect the degree of resistance to electrochemical reaction under low temperature conditions and is an important basis for distinguishing the performance differences of battery cells.

[0029] In this embodiment, the internal resistance-temperature coupling characteristic value is a composite feature obtained by mapping the equivalent internal resistance value with the corresponding average temperature value. It is used to describe the degree of influence of temperature change on internal resistance, such as the magnitude of internal resistance increase when the temperature decreases.

[0030] In this embodiment, the feature compensation sub-vector is a multi-dimensional data set formed by combining the temperature-corrected voltage feature value, current feature value, and internal resistance-temperature coupling feature value in a fixed order. This vector can comprehensively reflect the actual working state of the battery under low temperature conditions.

[0031] In this embodiment, the feature compensation vector is a continuous sequence formed by splicing the feature compensation sub-vectors under each time window in chronological order. This sequence contains both the state information at each moment and the trend information of the state changing over time.

[0032] It should be noted that the above processing procedure decomposes the effects of low temperature into multiple independently manageable factors by applying temperature-related corrections to voltage, current, and internal resistance. This avoids over-response to temperature changes when controlling directly based on the original data, making the subsequent adjustment process more stable.

[0033] It should be noted that by constructing the coupling characteristics between internal resistance and temperature, the battery state not only reflects instantaneous values, but also the influence of temperature on performance, thus providing a more physically meaningful data basis for the subsequent state correction process and improving the stability of subsequent processing.

[0034] It should be noted that the beneficial effect of this technology is that by performing segmented analysis, temperature correction, and multi-feature fusion on battery state data under low-temperature conditions, the interference of low temperature on voltage and current measurements can be effectively eliminated, allowing the state of different battery cells to be compared under the same evaluation benchmark. This reduces the risk of misjudgment caused by temperature differences and improves the accuracy of subsequent energy regulation and equalization control.

[0035] The technical solution of this invention transforms the distorted original feature data under low temperature conditions into feature data with consistency and comparability by constructing a feature compensation vector, so that the subsequent energy regulation process is based on a unified state and provides a stable input for subsequent state correction, thereby improving the reliability and adaptability of the overall control process.

[0036] S2. Using the feature compensation vector, the particle swarm optimization algorithm is used to analyze the adjustment amplitude and direction identifier of each energy storage node in the energy distribution process, and the first distributed control vector set is generated in vector form.

[0037] In this embodiment, the step of using feature compensation vectors to analyze the adjustment amplitude and direction identifier of each energy storage node in the energy allocation process through particle swarm optimization algorithm, and generating a first distributed control vector set in vector form, specifically involves: splitting the feature compensation vector of each energy storage node into a candidate feature vector set by window, and using each candidate feature vector in the candidate feature vector set as an initial position vector of a particle; assigning a combination of adjustment amplitude and direction identifier to the initial position vector of each particle, and analyzing the corresponding feature change deviation, fluctuation, and internal resistance change penalty based on the feature compensation vector; performing a weighted summation of the feature change deviation, fluctuation, and internal resistance change penalty to obtain the particle fitness value of each particle; sorting all particles according to the particle fitness value of each particle to determine the individual optimal adjustment vector and the global optimal adjustment vector of each energy storage node; iteratively updating the velocity vector and position vector of each particle based on the vector difference between the individual optimal adjustment vector and the global optimal adjustment vector until a preset number of iterations is reached to obtain the target adjustment vector of each energy storage node; and arranging the target adjustment vectors of all energy storage nodes into the first distributed control vector set.

[0038] In this embodiment, the candidate feature vector set is a set of multiple sub-sequences formed by splitting the feature compensation vector of each battery cell according to a time window. Each sub-sequence corresponds to a continuous state change process within a time window, which is used to describe the local operating characteristics of the battery in a low-temperature environment during that time period, thereby avoiding the loss of local differences due to the averaging of features caused by directly using the whole data.

[0039] In this embodiment, the initial position vector of the particle is a multi-dimensional numerical vector formed by combining the voltage feature value, current feature value and internal resistance-temperature coupling feature value in the candidate state sequence in a fixed order. This vector serves as the search starting point in the particle swarm optimization algorithm and represents the initial operating state of the current battery in a certain time window.

[0040] In this embodiment, the combination parameter of the adjustment amplitude and direction identifier is a set of adjustment amplitude and direction identifiers randomly or according to rules assigned to each particle. This combination is used to simulate the adjustment strategy that may be taken under the current battery state and serves as the basis for subsequent evaluation of its merits.

[0041] In this embodiment, the characteristic change deviation is the degree of difference in characteristic changes before and after the strategy is applied, after applying a combination of parameters of a certain set of adjustment amplitude and direction indicator bits. For example, the degree of difference between the voltage change trend and the expected stable voltage is used to reflect whether the strategy will cause abnormal voltage fluctuations or excessive deviation.

[0042] In this embodiment, the fluctuation amount is an index obtained within the same time window based on the dispersion of the current change amplitude after the adjustment strategy is applied, for example, to measure whether the current is stable during charging and discharging, so as to avoid further deterioration of battery performance due to drastic current fluctuations in low temperature environments.

[0043] In this embodiment, the internal resistance change penalty is a constraint index set based on the change range of the equivalent internal resistance during the adjustment process. When a certain adjustment strategy causes a significant increase in internal resistance, the corresponding penalty value increases to limit the selection of strategies that have an adverse effect on the internal structure of the battery.

