An intelligent linkage decision system and method for power grid dispatching in extreme weather

Abstract

The application discloses an intelligent linkage decision system and method for power grid dispatching in extreme weather, which deploys a micro weather station cluster in a transformer substation, integrates environmental temperature, wind speed, solar radiation and other multi-parameter sensors, combines an edge computing unit, is based on a transformer thermal circuit model, comprehensively considers the dynamic changes of a load rate, an environmental temperature and transformer loss, accurately calculates winding hot spot temperature, calculates an aging acceleration factor based on an Arrhenius equation, calculates cumulative life loss of transformer insulation, combines a Weibull distribution to calculate insulation aging probability, further fuses a thermal stress correction factor, an electric stress correction factor and a time cumulative effect factor, constructs a complete insulation breakdown probability calculation model, adopts a multi-objective optimization algorithm to obtain a deduction dispatching result, issues a dispatching instruction according to the deduction dispatching result, and realizes global safety joint dispatching in over-temperature weather.

Description

Technical Field

[0001] This invention relates to the field of intelligent linkage decision-making for power grid dispatch in extreme weather, and particularly to an intelligent linkage decision-making system and method for power grid dispatch in extreme high-temperature weather. Background Technology

[0002] Power system dispatching is a core technology for ensuring the safe, economical, and high-quality operation of the power grid. In recent years, extreme heat events have become increasingly frequent globally. When temperatures exceed 40°C, grid loads repeatedly reach new highs, and critical equipment such as transformers experience frequent failures due to overheating, especially insulation degradation leading to breakdowns. Traditional dispatching systems mainly rely on macroscopic forecast data from meteorological departments, which struggle to accurately reflect the micro-environmental conditions of key nodes such as substations and transmission corridors, resulting in delayed warnings of equipment overheating risks.

[0003] As a core piece of equipment in the power grid, the aging and failure of the transformer's insulation system (mainly oil-impregnated paper insulation) is a complex process involving multiple stress couplings, including thermal, electrical, mechanical, and environmental factors. The hot spot temperature of the transformer windings is a key driving factor for insulation aging; according to the Arrhenius equation, the insulation aging rate approximately doubles for every 6-10°C increase in temperature. However, existing dispatching systems lack real-time sensing capabilities for the internal hot spot temperature of transformers, typically only monitoring the top oil temperature. This makes it impossible to accurately assess the hottest temperature of the windings and its cumulative impact on insulation life, hindering dispatching systems from making decisions directly based on fault risk and leaving them stuck in a passive defense stage of "over-limit alarms."

[0004] Existing dispatching and equipment condition monitoring systems still rely on deterministic temperature limits or short-term overload multiples, failing to consider the health status of transformer core insulation equipment under extreme high-temperature conditions. Furthermore, dispatching processes typically do not consider the cascading effects and scope of impact after a fault, making it difficult to achieve risk balancing and minimize overall system losses across multiple transformers. In other words, there is a lack of methods for coordinated dispatching based on transformer insulation status. In emergency situations such as extreme high temperatures, dispatchers face the dual pressures of information overload and tight decision-making time, urgently requiring intelligent decision support tools. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an intelligent linkage decision-making system and method for power grid dispatching in extreme weather. This system deploys a cluster of micro-weather stations in substations, integrating multiple parameter sensors such as ambient temperature, wind speed, and solar radiation. Combined with edge computing units, and based on a transformer thermal circuit model, it comprehensively considers the dynamic changes in load rate, ambient temperature, and transformer losses to accurately calculate winding hot spot temperatures. It calculates the aging acceleration factor based on the Arrhenius equation, calculates the cumulative lifetime loss of transformer insulation, and combines this with the Weibull distribution to calculate the insulation aging probability. Furthermore, it integrates thermal stress correction factors, electrical stress correction factors, and time cumulative effect factors to construct a complete insulation breakdown probability calculation model. A multi-objective optimization algorithm is used to obtain the simulated dispatching results, and dispatching commands are issued based on these results to achieve full-area safe joint dispatching under extreme weather conditions.

[0006] This invention provides a smart coordinated decision-making method for power grid dispatching in extreme weather, the method comprising:

[0007] The substation edge processor acquires real-time environmental parameters, including ambient temperature, sunshine and wind speed at the location of the substation. When the ambient temperature exceeds a preset threshold, the substation edge processor calculates the transformer winding hot spot temperature based on the acquired environmental parameters.

[0008] The substation edge processor calculates the insulation breakdown probability based on the winding hot spot temperature and sends its own insulation breakdown probability to the joint dispatch center.

[0009] The joint dispatch center performs fault deduction based on a knowledge graph by receiving insulation breakdown probabilities uploaded by all substations in the dispatch area whose ambient temperature exceeds a preset threshold, and performs coordinated dispatch based on the fault deduction results.

[0010] Specifically, the substation edge processor calculates the transformer winding hot spot temperature based on the collected environmental parameters, including:

[0011] The substation edge processor calculates the transformer winding hot spot temperature T based on the collected environmental parameters. The specific calculation formula is as follows:

[0012] ;

[0013] Among them, t a For ambient temperature, Temperature correction for sunlight, For the top oil temperature rise relative to the environment, This represents the temperature difference between the hot spot of the winding and the top layer oil.

[0014] Specifically, the The formula for calculating the solar radiation correction temperature is as follows: ;

[0015] in, denoted as the solar radiation absorption coefficient of the transformer surface; S is the collected solar radiation intensity; A is the effective radiation-receiving area; G is the solar radiation absorption coefficient of the transformer surface; S is the solar radiation absorption coefficient of the transformer surface; G is the solar radiation absorption coefficient of the transformer surface; S is the solar radiation intensity of the transformer surface; A is the effective radiation-receiving area; G is the solar radiation absorption coefficient of the transformer surface ... th The total heat dissipation thermal conductivity of the transformer;

[0016] in, ;

[0017] P0 is the no-load loss; P k This is the rated load loss; This is the rated top oil temperature rise under windless standard conditions.

[0018] Specifically, the The specific calculation method for the top-layer oil temperature rise relative to the environment includes:

[0019] The edge processor collects the real-time current of the transformer and calculates the load factor K under the current ambient temperature.

[0020] ;

[0021] Among them, I 实 For the real-time current of the transformer, I 额 This refers to the rated current on the transformer's nameplate.

[0022] The top oil temperature rise relative to the environment The calculation formula is: ;

[0023] in, R represents the oil temperature rise under wind speed correction; R is the ratio of rated load loss to no-load loss; x is the oil index of the oil-immersed transformer.

[0024] Oil temperature rise under wind speed correction The calculation formula is:

[0025] ;

[0026] in ν is the rated top layer oil temperature rise under windless standard conditions; v is the normalized wind speed value; a and b are fitting coefficients.

[0027] Specifically, the temperature difference between the winding hot spot and the top layer oil The calculation formula is:

[0028] ;

[0029] in y is the hotspot gradient; K is the load factor at the current ambient temperature; y is the winding index.

[0030] Specifically, the substation edge processor calculates the insulation breakdown probability based on the winding hotspot temperature, including:

[0031] Calculate the transformer insulation aging factor F according to IEEE standards:

[0032] ;

[0033] Where 15000 is the aging rate constant recommended by the IEEE standard, and 383 is the reference temperature of 110°C. 0 The absolute temperature value at C, where T is the winding hot spot temperature;

[0034] Based on the transformer insulation aging factor F, the cumulative average equivalent aging factor F of the transformer can be calculated. j :

[0035] ;

[0036] Where n is the sampling order, n is a natural number, and N is the total number of samples; The duration of the nth sampling; The aging factor is calculated based on the average winding hot spot temperature during the sampling time at the nth sampling time.

[0037] The percentage of transformer insulation life loss The calculation formula is: ;

[0038] Where t is the transformer's operating time; L is the total test insulation life of the transformer.

[0039] Specifically, based on the percentage of transformer insulation life loss. Calculate the probability of insulation aging. :

[0040] ,

[0041] in, The shape parameter characterizes the steepness of the aging process; is a scale parameter that characterizes the percentage of insulation life loss when the aging probability reaches a preset threshold.

[0042] The breakdown probability D is obtained based on the insulation aging probability calculation:

[0043] ;

[0044] Where f(T) is the thermodynamic correction function; f(E) is the electric field stress correction function; and f(t) is the time accumulation correction function.

[0045] Specifically, the formula for calculating the thermodynamic correction function f(T) is as follows: ;

[0046] Where T is the winding hot spot temperature; 110 is the IEEE standard recommended reference temperature of 110°C. 0 C; 10 is an empirical constant;

[0047] The formula for calculating the thermodynamic correction function f(E) is as follows: ;

[0048] Among them, E f E represents the actual local electric field strength. r Z represents the rated electric field strength; z represents the properties of the insulating material.

[0049] The formula for calculating the thermodynamic correction function f(T) is as follows: ;

[0050] Where t c This represents the duration of the high-temperature state when the current ambient temperature exceeds a preset threshold. This is the characteristic time constant.

[0051] Specifically, the joint dispatch center implements coordinated dispatching, including:

[0052] After receiving the insulation breakdown probability uploaded by the substation, the joint dispatch center determines whether the insulation breakdown probability is greater than a first threshold. If it is greater, it uses the substation as the center and a preset distance as the radius to deduce the insulation breakdown fault evolution link based on the substation-related nodes in the knowledge graph. It then uses a multi-objective optimization algorithm to obtain the deduction and dispatch results and issues dispatch instructions based on the deduction and dispatch results.

[0053] The present invention also provides an intelligent linkage decision-making system for power grid dispatch in extreme weather, the system comprising: a joint dispatch center and multiple substation substations;

[0054] The substation substation includes an environmental information acquisition module, an edge processor, and a communication module; the substation edge processor acquires environmental parameters collected in real time by the environmental information acquisition module, including the ambient temperature, sunshine, and wind speed at the location of the substation.

[0055] When the ambient temperature exceeds a preset threshold, the substation edge processor calculates the transformer winding hot spot temperature based on the collected environmental parameters; and calculates the insulation breakdown probability based on the calculated winding hot spot temperature, and sends its own insulation breakdown probability to the joint dispatch center through the communication module.

[0056] The joint dispatch center includes a communication unit, a cloud server, and a database;

[0057] The cloud server performs fault deduction based on the insulation breakdown probability uploaded by all substations in the dispatch area whose ambient temperature exceeds a preset threshold, which is received by the communication unit. Based on the knowledge graph stored in the database, the cloud server generates dispatch instructions based on the fault deduction results and sends them to the corresponding substation substations through the communication unit.

[0058] The beneficial effects of this invention are as follows:

[0059] 1. Traditional transformer load management mainly relies on standards and guidelines such as IEC 60076-7 or IEEE C57.91, calculating winding hot spot temperatures using a fixed ambient temperature or a simplified temperature rise model. This invention utilizes substation-level micro-weather stations and edge computing nodes. A cluster of micro-weather stations is deployed in the substation, integrating multiple parameters such as ambient temperature, wind speed, and solar radiation sensors. Combined with edge computing units, based on a transformer thermal circuit model, it comprehensively considers the dynamic changes in load rate, ambient temperature, and transformer losses. This allows for management even when the ambient temperature exceeds a preset threshold, such as 40°C. 0 At time C, the winding hot spot temperature is accurately calculated, realizing the acquisition of micro-environmental parameters at the kilometer and minute levels and the real-time calculation of hot spot temperature. Compared with the traditional method that relies on macro-forecasts from meteorological departments, this invention improves the spatial resolution to 1-3 kilometers and the temporal resolution to the minute level, significantly reducing the error in hot spot temperature calculation.

[0060] 2. Transformer losses (no-load loss and load loss) are used as direct inputs for the top-layer oil temperature rise. Steady-state and dynamic temperature rises are calculated using a thermal circuit model, thereby obtaining the winding hot spot temperature. An aging acceleration factor is calculated based on the Arrhenius equation, and the insulation aging probability is calculated by calculating the cumulative life loss of transformer insulation and combining it with the Weibull distribution. Then, thermal stress correction factors, electrical stress correction factors, and time cumulative effect factors are further integrated to construct a complete insulation breakdown probability calculation model. The insulation breakdown probability value directly reflects the fault risk of transformer insulation breakdown under the current extreme ambient temperature, providing a comparable and cumulative decision-making basis for the dispatching system, upgrading the operation mode from "passive over-temperature protection" to "proactive risk management."

[0061] 3. Power grid equipment, loads, and power sources are modeled as knowledge graph nodes, while power supply relationships, protection coordination, and transfer paths are modeled as graph edges. When the predicted transformer breakdown probability exceeds a threshold, a graph propagation algorithm is used to search all downstream affected nodes, calculate the fault coupling strength, and obtain the expected loss of cascaded failures, providing a calculation basis for minimizing the total loss of the fault evolution link. Simultaneously, based on the knowledge graph, the minimum fault impact range is quantified, and the overall system scheduling cost is minimized to achieve multi-objective optimization joint scheduling. A multi-objective optimization algorithm is used to solve the optimal load allocation scheme, which can actively balance the insulation aging rate among different transformers, reduce the overall fault risk in the entire scheduling domain, and extend the service life of the transformer group.

[0062] 4. The NSGA-III algorithm is used to solve for the Pareto optimal solution set, with three objectives corresponding to: reducing economic losses caused by failures (loss dimension), controlling the social impact of failures (scope dimension), and reducing system operating costs (economic dimension). Dispatchers can select different Pareto solutions according to the warning level (red / orange / yellow / green), realizing flexible switching from "safety priority" to "economic priority", and achieving full-domain safe joint dispatching under extreme temperature weather. Attached Figure Description

[0063] Figure 1 This is a flowchart of the dynamic monitoring method for the power distribution network dispatching system;

[0064] Figure 2 This is a schematic diagram of the dynamic monitoring system structure of the power distribution network dispatching system. Detailed Implementation

[0065] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0066] This invention provides a smart linkage decision-making method for power grid dispatching in extreme weather, such as... Figure 1 As shown, the method includes:

[0067] The substation edge processor acquires real-time environmental parameters, including ambient temperature, sunshine and wind speed at the location of the substation. When the ambient temperature exceeds a preset threshold, the substation edge processor calculates the transformer winding hot spot temperature based on the acquired environmental parameters.

[0068] The substation edge processor calculates the insulation breakdown probability based on the winding hot spot temperature and sends its own insulation breakdown probability to the joint dispatch center.

[0069] The joint dispatch center performs fault deduction based on a knowledge graph by receiving insulation breakdown probabilities uploaded by all substations in the dispatch area whose ambient temperature exceeds a preset threshold, and performs coordinated dispatch based on the fault deduction results.

[0070] Specifically, based on the winding hotspot temperature calculation model provided by the IEEE standard, the substation edge processor calculates the transformer winding hotspot temperature T based on the collected environmental parameters, including:

[0071] ;

[0072] Among them, t a For ambient temperature, Temperature correction for sunlight, For the top oil temperature rise relative to the environment, This refers to the temperature difference between the hot spot of the winding and the top layer oil.

[0073] Specifically, regarding the top oil temperature rise relative to the environment First, the edge processor collects the real-time current of the transformer and calculates the load factor K under the current ambient temperature:

[0074] ;

[0075] Among them, I 实 For the real-time current of the transformer, I 额 This refers to the rated current on the transformer's nameplate.

[0076] The top oil temperature rise relative to the environment The calculation formula is: ;

[0077] in, R represents the oil temperature rise under wind speed correction; R is the ratio of rated load loss to no-load loss, both of which can be obtained from the transformer nameplate; x is the oil index of the oil-immersed transformer, which can be taken as 0.8 for self-cooled transformers and 1.0 for air-cooled transformers.

[0078] ;

[0079] in The rated top oil temperature rise under windless standard conditions can be obtained from the transformer nameplate or by experiment; v is the normalized wind speed value; a and b are fitting coefficients, for example, a=0.2 and b=0.5;

[0080] Specifically, Temperature correction for sunlight: ;

[0081] in, The solar radiation absorption coefficient of the transformer surface. S represents the collected solar radiation intensity; A represents the effective radiation-receiving area, which is the projected area of ​​the total surface area of ​​the fuel tank and radiator in the direction of the sun; G th The total heat dissipation thermal conductivity of the transformer;

[0082] in, P0 is the no-load loss; P k This is the rated load loss; The rated top oil temperature rise under windless standard conditions can be obtained from the transformer nameplate or by experiment, i.e., the rated steady-state oil temperature rise.

[0083] Specifically, The temperature difference between the hot spot of the winding and the top layer oil: ;

[0084] The hotspot gradient can be obtained from the transformer nameplate or through experiments; K is the load factor at the current ambient temperature; y is the winding index, which is selected and set according to the transformer type, and can be 1.3-1.6.

[0085] Specifically, the substation edge processor calculates the insulation breakdown probability based on the winding hotspot temperature, including:

[0086] The IEEE standard also provides a model for calculating the cumulative insulation loss of a transformer based on the temperature of the hottest spot in the winding. Because the internal temperature distribution of the transformer is uneven, the hottest spot is considered in the calculation. The Arrhenius chemical reaction rate theory is the source of the IEEE standard experimental equation for calculating the cumulative insulation loss of a transformer.

[0087] According to the IEEE standard, the formula for calculating the transformer insulation aging factor F is:

[0088] ;

[0089] Where 15000 is the aging rate constant recommended by the IEEE standard, and 383 is the reference temperature of 110°C. 0 The absolute temperature value at C, where T is the winding hot spot temperature;

[0090] During transformer operation, the hot spot temperature varies over time. It is necessary to perform a time-weighted average of the insulation aging factors at different times. Based on the transformer insulation aging factor F, the cumulative average equivalent aging factor F of the transformer can be calculated. j :

[0091] ;

[0092] Where n is the sampling order, n is a natural number, and N is the total number of samples; The duration of the nth sampling; The aging factor is calculated based on the average winding hot spot temperature during the sampling time at the nth sampling time.

[0093] The percentage of transformer insulation life loss The calculation formula is: ;

[0094] Where t is the transformer's operating time; L is the total test insulation life of the transformer;

[0095] Specifically, based on the percentage of transformer insulation life loss. Calculate the probability of insulation aging. :

[0096] ,

[0097] in, The shape parameter characterizes the steepness of the aging process; is a scale parameter that characterizes the percentage of insulation life loss when the aging probability reaches a preset threshold.

[0098] in, For shape parameters, The higher the value, the greater the probability of insulation aging. < The lower the time, the more... > The higher the time, the higher the A higher value indicates a steeper aging process and a more sudden end to the lifespan; older transformers should ideally have a higher value. Values, such as 3-4; For scale parameters, The smaller, the same The higher the probability of aging, the lower the required temperature for harsh environments / old transformer equipment. , The value can usually be 80%-95%;

[0099] In one embodiment, since the hotspot temperature calculation model indirectly infers the insulation aging probability model rather than directly measuring the actual aging state of the insulation, and the insulation aging probability model calculation involves multiple calculation steps, which can easily lead to cumulative bias, a cross-validation mechanism is needed to ensure the accuracy of the insulation aging probability model. The DGA gas ratio method is used for cross-validation.

[0100] Specifically, the cellulose insulation paper in the transformer decomposes under heat, with the main products being carbon oxides (CO, CO2) and a small amount of hydrocarbon gases. Therefore, the insulation aging factor F output by the hot spot temperature calculation model can be mapped to the theoretically required total amount of CO+CO2, and compared with the actual measured gas production for cross-validation.

[0101] Among them, the CO2 / CO ratio is closely related to the aging temperature: when the insulating paper is slowly thermally aged (low temperature, long period of time), CO2 generation is dominant and the CO2 / CO ratio is high; when local overheating occurs (high temperature, rapid aging), CO generation increases significantly and the CO2 / CO ratio decreases; therefore, the CO2 / CO ratio is also used as a basis for auxiliary verification.

[0102] