Self-healing control and energy efficiency optimization method and system for box-type substation

Abstract

The invention belongs to the technical field of electric power automation, and particularly relates to a self-healing control and energy efficiency optimization method and system for a box-type substation, and the method comprises the steps: collecting the three-phase current, voltage waveform and temperature parameters of equipment in the box-type substation, constructing a space-time correlation feature matrix, and extracting a dynamic operation mode; constructing an impedance parameter nonlinear propagation model, and calculating a node credibility propagation value; establishing a multi-objective optimization model for dynamic weight optimization, and solving a topology reconstruction strategy under an asymmetric constraint condition; on the basis of a topology reconstruction strategy, voltage in the energy storage scheduling parameters is optimized, and an inertia weight self-adaptive control instruction is generated in combination with an optical fiber self-healing ring network communication channel; and correcting the dynamic weight and the voltage in the energy storage scheduling parameters in real time through a closed-loop feedback mechanism, and completing dynamic performance compensation and energy efficiency optimization iteration. By monitoring and optimizing the operation parameters of the box-type substation in real time, rapid and accurate fault diagnosis, self-healing control and dynamic energy efficiency optimization are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power automation, and in particular relates to a method and system for self-healing control and energy efficiency optimization of a box-type substation. Background Art

[0002] With the continuous expansion of the scale of power systems and the increasing requirements of users for power supply reliability and energy efficiency, the stable operation and high efficiency of box-type substations, as a key link in power distribution, are becoming increasingly important. Box-type substations are widely used in many scenarios such as urban power grids, industrial parks, and residential communities, and undertake the important task of converting high-voltage electricity into low-voltage electricity and distributing it.

[0003] Most of the existing box-type substation monitoring and control systems monitor and manage the equipment operating status through regular inspections, manual troubleshooting, and simple fault alarm mechanisms. Some relatively advanced systems use some basic sensors to collect routine data such as temperature, current, voltage, etc., and then determine whether the equipment is abnormal based on the preset fixed threshold. Once the threshold is exceeded, an alarm will be issued to prompt the operation and maintenance personnel to handle it. However, in the face of dynamic changes in the operating status of the power grid, such as sudden increases and decreases in loads, sudden changes in the external environment, etc., the existing power system has problems such as difficulty in making effective adjustments quickly and poor dynamic response adaptability. Summary of the invention

[0004] In view of the above-mentioned problems, the purpose of the present invention is to provide a method and system for self-healing control and energy efficiency optimization of a box-type substation, which realizes fast and accurate fault diagnosis and self-healing control, as well as dynamic energy efficiency optimization by real-time monitoring and optimization of the operating parameters of the box-type substation, thereby significantly improving the power supply reliability and energy utilization efficiency of the power system.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is a box-type substation self-healing control and energy efficiency optimization method, comprising the following steps: Step 1: Collect the three-phase current, voltage waveform and temperature parameters of the equipment in the box-type substation through a distributed sensor array, construct a spatiotemporal correlation feature matrix, and extract the dynamic operation mode; Step 2: Based on the dynamic operation mode, a nonlinear propagation model of impedance parameters is constructed to calculate the node credibility propagation value. When the node credibility propagation value exceeds the preset threshold, a circuit breaker tripping instruction is automatically generated to isolate the faulty link and trigger the backup line switching mechanism to restore power supply to the non-faulty area. Step 3: According to the faulty link, a multi-objective optimization model for dynamic weight optimization is established to solve the topology reconstruction strategy under asymmetric constraints; Step 4: Based on the topology reconstruction strategy, the voltage in the energy storage scheduling parameters is optimized, and the inertia weight adaptive control instructions are generated in combination with the optical fiber self-healing ring network communication channel; Step 5: Use a closed-loop feedback mechanism to correct the dynamic weight and voltage in the energy storage scheduling parameters in real time to complete the iteration of dynamic performance compensation and energy efficiency optimization.

[0006] Preferably, in step 1, the spatiotemporal correlation feature matrix It is expressed as: ; in, for The current components in the coordinate system, is the voltage component in the dq coordinate system, represents the Klaunik product, is the temperature gradient tensor, is the magnetic flux of the g-th sensor, is the total number of sensors, t is the current time, is the insulation aging index, , is the initial time of equipment commissioning, for The dielectric conductivity of the time, for The electric field strength over time, is the time-integrated variable, .

[0007] Preferably, in step 2, the node credibility propagation value is calculated as follows: ; in, For Node The credibility propagation value of is the i-th node in the fault propagation network, For Node The initial confidence level of , is the initial confidence benchmark value, is the flux change sensitivity coefficient, is the sensor number belonging to the i-th node, The total number of sensors belonging to the i-th node, It represents the multiplication of all neighbor nodes j of node i, and the total number of nodes is N. represents the set of all neighbor nodes of node i, e is a natural constant, is the impedance between node i and node j, is the reference impedance, is the nonlinear attenuation factor of fault propagation, is the fault propagation coefficient between node i and node j, , is the standard deviation of characteristic impedance, is the line resistance from node i to node j, is the line reactance from node i to node j.

[0008] Preferably, in step 3, the objective function of the multi-objective optimization model is: ; in, is the dynamic weight adjustment coefficient, , is the power loss, is the reference power, is the weight of the voltage deviation term, is the voltage deviation, is the rated voltage, is the weight adjustment factor, is the deviation between the current power loss and the reference value, e is a natural constant, , V is voltage, t is current time, is the average winding temperature.

[0009] Preferably, the asymmetric constraint is: ; in, is the mth voltage constraint function, is the voltage of the ith node, is the reference voltage, is the harmonic current change rate, is the harmonic current, is the maximum constraint allowed threshold, is the dynamic relaxation factor, , N is the total number of nodes, is the frequency deviation of the ith node, is the system reference frequency, is the error function.

[0010] Preferably, in step 4, the inertia weight is self-adapted based on a genetic algorithm, and the adjustment formula is: ; in, is the inertia weight, is the minimum allowed weight, is the maximum allowed weight, is the quantum bit state value, , is the initial value of the quantum rotation angle, e is a natural constant, is the number of iterations, is the convergence rate adjustment factor, , is the change of fitness function, is the average fitness value.

[0011] Preferably, the initial value of the quantum rotation angle is The method for determining is: ; in, is the reference impedance, is the total impedance change, , K is the total number of devices, is the impedance of the kth device at time t, For the kth device at time The impedance, is the sampling time interval.

[0012] Preferably, in step 5, in the closed-loop feedback mechanism, the function of parameter correction is: ; in, is the corrected proportionality coefficient, is the proportionality coefficient, is the voltage deviation sign function, is the voltage deviation, e is a natural constant, is the threshold voltage, , is the rated voltage, V is the voltage, is the current time, is the sampling time window, is the short-term integration variable, is a short-term integration variable The voltage at the time.

[0013] Preferably, the closed-loop feedback mechanism also includes a harmonic energy redistribution algorithm. When the dynamic weight and the voltage in the energy storage scheduling parameter are corrected in real time through the closed-loop feedback mechanism, the harmonic energy redistribution algorithm is triggered to adjust the harmonic energy distribution. The harmonic energy redistribution algorithm is expressed as: ; in, is the corrected harmonic energy, is the harmonic energy, is the effective value of the fundamental voltage, is the effective value of the nth harmonic voltage, .

[0014] The box-type substation self-healing control and energy efficiency optimization system is used to implement the above method, including: The matrix construction module is used to collect the three-phase current, voltage waveform and temperature parameters of the equipment in the box-type substation through a distributed sensor array, construct a time-space correlation feature matrix, and extract dynamic operation modes; The credibility propagation value calculation module is used to construct a nonlinear propagation model of impedance parameters based on dynamic operation modes, calculate the node credibility propagation value, and automatically generate a circuit breaker trip command to isolate the faulty link when the node credibility propagation value exceeds the preset threshold, and trigger the backup line switching mechanism to restore power supply to the non-faulty area; A model building module is used to build a multi-objective optimization model for dynamic weight optimization based on the faulty link and solve the topology reconstruction strategy under asymmetric constraints; The weight adaptation module is used to optimize the voltage in the energy storage scheduling parameters based on the topology reconstruction strategy, and generate control instructions for inertia weight adaptation in combination with the optical fiber self-healing ring network communication channel; The parameter correction module is used to correct the voltage in the dynamic weight and energy storage scheduling parameters in real time through a closed-loop feedback mechanism to complete the iteration of dynamic performance compensation and energy efficiency optimization.

[0015] The beneficial effects of the present invention are: The present invention constructs a spatiotemporal correlation feature matrix, uses Clarke-Park transformation and dynamic mode decomposition technology, and extracts multi-dimensional features such as current, voltage, and temperature in real time, thereby enhancing the identifiability and spatiotemporal correlation of fault features. This enables the system to extract fault features and accurately perceive the state within milliseconds, quickly locate fault links (such as short circuits, insulation breakdown, etc.), and automatically generate circuit breaker trip instructions to isolate the fault area, while triggering the backup line switching mechanism to restore power supply to non-fault areas. This rapid response mechanism significantly improves the power supply reliability of the power system, reduces fault expansion and power outage time, and is particularly suitable for scenarios such as hospitals and data centers that have extremely high requirements for power supply continuity.

[0016] The present invention breaks through the limitations of traditional binary logic and achieves accurate positioning of fault links under complex working conditions by establishing a nonlinear fault propagation model, integrating impedance parameters and improved fuzzy Petri nets, and calculating node credibility propagation values. Combined with the dynamic constraint relaxation mechanism and the error function to dynamically adjust the weight coefficient and the constraint upper limit, it is possible to balance the conflict of multi-objective optimization under asymmetric constraints and search for the global optimal solution. This greatly improves the optimization success rate of the system in scenarios with high penetration of new energy, significantly improves energy utilization, and ensures the stability and security of the system.

[0017] The present invention uses a closed-loop feedback mechanism to correct dynamic weights and energy storage scheduling parameters in real time, completing dynamic performance compensation and energy efficiency optimization iteration. The introduction of the harmonic energy redistribution algorithm further reduces harmonic distortion, improves power quality, protects equipment, and extends equipment life. In addition, the system achieves multi-objective dynamic balance and achieves Pareto optimality among conflicting objectives such as voltage deviation, loss, and equipment life. From the initial stage of commissioning to the aging stage, the parameters are continuously self-calibrated to ensure that the system remains stable and efficient under different operating conditions, significantly improving the overall performance and economy of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 Schematic diagram of the modules of the system of the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0020] Example 1: Figure 1 As shown, the box-type substation self-healing control and energy efficiency optimization method includes the following steps: Step 1: Multi-dimensional data collection and dynamic operation mode extraction: The three-phase current, voltage waveform and temperature parameters of the equipment in the box-type substation are collected through a distributed sensor array, and a spatiotemporal correlation feature matrix is ​​constructed to extract the dynamic operation mode.

[0021] The three-phase current of the equipment (transformer, high-voltage switchgear, low-voltage switchgear, reactive power compensation device, current transformer and voltage transformer, etc.) in the box-type substation is collected in real time through a distributed sensor array. ,Voltage and temperature , using the Clarke-Park transform to convert it into Decoupling components in the stationary coordinate system and the dq rotating coordinate system , . Combined magnetic flux , temperature gradient and insulation aging index , construct the spatiotemporal correlation feature matrix : ; in, for The current components in the coordinate system, is the voltage component in the dq coordinate system, represents the Klaunik product, is the temperature gradient tensor, is the magnetic flux of the g-th sensor, is the total number of sensors, t is the current time, is the insulation aging index, , is the initial time of equipment commissioning, for The dielectric conductivity of the time, for The electric field strength over time, is the time-integrated variable, .

[0022] In this embodiment, the functions of the dynamic operation mode are: Fault feature enhancement: through The transformation eliminates three-phase unbalanced interference and separates transient features such as high-frequency harmonics and voltage sags from the power frequency signal.

[0023] Spatiotemporal correlation analysis: Klaunik product ( ) integrates the spatiotemporal coupling characteristics of current and voltage, such as The correlation between power fluctuation and impedance change can be characterized.

[0024] Long-term aging monitoring: Quantify the cumulative degradation of insulation materials and provide early warning of breakdown risks.

[0025] The spatiotemporal correlation feature matrix is ​​the "data center" of the entire system. (corresponding to ) is directly used in step 2 to reflect the magnetic field mutation and serves as the input feature of the short-circuit fault.

[0026] The dynamic operation mode can reduce the number of false alarms and avoid unnecessary power outages caused by sensor noise. It is especially suitable for scenarios such as hospitals and data centers that have extremely high requirements for power supply continuity.

[0027] Step 2: Nonlinear fault propagation diagnosis, i.e. core fault location: Based on the dynamic operation mode, a nonlinear propagation model of impedance parameters is constructed to calculate the node credibility propagation value. When the node credibility propagation value exceeds the preset threshold, a circuit breaker tripping instruction is automatically generated to isolate the faulty link and trigger the backup line switching mechanism to restore power supply to the non-faulty area.

[0028] A nonlinear propagation model of impedance parameters is established, that is, a calculation model of node credibility propagation value, specifically: ; in, For Node The credibility propagation value is between 0 and 1, and the closer it is to 1, the higher the probability of failure. is the i-th node in the fault propagation network, For Node The initial confidence level of , is the initial confidence benchmark value, is the flux change sensitivity coefficient, is the sensor number belonging to the i-th node, The total number of sensors belonging to the i-th node, is the product of all neighbor nodes j of node i, the total number of nodes is N, {1, 2, ..., i, j ..., N}, i ≠ j, represents the set of all neighbor nodes of node i, e is a natural constant, is the impedance between node i and node j, is the reference impedance, is the nonlinear attenuation factor of fault propagation, which is obtained by training with historical fault data. Its typical value is 1.2~1.8. is the fault propagation coefficient between node i and node j, , is the standard deviation of characteristic impedance, is the line resistance from node i to node j, is the line reactance from node i to node j.

[0029] The space-time matrix of step 1 is used to provide the dynamic operating mode, that is, (inter-node impedance), (impedance standard deviation) and other key parameters are used as input data of the model.

[0030] when When the fault is isolated,

[0031] Step 2 can accurately locate the faulty link (such as short circuit, insulation breakdown), greatly reducing the positioning error. of When a fault occurs, the node is automatically isolated and the backup line is started. Compared with traditional manual inspection, the efficiency is greatly improved, and the failure expansion accidents are reduced. It is particularly suitable for unmanned substations in harsh environments in overseas projects.

[0032] Step 3: Dynamic constraint energy efficiency optimization: Based on the faulty links, a multi-objective optimization model for dynamic weight optimization is established to solve the topology reconstruction strategy under asymmetric constraints.

[0033] Dynamic weights are used to adjust the weight distribution of each objective in real time. For example, when some links fail, dynamic weights can reduce the weight of these links in the optimization objective while increasing the weight of other normal links, thus ensuring that the optimization strategy can adapt to the dynamic changes of the network.

[0034] The objective function of the multi-objective optimization model is: ; in, is the dynamic weight adjustment coefficient, , is the power loss, is the reference power, is the weight of the voltage deviation term, is the voltage deviation, is the rated voltage, is the weight adjustment factor, is the deviation between the current power loss and the reference value, , is the voltage change rate, V is the voltage, is the average winding temperature.

[0035] The dynamic constraint relaxation mechanism, i.e. the asymmetric constraint condition, is constructed as: ; in, is the mth voltage constraint function, is the voltage of the ith node, is the reference voltage, is the harmonic current change rate, is the harmonic current, is the maximum constraint allowed threshold, is the dynamic relaxation factor, , N is the total number of nodes, is the frequency deviation of the ith node, is the system reference frequency, is the error function.

[0036] The fault location results in step 2 have determined which nodes are excluded from the optimization model. The temperature gradient data from step 1 is automatically reduced at high temperatures Weight, give priority to ensuring equipment safety.

[0037] For example, when a branch is triggered by overload When it drops to 0.7, the system allows the voltage deviation of the branch to be temporarily relaxed to 15%, and at the same time starts the energy storage discharge to compensate for the power shortage to avoid global optimization failure.

[0038] Step 4: Generate quantum intelligent control strategy, that is, search for the global optimal solution: Based on the topology reconstruction strategy, optimize the voltage in the energy storage scheduling parameters, and generate inertia weight adaptive control instructions in combination with the optical fiber self-healing ring network communication channel.

[0039] Due to the dynamic adjustment of the inertia weight, the genetic algorithm in this embodiment can quickly explore the global solution space in the early stage and focus on local search in the later stage, thereby improving the convergence speed and accuracy of the algorithm.

[0040] Optimize energy storage scheduling and topology reconstruction. The corresponding inertia weight adaptive adjustment formula based on genetic algorithm is: ; in, is the inertia weight, is the minimum allowed weight, is the maximum allowed weight, is the quantum bit state value, , is the initial value of the quantum rotation angle, is the number of iterations, is the convergence rate adjustment factor, , is the change of fitness function, is the average fitness value, and ln is the logarithm with base e.

[0041] Initial value of quantum rotation angle The method for determining is: ; in, is the total impedance change ( This is the reference impedance of the corresponding device). , K is the total number of devices, is the impedance of the kth device at time t, For the kth device at time The impedance, is the sampling time interval.

[0042] The strategy output is to generate circuit breaker action sequence, energy storage charging and discharging power, capacitor switching instructions, etc.

[0043] Among them, the optimization objective function directly inherits step 3 ; Constraints The dynamic update frequency is synchronized with the data acquisition in step 1 (1000 times per second).

[0044] For example, when a sudden drop in renewable energy output is detected, the algorithm generates a strategy of "cutting off non-critical loads + maximum discharge of energy storage + interconnection line power support" within 50ms, shortening the voltage recovery time from 2 seconds in traditional methods to 0.3 seconds. This greatly improves the power supply recovery speed of the box-type transformer in extreme weather such as typhoons and lightning strikes, reducing economic losses.

[0045] Step 5: Closed-loop feedback and dynamic iteration, i.e., system self-update: The dynamic weight and voltage in the energy storage scheduling parameters are corrected in real time through a closed-loop feedback mechanism to complete dynamic performance compensation and energy efficiency optimization iteration.

[0046] In the closed-loop feedback mechanism, the function of parameter correction is: ; in, is the corrected proportionality coefficient, is the proportionality coefficient, is the voltage deviation sign function, is the voltage deviation, is the threshold voltage, , is the rated voltage, is the sampling time window, is a short-term integration variable The voltage at is the short-term integration variable.

[0047] The closed-loop feedback mechanism also includes a harmonic energy redistribution algorithm. When the dynamic weight and the voltage in the energy storage dispatch parameter are corrected in real time through the closed-loop feedback mechanism, the harmonic energy redistribution algorithm is triggered to adjust the harmonic energy distribution. The harmonic energy redistribution algorithm is expressed as: ; in, is the corrected harmonic energy, is the harmonic energy, is the effective value of the fundamental voltage, is the effective value of the nth harmonic voltage, . The spatiotemporal correlation feature matrix from step 1, Harmonic distortion rate parameters derived from the spatiotemporal correlation feature matrix.

[0048] The harmonic energy redistribution algorithm reduces harmonic distortion, improves power quality, protects equipment, extends equipment life, and optimizes system performance by real-time monitoring and adjusting the harmonic energy distribution in the system, ensuring that the system remains stable and efficient under different operating conditions.

[0049] This embodiment achieves multi-objective dynamic balance: achieving Pareto optimality among conflicting objectives such as voltage deviation, loss, and equipment life. Full life cycle adaptation: From the initial operation to the aging stage, parameters are continuously self-calibrated.

[0050] Example 2: Figure 2 As shown, the box-type substation self-healing control and energy efficiency optimization system is used to implement the method in Example 1, including: The matrix construction module is used to collect the three-phase current, voltage waveform and temperature parameters of the equipment in the box-type substation through a distributed sensor array, construct a time-space correlation feature matrix, and extract dynamic operation modes; The credibility propagation value calculation module is used to construct a nonlinear propagation model of impedance parameters based on dynamic operation modes, calculate the node credibility propagation value, and automatically generate a circuit breaker trip command to isolate the faulty link when the node credibility propagation value exceeds the preset threshold, and trigger the backup line switching mechanism to restore power supply to the non-faulty area; A model building module is used to build a multi-objective optimization model for dynamic weight optimization based on the faulty link and solve the topology reconstruction strategy under asymmetric constraints; The weight adaptation module is used to optimize the voltage in the energy storage scheduling parameters based on the topology reconstruction strategy, and generate control instructions for inertia weight adaptation in combination with the optical fiber self-healing ring network communication channel; The parameter correction module is used to correct the voltage in the dynamic weight and energy storage scheduling parameters in real time through a closed-loop feedback mechanism to complete the iteration of dynamic performance compensation and energy efficiency optimization.

Claims

1. The self-healing control and energy efficiency optimization method of box-type substation is characterized by: The following steps are involved: Step 1: Collect the three-phase current, voltage waveform and temperature parameters of the equipment in the box-type substation through a distributed sensor array, construct a spatiotemporal correlation feature matrix, and extract the dynamic operation mode; Step 2: Based on the dynamic operation mode, a nonlinear propagation model of impedance parameters is constructed to calculate the node credibility propagation value. When the node credibility propagation value exceeds the preset threshold, a circuit breaker tripping instruction is automatically generated to isolate the faulty link and trigger the backup line switching mechanism to restore power supply to the non-faulty area. Step 3: According to the faulty link, a multi-objective optimization model for dynamic weight optimization is established to solve the topology reconstruction strategy under asymmetric constraints; Step 4: Based on the topology reconstruction strategy, the voltage in the energy storage scheduling parameters is optimized, and the inertia weight adaptive control instructions are generated in combination with the optical fiber self-healing ring network communication channel; Step 5: Use a closed-loop feedback mechanism to correct the dynamic weight and voltage in the energy storage scheduling parameters in real time to complete the iteration of dynamic performance compensation and energy efficiency optimization.

2. The self-healing control and energy efficiency optimization method of a box-type substation according to claim 1, characterized in that: In step 1, the spatiotemporal correlation feature matrix It is expressed as: ； in, for The current components in the coordinate system, is the voltage component in the dq coordinate system, represents the Klaunik product, is the temperature gradient tensor, is the magnetic flux of the g-th sensor, is the total number of sensors, t is the current time, is the insulation aging index, , is the initial time of equipment commissioning, for The dielectric conductivity of the time, for The electric field strength over time, is the time-integrated variable, .

3. The self-healing control and energy efficiency optimization method of a box-type substation according to claim 1, characterized in that: In step 2, the calculation method of node credibility propagation value is: ； in, For Node The credibility propagation value of is the i-th node in the fault propagation network, For Node The initial confidence level of , is the initial confidence benchmark value, is the flux change sensitivity coefficient, is the sensor number belonging to the i-th node, The total number of sensors belonging to the i-th node, It represents the multiplication of all neighbor nodes j of node i, and the total number of nodes is N. represents the set of all neighbor nodes of node i, e is a natural constant, is the impedance between node i and node j, is the reference impedance, is the nonlinear attenuation factor of fault propagation, is the fault propagation coefficient between node i and node j, , is the standard deviation of characteristic impedance, is the line resistance from node i to node j, is the line reactance from node i to node j.

4. The self-healing control and energy efficiency optimization method of a box-type substation according to claim 1, characterized in that: In step 3, the objective function of the multi-objective optimization model is: ； in, is the dynamic weight adjustment coefficient, , is the power loss, is the reference power, is the weight of the voltage deviation term, is the voltage deviation, is the rated voltage, is the weight adjustment factor, is the deviation between the current power loss and the reference value, e is a natural constant, , V is voltage, t is current time, is the average winding temperature.

5. The self-healing control and energy efficiency optimization method of a box-type substation as claimed in claim 4 is characterized in that: The asymmetric constraints are: ； in, is the mth voltage constraint function, is the voltage of the ith node, is the reference voltage, is the harmonic current change rate, is the harmonic current, is the maximum constraint allowed threshold, is the dynamic relaxation factor, , N is the total number of nodes, is the frequency deviation of the ith node, is the system reference frequency, is the error function.

6. The self-healing control and energy efficiency optimization method of a box-type substation according to claim 1, characterized in that: In step 4, the inertia weight adaptation is based on a genetic algorithm, and its adjustment formula is: ； in, is the inertia weight, is the minimum allowed weight, is the maximum allowed weight, is the quantum bit state value, , is the initial value of the quantum rotation angle, e is a natural constant, is the number of iterations, is the convergence rate adjustment factor, , is the change of fitness function, is the average fitness value.

7. The self-healing control and energy efficiency optimization method of a box-type substation as claimed in claim 6, characterized in that: Initial value of quantum rotation angle The method to determine is: ； in, is the reference impedance, is the total impedance change, , K is the total number of devices, is the impedance of the kth device at time t, For the kth device at time The impedance, is the sampling time interval.

8. The self-healing control and energy efficiency optimization method of a box-type substation according to claim 1, characterized in that: In step 5, in the closed-loop feedback mechanism, the function of parameter correction is: ； in, is the corrected proportionality coefficient, is the proportionality coefficient, is the voltage deviation sign function, is the voltage deviation, e is a natural constant, is the threshold voltage, , is the rated voltage, V is the voltage, is the current time, is the sampling time window, is the short-term integration variable, is a short-term integration variable The voltage at the time.

9. The method for self-healing control and energy efficiency optimization of box-type substation according to claim 1, characterized in that: The closed-loop feedback mechanism also includes a harmonic energy redistribution algorithm. When the dynamic weight and the voltage in the energy storage dispatch parameter are corrected in real time through the closed-loop feedback mechanism, the harmonic energy redistribution algorithm is triggered to adjust the harmonic energy distribution. The harmonic energy redistribution algorithm is expressed as: ； in, is the corrected harmonic energy, is the harmonic energy, is the effective value of the fundamental voltage, is the effective value of the nth harmonic voltage, .

10. A box-type substation self-healing control and energy efficiency optimization system, used to implement the method described in any one of claims 1 to 9, characterized in that: include: The matrix construction module is used to collect the three-phase current, voltage waveform and temperature parameters of the equipment in the box-type substation through a distributed sensor array, construct a time-space correlation feature matrix, and extract dynamic operation modes; The credibility propagation value calculation module is used to construct a nonlinear propagation model of impedance parameters based on dynamic operation modes, calculate the node credibility propagation value, and automatically generate a circuit breaker trip command to isolate the faulty link when the node credibility propagation value exceeds the preset threshold, and trigger the backup line switching mechanism to restore power supply to the non-faulty area; A model building module is used to build a multi-objective optimization model for dynamic weight optimization based on the faulty link and solve the topology reconstruction strategy under asymmetric constraints; The weight adaptation module is used to optimize the voltage in the energy storage scheduling parameters based on the topology reconstruction strategy, and generate control instructions for inertia weight adaptation in combination with the optical fiber self-healing ring network communication channel; The parameter correction module is used to correct the voltage in the dynamic weight and energy storage scheduling parameters in real time through a closed-loop feedback mechanism to complete the iteration of dynamic performance compensation and energy efficiency optimization.

Images