[0044] In this embodiment, the particle fitness value is the result of comprehensively evaluating the characteristic change deviation, fluctuation amount and internal resistance change penalty amount according to a preset weight. This result is used to characterize the adjustment quality of the first distributed control vector corresponding to the current particle. The better the value, the more suitable the adjustment strategy is for the battery state under the current low temperature environment.

[0045] In this embodiment, the individual optimal adjustment vector is the combination of the adjustment amplitude and direction identifier corresponding to the best fitness value obtained by each particle in history during the particle iteration process, which is used to guide the particle's subsequent search direction.

[0046] In this embodiment, the global optimal adjustment vector is a combination of the adjustment magnitude and direction identifier that is currently the best fit among all particles. It is used to guide the entire particle swarm to converge toward a better solution region, thereby improving the overall optimization efficiency.

[0047] In this embodiment, the velocity vector is a parameter describing the direction and magnitude of particle movement in the search space. Its update process reflects the adjustment trend of the particle between individual optimality and global optimality, and is used to control the search step size and direction.

[0048] In this embodiment, the position vector is a combination of the adjustment amplitude and direction identifier corresponding to the particle. By continuously updating the position vector, the particle gradually approaches the optimal adjustment strategy.

[0049] In this embodiment, the preset number of iterations is the maximum number of loops set according to the computational complexity and control response time requirements. This is used to limit the running time of the optimization process and ensure that control results can be output in a timely manner in practical applications.

[0050] In this embodiment, the target adjustment vector is the optimal combination of adjustment amplitude and direction identifier determined after particle iteration. This combination can balance voltage stability, current smoothing and internal resistance control under the current low temperature conditions.

[0051] In this embodiment, the first distributed control vector set is an overall control scheme formed by uniformly arranging the target adjustment vectors corresponding to all battery cells according to the battery number or spatial location, which is used to guide the subsequent battery equalization adjustment process.

[0052] It should be noted that in the above particle construction process, the charging and discharging current is not directly searched blindly. Instead, by introducing three dimensions of constraint indicators—voltage, current, and internal resistance—the optimization process is always focused on the key factors of battery performance degradation under low temperature conditions, thereby improving the actual usability of the search results.

[0053] It should be noted that by transforming the state characteristics within the time window into the initial position of the particles, the search starting point of each particle has a clear physical meaning, avoiding the instability problem caused by random initialization in traditional optimization algorithms, and also enhancing the comparability of the optimization process between different battery cells.

[0054] It should be noted that the beneficial effect of this technology is that by constructing a multi-constraint optimization mechanism with voltage stability, current smoothness and internal resistance change as the core, the charging and discharging strategy generated in the low temperature environment no longer depends on a single index. This avoids the problem of over-adjustment caused by temperature differences, improves the consistency and controllability of the adjustment strategy of each battery cell, and provides a more reasonable control input for subsequent state correction.

[0055] The technical solution of this invention introduces a particle swarm optimization algorithm to transform the feature compensation vector into a searchable adjustment parameter space, thereby changing the distributed control vector set from empirical setting to a state-driven adaptive generation method. This enables the generation of differentiated but controlled adjustment strategies for different battery cells in low-temperature environments, reducing system complexity and improving the overall balancing effect, while providing a stable input basis for subsequent state correction processes.

[0056] S3. Real-time acquisition of target ambient temperature and internal temperature of energy storage nodes, construction of temperature change gradient matrix, and matrix transformation and adaptive correction of the first distributed control vector set to obtain the second distributed control vector set.

[0057] In this embodiment, the real-time acquisition of the target ambient temperature and the internal temperature of the energy storage node, the construction of a temperature change gradient matrix, and the matrix transformation and adaptive correction of the first distributed control vector set to obtain the second distributed control vector set are specifically as follows: the real-time acquired target ambient temperature and the internal temperature of the energy storage node are formed into a temperature difference sequence at preset time intervals; the temperature difference sequence is segmented into sliding windows to construct a temperature change gradient matrix; the temperature change gradient matrix is ​​aligned with the first distributed control vector set on the time axis to construct a temperature-regulation mapping relationship; the temperature change gradient value within each window is analyzed using the temperature change gradient matrix and the temperature-regulation mapping relationship to obtain a temperature correction transformation matrix; the target regulation vector within each window of the first distributed control vector set is offset and corrected using the temperature correction transformation matrix to obtain a temperature-corrected regulation vector set; and the temperature-corrected regulation vector set is filtered and reorganized under the constraints of minimizing voltage fluctuation and minimizing energy loss to generate the second distributed control vector set.

[0058] In this embodiment, the target ambient temperature is the real-time temperature data of the external environment where the pole-mounted vacuum circuit breaker is located. This temperature is used to reflect the degree of influence of the extremely cold environment on the battery's external conditions, while the internal temperature of the energy storage node is the actual thermal state inside the battery cell. The difference between the two can reflect the response hysteresis characteristics of the battery after being affected by the external low temperature.

[0059] In this embodiment, the temperature difference sequence is a time series data formed by subtracting the ambient temperature from the internal temperature of the energy storage node at the same time point. This sequence is used to characterize the thermal deviation between the battery's internal and external environments. For example, when there is a sudden drop in low temperature, the external temperature drops rapidly while the internal temperature drops more slowly, thus forming a large temperature difference value.

[0060] In this embodiment, the preset time interval is the time sampling interval set according to the rate of change of ambient temperature and the thermal conductivity of the battery. This is used to control the update frequency of the temperature difference sequence so that it can reflect the temperature change process and avoid data jitter caused by excessive sampling.

[0061] In this embodiment, the temperature change gradient matrix is ​​a set of multiple subsequences obtained by segmenting the temperature difference sequence through a sliding window. Each subsequence corresponds to the temperature change process within a time window and is used to extract the local temperature change trend.

[0062] In this embodiment, time axis alignment involves synchronously mapping the temperature change gradient matrix and the first distributed control vector set according to a unified time window. This ensures that the temperature change information within each time window corresponds one-to-one with the combination of the corresponding adjustment amplitude and direction identifier, thereby guaranteeing time consistency in the subsequent correction process.

[0063] In this embodiment, the temperature-regulation mapping relationship is to establish a correspondence between temperature change characteristics and corresponding first distributed control vectors within each time window. This relationship is used to describe the regulation strategy adopted under specific temperature change conditions.

[0064] In this embodiment, the temperature change gradient value is a rate of change index calculated based on the temperature change gradient matrix within each time window. It is obtained by comparing the changing trend of the temperature difference between the beginning and end of the window and is used to reflect the speed and direction of temperature change. For example, rapid cooling corresponds to a larger negative gradient.

[0065] In this embodiment, offset correction is a process of increasing or decreasing the original target adjustment vector according to the temperature change gradient value. For example, when the temperature drops rapidly, the charging current is appropriately reduced to avoid battery overload, and the current amplitude is gradually restored when the temperature rises.

[0066] In this embodiment, the temperature-corrected adjustment vector set is a combination of new adjustment amplitude and direction identifier parameters obtained by correcting the original target adjustment vector through temperature gradient. This combination can more accurately adapt to the current temperature change conditions.

[0067] In this embodiment, voltage fluctuation minimization is an index obtained by evaluating the stability of voltage changes after applying a combination of a certain adjustment amplitude and direction indicator bit. This index reflects that the smaller the voltage fluctuation, the better the value, indicating that the voltage is more stable.

[0068] In this embodiment, energy loss is minimized as the degree of change in energy loss caused by changes in internal resistance and current after executing a certain combination of adjustment amplitude and direction flag. This value is used to measure whether the strategy introduces additional energy loss, for example, high current at low temperature may lead to higher loss.

[0069] In this embodiment, vector selection is a process of choosing the combination with the smallest voltage fluctuation and the smallest change in energy loss from all candidate schemes, thereby ensuring that the selected adjustment vector achieves a balance between stability and efficiency.

[0070] In this embodiment, the second distributed control vector set is a complete control sequence formed by splicing and recombining the selected adjustment vectors according to the time window order. This sequence can keep synchronized with real-time temperature changes, thereby realizing dynamic adjustment.

[0071] It should be noted that the above process is corrected by introducing a temperature change gradient matrix instead of a single temperature value, so that the adjustment vector can respond to the temperature change trend, rather than just making static adjustments to the current temperature state, thereby improving the response sensitivity in extremely cold environments.

[0072] It should be noted that by performing a secondary temperature-driven correction on the particle optimization results, the problem of the initial optimization results becoming invalid in scenarios with rapid temperature changes is avoided, enabling the strategy to have dynamic adaptive capabilities. At the same time, it is not necessary to re-execute the complete optimization process, thus reducing computational complexity.

[0073] It should be noted that the beneficial effect of this technology is that by introducing the temperature change process into the adjustment vector correction mechanism, the battery regulation can be adjusted synchronously with the temperature change under the condition of drastic temperature fluctuation in cold regions. This effectively suppresses voltage fluctuations caused by sudden temperature changes and reduces the additional energy loss caused by unreasonable current regulation, thereby improving the overall system stability and energy efficiency.

[0074] The technical solution of this invention establishes a dynamic mapping relationship between temperature changes and adjustment strategies, enabling the original adjustment strategies based on state optimization to further adapt to environmental changes and avoid adjustment lag or over-response problems in low-temperature sudden change scenarios, thereby ensuring the continuity and reliability of subsequent battery energy balance control in complex temperature environments.

[0075] S4. Based on the distributed numerical recursive model, the second distributed control vector set is distributed to each energy storage node in parallel. Each energy storage node independently performs feature vector recursive update and collects the corrected feature vector of each energy storage node after the recursive update.

[0076] In this embodiment, based on the distributed numerical recursive model, the second distributed control vector set is distributed in parallel to each energy storage node. Each energy storage node independently performs feature vector recursive updates and collects the corrected feature vectors of each energy storage node after the recursive update. Specifically, based on the distributed numerical recursive model, the second distributed control vector set is split into individual adjustment vector sequences according to the energy storage node number, and matched one-to-one with the feature compensation vectors within a time window to form a node-window control mapping matrix. Based on the node-window control mapping matrix, the second distributed control vector set is distributed in parallel to each energy storage node. Each energy storage node independently performs numerical recursive analysis locally to obtain the instantaneous feature change vector of each energy storage node in each window. The instantaneous feature change vectors are accumulated according to the time window to obtain the feature update vector sequence of the corresponding energy storage node. The feature update vector sequence is jointly analyzed with the internal resistance-temperature coupled eigenvalues ​​to generate a feature drift correction vector. The feature drift correction vector and the feature update vector sequence are superimposed to obtain the corrected feature node vector for each energy storage node. The corrected feature node vector is then uploaded to the central processing unit, which summarizes the corrected feature node vectors of all energy storage nodes in chronological order into a corrected feature vector.

[0077] In this embodiment, the distributed numerical recursive model is a parallel computing architecture that decomposes the global computing task into multiple independent computing nodes, and each node iteratively calculates the change in state at the next moment based solely on its current state and the received control instructions using a fixed mathematical recursive formula.

[0078] In this embodiment, the single-cell adjustment vector sequence is an independent current control sequence formed by splitting the second distributed control vector set according to the energy storage node number, such as the battery cell number. Each sequence corresponds to the charging current and discharging current change process of a battery cell within a continuous time window, which is used to realize differentiated adjustment of different battery cells.

[0079] In this embodiment, the node-window control mapping matrix is ​​a relational structure that establishes a one-to-one correspondence between the adjustment execution sequence of each battery cell and the time window in the feature compensation vector. This mapping relationship is used to ensure that each control current acts on the actual battery state under the corresponding time window, thereby avoiding control and state misalignment.

[0080] In this embodiment, the instantaneous feature change vector is obtained by breaking down the overall control command into a dedicated command for each node in each time window according to the pre-established "node-window" correspondence and sending it to each energy storage node at the same time. After receiving the command, each node independently analyzes the instantaneous state change that it should generate in each time window.

[0081] In this embodiment, the feature update vector sequence is a sequence formed by accumulating the instantaneous feature change vectors within each time window in chronological order. This sequence can reflect the feature change trajectory of the battery throughout the entire adjustment process, such as continuous charging leading to a gradual increase in energy or discharging leading to a decrease in energy.

[0082] In this embodiment, the feature drift correction vector is the correction result obtained by jointly analyzing the feature update vector sequence and the internal resistance-temperature coupled feature value. It is used to eliminate the feature shift phenomenon caused by low temperature and internal resistance change, such as the case where the power estimation is too low due to the increase of internal resistance at low temperature.

[0083] In this embodiment, the corrected feature node vector is a multi-dimensional state representation formed by fusing the feature drift correction vector and the feature update vector sequence. This vector contains both the corrected features and the original change trend, and is used to comprehensively reflect the true state of the battery.

[0084] In this embodiment, the corrected feature vector is a set of data formed by summarizing the corrected feature node vectors of all battery cells in chronological order. This data is used to describe the dynamic feature state changes of the entire battery pack during the adjustment process.

[0085] It should be noted that by introducing the internal resistance-temperature coupling feature to correct the feature update vector sequence, the energy estimation error caused by the change in internal resistance under low temperature environment can be effectively compensated, thereby improving the accuracy of the state data and providing a reliable basis for subsequent control.

[0086] It should be noted that the beneficial effect of this part of the technology is that by transforming the second distributed control vector set into a characteristic change process, the adjustment result of each battery cell can be quantitatively expressed, and the internal resistance effect under low temperature environment is corrected, thereby avoiding energy assessment deviation caused by temperature difference, improving the consistency of state comparison between batteries, and providing accurate input for subsequent energy balance control.

[0087] The technical solution of this invention realizes the mapping from control strategy to actual battery state changes, concretizes the adjustment result into characteristic change trajectory, and eliminates the influence of low temperature through state correction, so that the entire system can still accurately grasp the real state of each battery cell in complex environments, thereby supporting the fine execution of subsequent energy balance control.

[0088] S5. The corrected eigenvector also includes, optionally, a low-temperature state transition matrix under low-temperature conditions constructed based on extended Kalman filtering, and the corrected eigenvectors of all energy storage nodes are recursively corrected to obtain the eigenvector correction data matrix.

[0089] In this embodiment, the corrected feature vector further includes, optionally, constructing a low-temperature state transition matrix under low-temperature conditions based on extended Kalman filtering, and recursively correcting the corrected feature vectors of all energy storage nodes to obtain a feature vector correction data matrix. Specifically, the corrected feature vector sequence matrix is ​​divided into measurement feature vector sequences according to the energy storage node number and time window; the measurement feature vector sequences are paired with internal resistance-temperature coupling feature values ​​according to the time window order, and a numerical recursive relationship of extended Kalman filtering is constructed based on the pairing results. Simultaneously, the difference between the target ambient temperature and the internal temperature of the energy storage node is introduced into the temperature correction term of the numerical recursive relationship to generate a low-temperature state transition matrix; the measurement feature vector sequences are initialized as the initial state sequence of system feature variables according to the time window order, and... Under the cryogenic state transition matrix, window-by-window recursive propagation is performed to execute feature prediction analysis, obtaining the predicted feature vector sequence for each energy storage node. The predicted feature vector sequence and the measured feature vector sequence are then mapped point-by-point according to time windows to generate a feature residual vector sequence. The cumulative residual result of the feature correction update sequence of the previous time window and the propagation error diffusion of the cryogenic state transition matrix are obtained, and a prediction error covariance matrix is ​​constructed. An adaptive Kalman gain vector is generated based on the feature residual vector sequence and the prediction error covariance matrix. The predicted feature vector sequence is corrected and updated based on the adaptive Kalman gain vector to generate the feature correction update sequence for each energy storage node under each time window. The feature correction update sequences are concatenated and recombined in chronological order to form a feature vector correction data matrix.

[0090] In this embodiment, the measurement feature vector sequence is a segmented sequence formed by dividing the modified feature vector according to the battery cell number and time window. Each sequence corresponds to the state change process of a battery cell within a certain time window, which is used to characterize the local dynamic features of the battery state under low temperature environment.

[0091] In this embodiment, the internal resistance-temperature coupling characteristic value is a characteristic parameter obtained by analyzing the relationship between the battery's equivalent internal resistance and temperature. It is used to characterize the nonlinear increase of internal resistance as temperature decreases in a low-temperature environment, thereby reflecting the coupling relationship between the battery's internal impedance and thermal state.

[0092] In this embodiment, the numerical recursive relationship is a state evolution description relationship constructed based on the internal resistance-temperature coupling characteristic value, which is used to express the change law of battery state between adjacent time windows. This relationship is used to constrain the recursive direction and magnitude of the state under low temperature environment.

[0093] In this embodiment, the low-temperature state transition matrix is ​​a matrix structure formed by introducing a correction term for the difference between the target ambient temperature and the internal temperature based on the numerical recursive relationship. It is used to correct the lag effect in the state propagation process under low-temperature conditions, such as the state delay caused by the slower decrease of the internal temperature of the battery compared to the external environment.

[0094] In this embodiment, the initial state sequence of system feature variables is an initial state input sequence formed by arranging the measurement feature vector sequence in time window order. This sequence serves as the starting reference for the state recursion process and is used to initiate subsequent state propagation calculations.

[0095] In this embodiment, feature prediction analysis is a process of recursively calculating system feature variables over time windows under the constraint of the low-temperature state transition matrix, which is used to predict the state change trend of the battery in future time windows.

[0096] In this embodiment, the predicted feature vector sequence is the feature evolution result of each battery cell under different time windows obtained through feature prediction analysis, which is used to describe the theoretical state trajectory before the introduction of observation correction.

[0097] In this embodiment, the feature residual vector sequence is a sequence obtained by calculating the point-by-point difference between the predicted feature vector sequence and the measured feature vector sequence within the same time window. It is used to characterize the degree of deviation between the prediction result and the actual observation, such as the difference between the predicted state of charge and the actual correction value at low temperatures.

[0098] In this embodiment, the residual accumulation result is the amount of error accumulation formed by the accumulation of residuals over time in the feature correction update sequence of the previous time window, which is used to reflect the system deviation that was not completely eliminated during the historical correction process.

[0099] In this embodiment, the propagation error diffusion amount is the uncertainty expansion amount caused by the low temperature state transition matrix during the recursive process, which is used to characterize the degree of diffusion of state propagation error over time under low temperature conditions.

[0100] In this embodiment, the prediction error covariance matrix is ​​an error uncertainty description information constructed by the cumulative residual result and the propagation error diffusion amount, which is used to characterize the stability and diffusion degree of the error distribution during the state prediction process.

[0101] In this embodiment, the Kalman gain update is a weight adjustment calculated based on the feature residual vector sequence and the prediction error covariance matrix, which is used to balance the influence ratio of the predicted state and the observed state in the final correction.

[0102] In this embodiment, the adaptive Kalman gain vector is a coefficient dynamically adjusted by the Kalman gain update. This coefficient can automatically adjust the correction strength according to the error changes under low temperature conditions, thereby improving the stability of state estimation.

[0103] In this embodiment, the feature correction update sequence is a feature result sequence obtained by correcting the predicted feature vector sequence based on the adaptive Kalman gain coefficient, which is used to describe the change process of the battery's true state after observation correction.

[0104] In this embodiment, the feature vector correction data matrix is ​​a complete set of state data formed by splicing and recombining the feature correction update sequences of all battery cells in chronological order, which is used as the input basis for subsequent energy balance control.

[0105] It should be noted that by introducing the internal resistance-temperature coupling characteristic to constrain the state transition process, the state propagation no longer depends on a fixed linear relationship, but can dynamically adjust the direction of state evolution with temperature changes, thereby adapting to the characteristics of significant changes in battery performance in cold regions.

[0106] It should be noted that the beneficial effect of this part of the technology is that by constructing a state transition mechanism that includes temperature difference correction and error covariance information based on residual accumulation, the extended Kalman filter no longer relies on a single initial state estimate in low-temperature environments. This reduces the impact of inaccurate initial states on the filtering results, improves the convergence stability of battery state estimation, and enhances the ability to identify differences in individual battery cells in extremely cold environments, providing a more reliable state basis for distributed equilibrium control.

[0107] The technical solution of this invention introduces the modified feature vector, the compensation feature, and the temperature coupling relationship into the filtering recursive process, so that the state estimation is transformed from static observation correction to dynamic error propagation correction mechanism. This solves the problems of state drift accumulation and initial error sensitivity under low temperature conditions, and ensures that the battery state correction results can truly reflect the operating state of each cell in the extremely cold environment, providing a stable and consistent state input basis for subsequent energy balance control.

[0108] S6. Generate a distributed eigenvector control matrix based on the eigenvector correction data matrix and the second distributed control vector set, and apply the distributed eigenvector control matrix to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker.

[0109] In this embodiment, the step of generating a distributed feature vector control matrix based on the feature vector correction data matrix and the second distributed control vector set, and applying the distributed feature vector control matrix to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker, specifically involves: dividing the feature vector correction data matrix into a set of corrected feature vectors according to the energy storage node number and time window; aligning and matching the corrected feature vectors with the second distributed control vector set to form a node-window joint control matrix; analyzing the deviation vector between the actual feature value and the preset target feature value of each energy storage node under each window based on the node-window joint control matrix, and generating a feature deviation matrix; accumulating the feature deviation matrix according to the time window to obtain a cumulative feature offset vector; constructing a numerical balance correction matrix based on the cumulative feature offset vector; performing matrix fusion of the numerical balance correction matrix with the second distributed control vector set to generate a final execution adjustment matrix; normalizing and sorting the final execution adjustment matrix to generate a distributed feature vector control matrix, and applying the distributed feature vector control matrix to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker.

[0110] In this embodiment, the corrected feature vector set is a sequence set formed by dividing the feature vector correction data matrix according to the battery cell number and time window. Each sequence corresponds to the feature change process of a single battery cell after filtering and correction under each time window, which is used to reflect the actual operating state after correction in low temperature environment.

[0111] In this embodiment, the node-window joint control matrix is ​​a joint sequence structure formed by aligning the corrected feature vector with the charging and discharging current sequence of the corresponding time window in the second distributed control vector set. This structure is used to establish a one-to-one correspondence between feature information and control behavior.

[0112] In this embodiment, the actual characteristic value is the actual characteristic change generated by the charging and discharging current acting on the battery cell within the time window, which is used to characterize the actual energy level output or absorbed by the battery under the current control strategy.

[0113] In this embodiment, the preset target feature value is the expected energy change amount preset according to the second distributed control vector set. This value is used to represent the feature adjustment target that a single battery cell should achieve within the current time window, and is used to measure the effect of the control strategy execution.

[0114] In this embodiment, the feature deviation matrix is ​​a sequence formed by calculating the difference between the actual feature values ​​and the target energy values ​​within each time window, and is used to characterize the degree to which the control execution deviates from the target within each time window.

[0115] In this embodiment, the cumulative feature offset vector is the result of gradually accumulating the feature deviation matrix over a time window. It is used to reflect the cumulative trend of feature deviation of a single battery cell during continuous control, such as long-term bias charging or discharging.

[0116] In this embodiment, the numerical balancing correction matrix is ​​used to describe the degree of deviation that needs to be corrected in the overall feature allocation of a single battery cell, and is used to correct and compensate for subsequent charging and discharging currents.

[0117] In this embodiment, the final execution adjustment matrix is ​​an adjustment matrix formed by fusing and adjusting the numerical balancing correction matrix with the original second distributed control vector set. This sequence is used to actually guide the charging and discharging process of the battery cells.

[0118] In this embodiment, the distributed feature vector control matrix is ​​a matrix structure formed by uniformly scheduling and sorting the final execution adjustment matrices of all battery cells, and is used to describe the feature allocation relationship of the entire battery pack under each time window.

[0119] In this embodiment, the distributed feature vector control matrix is ​​used as a unified energy regulation command for the self-powered power supply of the pole-mounted vacuum circuit breaker in a low-temperature environment.

[0120] It should be noted that the above process aligns the modified feature information with the control strategy at the time window level, so that the battery state and control behavior form a strict correspondence, thereby avoiding the energy distribution deviation problem caused by state lag in traditional control and improving control synchronization in low temperature environments.

[0121] It should be noted that by introducing a cumulative feature offset vector to construct a correction coefficient, feature adjustment no longer depends on a single time window error, but is based on historical deviations for long-term correction, thereby effectively suppressing the problem of continuous bias accumulation in low-temperature environments.

[0122] It should be noted that the beneficial effect of this technology is that by fusing the corrected feature vector with the second distributed control vector set and introducing an energy deviation accumulation correction mechanism, the battery pack can dynamically compensate based on the long-term energy deviation trend in cold environments, thereby achieving energy redistribution optimization among individual battery cells, reducing the amplification effect of individual cell differences caused by low temperature, and improving the stability and continuous power supply capability of the overall self-powered system.

[0123] The technical solution of this invention establishes a node-window joint control matrix and a characteristic deviation closed-loop correction mechanism, which transforms battery control from "instantaneous adjustment" to "cumulative deviation-driven adjustment", thereby solving the energy imbalance problem caused by the accumulation of single adjustment errors in low-temperature environments, and ultimately achieving long-term stable energy balance output of the pole-mounted vacuum circuit breaker's self-powered power supply under extremely cold conditions.

[0124] Example 2: The present invention provides a self-powered energy balance control system for pole-mounted vacuum circuit breakers, comprising a feature vector compensation module, a control vector set generation module, a control vector set correction module, a distributed recursive update module, a Kalman filter correction module, and a feature control matrix generation module. The feature vector compensation module is used to construct the initial feature vector of each energy storage node in the self-powered system of the pole-mounted vacuum circuit breaker, and to compensate the initial feature vector based on the temperature compensation algorithm to generate the feature compensation vector. The control vector set generation module is used to analyze the adjustment amplitude and direction identifier of each energy storage node in the energy distribution process using the feature compensation vector and the particle swarm optimization algorithm, and generate the first distributed control vector set in vector form. The control vector set correction module is used to collect the target ambient temperature and the internal temperature of the energy storage node in real time, construct a temperature change gradient matrix, and perform matrix transformation and adaptive correction on the first distributed control vector set to obtain the second distributed control vector set. The distributed recursive update module is used to distribute the second distributed control vector set to each energy storage node in parallel based on a distributed numerical recursive model. Each energy storage node independently performs feature vector recursive update and collects the corrected feature vector of each energy storage node after the recursive update. The Kalman filter correction module is used to construct a low-temperature state transition matrix under low-temperature conditions based on an optional extended Kalman filter, and recursively correct the corrected feature vectors of all energy storage nodes to obtain a feature vector correction data matrix. The feature control matrix generation module is used to generate a distributed feature vector control matrix based on the feature vector correction data matrix and the second distributed control vector set, and apply the distributed feature vector control matrix to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker.

[0125] The present invention also includes an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform a pole-mounted vacuum circuit breaker self-powered power supply energy balancing control method.

[0126] The present invention also includes a computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for energy balance control of a self-powered power supply for a pole-mounted vacuum circuit breaker.

[0127] In the embodiments provided by this invention, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0128] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0129] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0130] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0131] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0132] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0133] In the embodiments provided in this disclosure, it should be understood that the disclosed systems and methods can also be implemented in other ways. The system embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0134] It should be noted that, in this disclosure, 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 a process, method, article, or apparatus. Without further limitation, an element limited by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0135] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for energy balance control of a self-powered power supply in a pole-mounted vacuum circuit breaker, characterized in that, Includes the following steps: The initial feature vector of each energy storage node in the self-powered system of the pole-mounted vacuum circuit breaker is constructed, and the initial feature vector is compensated based on the temperature compensation algorithm to generate the feature compensation vector. The characteristic compensation vector is used to analyze the adjustment amplitude and direction identifier of each energy storage node in the energy distribution process through particle swarm optimization algorithm, and the first distributed control vector set is generated in vector form. The target ambient temperature and the internal temperature of the energy storage node are collected in real time to construct a temperature change gradient matrix. The first distributed control vector set is then subjected to matrix transformation and adaptive correction to obtain the second distributed control vector set. Based on the distributed numerical recursive model, the second distributed control vector set is distributed to each energy storage node in parallel. Each energy storage node independently performs feature vector recursive update and collects the corrected feature vector of each energy storage node after the recursive update. The corrected eigenvector also includes, optionally, a low-temperature state transition matrix under low-temperature conditions constructed based on extended Kalman filtering, and the corrected eigenvectors of all energy storage nodes are recursively corrected to obtain the eigenvector corrected data matrix. A distributed eigenvector control matrix is ​​generated based on the eigenvector correction data matrix and the second distributed control vector set, and then applied to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker.

2. The energy balance control method for self-powered power supply of pole-mounted vacuum circuit breaker according to claim 1, characterized in that, The initial feature vector of each energy storage node in the self-powered system of the pole-mounted vacuum circuit breaker is constructed, and the initial feature vector is compensated based on the temperature compensation algorithm to generate a feature compensation vector, specifically: Within a preset sampling period, the terminal voltage sequence, current sequence, and internal temperature sequence of each energy storage node are collected and aligned according to the timestamp to form an initial feature vector; The initial feature vector is segmented into sliding windows, and a window feature vector group with mean voltage, mean current, and mean temperature is constructed within each window; Based on the average temperature in each window feature vector group, the average voltage in each time window is corrected for temperature offset, and the average current is corrected for rate attenuation, generating a compensated two-dimensional feature sub-vector. Based on the analysis of the equivalent internal resistance value in each window using the compensated two-dimensional feature vector, the equivalent internal resistance value is jointly mapped with the average temperature to generate internal resistance-temperature coupled feature value. The compensated two-dimensional feature vector is concatenated with the internal resistance-temperature coupling feature value to form the feature compensation vector under each window. Stack the feature compensation subvectors of all windows in chronological order to generate the feature compensation vector.

3. The energy balance control method for self-powered power supply of pole-mounted vacuum circuit breaker according to claim 2, characterized in that, The process involves using feature compensation vectors to analyze the adjustment amplitude and direction identifier of each energy storage node during energy allocation using a particle swarm optimization algorithm, and generating a first distributed control vector set in vector form, specifically: The feature compensation vector of each energy storage node is split into a set of candidate feature vectors by a window, and each candidate feature vector in the set of candidate feature vectors is used as the initial position vector of a particle. Each particle's initial position vector is assigned a combination of adjustment amplitude and direction identifier, and the corresponding characteristic change deviation, fluctuation, and internal resistance change penalty are analyzed based on the characteristic compensation vector. The particle fitness value of each particle is obtained by weighted summing of the characteristic change deviation, fluctuation amount, and internal resistance change penalty amount. All particles are sorted according to their particle fitness values ​​to determine the individual optimal adjustment vector and the global optimal adjustment vector for each energy storage node. Based on the vector difference between the individual optimal adjustment vector and the global optimal adjustment vector, the velocity vector and position vector of each particle are iteratively updated until the preset number of iterations is reached, so as to obtain the target adjustment vector of each energy storage node. Arrange the target adjustment vectors of all energy storage nodes into the first distributed control vector set.

4. The energy balance control method for self-powered power supply of pole-mounted vacuum circuit breaker according to claim 3, characterized in that, The system collects the target ambient temperature and the internal temperature of the energy storage node in real time, constructs a temperature change gradient matrix, and performs matrix transformation and adaptive correction on the first distributed control vector set to obtain the second distributed control vector set, specifically as follows: The real-time collected target ambient temperature and the internal temperature of the energy storage node are used to form a temperature difference sequence at preset time intervals. The temperature difference sequence is then segmented by a sliding window to construct a temperature change gradient matrix. Align the temperature change gradient matrix with the first distributed control vector set along the time axis to construct a temperature-regulation mapping relationship. By analyzing the temperature change gradient matrix and the temperature-regulation mapping relationship within each window, the temperature change gradient value is obtained, and the temperature correction transformation matrix is ​​derived. The target adjustment vector in each window of the first distributed control vector set is offset and corrected using the temperature correction transformation matrix to obtain the temperature-corrected adjustment vector set. With the constraints of minimizing voltage fluctuations and minimizing energy loss, the temperature-corrected adjustment vector set is filtered and reorganized to generate a second distributed control vector set.

5. The energy balance control method for self-powered power supply of pole-mounted vacuum circuit breaker according to claim 4, characterized in that, The distributed numerical recursive model involves distributing the second distributed control vector set to each energy storage node in parallel. Each energy storage node independently performs feature vector recursive updates and collects the corrected feature vectors of each energy storage node after the recursive update. Specifically: Based on the distributed numerical recursive model, the second distributed control vector set is split into individual adjustment vector sequences according to the energy storage node number, and matched with the feature compensation vector one by one in time window to form a node-window control mapping matrix. Based on the node-window control mapping matrix, the second distributed control vector set is distributed in parallel to each energy storage node. Each energy storage node independently performs numerical recursive analysis locally to obtain the instantaneous characteristic change vector of each energy storage node in each window. Accumulate the instantaneous feature change vectors according to the time window to obtain the feature update vector sequence of the corresponding energy storage node; The feature update vector sequence is jointly analyzed with the internal resistance-temperature coupled eigenvalues ​​to generate a feature drift correction vector. The feature drift correction vector and the feature update vector sequence are superimposed to obtain the corrected feature node vector for each energy storage node. The corrected feature node vectors are uploaded to the central processing unit, which then aggregates the corrected feature node vectors of all energy storage nodes in chronological order into a corrected feature vector.

6. The energy balance control method for self-powered power supply of pole-mounted vacuum circuit breaker according to claim 5, characterized in that, The corrected eigenvector also includes, optionally, a cryogenic state transition matrix under cryogenic conditions constructed based on extended Kalman filtering, and recursively corrects the corrected eigenvectors of all energy storage nodes to obtain an eigenvector corrected data matrix, specifically: The modified eigenvector sequence matrix is ​​divided into measurement eigenvector sequences according to the energy storage node number and time window; The measured feature vector sequence is paired with the internal resistance-temperature coupling feature value in the order of time window, and the numerical recursive relationship of extended Kalman filter is constructed based on the pairing result. At the same time, the difference between the target ambient temperature and the internal temperature of the energy storage node is introduced into the temperature correction term of the numerical recursive relationship to generate a low temperature state transition matrix. The measured feature vector sequence is initialized as the initial state sequence of system feature variables in the order of time window, and then recursively propagated window by window under the low temperature state transition matrix to perform feature prediction analysis and obtain the predicted feature vector sequence of each energy storage node. The predicted feature vector sequence and the measured feature vector sequence are mapped point-by-point according to a time window to generate a feature residual vector sequence. Obtain the residual accumulation result of the feature correction update sequence of the previous time window and the propagation error diffusion amount of the low temperature state transition matrix, and construct the prediction error covariance matrix; Generate an adaptive Kalman gain vector based on the feature residual vector sequence and the prediction error covariance matrix; The predicted feature vector sequence is corrected and updated based on the adaptive Kalman gain vector, generating the feature correction and update sequence for each energy storage node in each time window. The feature correction update sequence is concatenated and recombined in chronological order to form a feature vector correction data matrix.

7. The energy balance control method for self-powered power supply of pole-mounted vacuum circuit breaker according to claim 6, characterized in that, The process of generating a distributed eigenvector control matrix based on the eigenvector correction data matrix and the second distributed control vector set, and then applying the distributed eigenvector control matrix to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker, specifically involves: The eigenvector correction data matrix is ​​divided into a set of corrected eigenvectors according to the energy storage node number and time window. The corrected feature vectors are aligned and matched with the second distributed control vector set to form a node-window joint control matrix; Based on the node-window joint control matrix analysis, the deviation vector between the actual characteristic value of each energy storage node and the preset target characteristic value under each window is analyzed to generate a characteristic deviation matrix; The feature deviation matrix is ​​accumulated over a time window to obtain the accumulated feature offset vector; A numerical balance correction matrix is ​​constructed based on the cumulative feature offset vector. The numerical balance correction matrix is ​​then fused with the second distributed control vector set to generate the final execution adjustment matrix. The final execution adjustment matrix is ​​vector normalized and sorted to generate a distributed eigenvector control matrix, which is then applied to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker.

8. A system using the energy balancing control method for a self-powered power supply of a pole-mounted vacuum circuit breaker as described in any one of claims 1-7, characterized in that, It includes an eigenvector compensation module, a control vector set generation module, a control vector set correction module, a distributed recursive update module, a Kalman filter correction module, and an eigencontrol matrix generation module; The feature vector compensation module is used to construct the initial feature vector of each energy storage node in the self-powered system of the pole-mounted vacuum circuit breaker, and to compensate the initial feature vector based on the temperature compensation algorithm to generate the feature compensation vector. The control vector set generation module is used to analyze the adjustment amplitude and direction identifier of each energy storage node in the energy distribution process using the feature compensation vector and the particle swarm optimization algorithm, and generate the first distributed control vector set in vector form. The control vector set correction module is used to collect the target ambient temperature and the internal temperature of the energy storage node in real time, construct the temperature change gradient matrix, and perform matrix transformation and adaptive correction on the first distributed control vector set to obtain the second distributed control vector set. The distributed recursive update module is used to distribute the second distributed control vector set to each energy storage node in parallel based on the distributed numerical recursive model. Each energy storage node independently performs feature vector recursive update and collects the corrected feature vector of each energy storage node after the recursive update. The Kalman filter correction module, used for correcting the feature vector, also includes, optionally, constructing a low-temperature state transition matrix under low-temperature conditions based on extended Kalman filter, and recursively correcting the corrected feature vectors of all energy storage nodes to obtain a feature vector correction data matrix. The feature control matrix generation module is used to generate a distributed feature vector control matrix based on the feature vector correction data matrix and the second distributed control vector set, and to apply the distributed feature vector control matrix to the energy balance control of the self-powered power supply of the pole-mounted vacuum circuit breaker.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; And, a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the energy balancing control method for the self-powered power supply of the pole-mounted vacuum circuit breaker as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the energy balance control method for the self-powered power supply of the pole-mounted vacuum circuit breaker as described in any one of claims 1 to 7.

Images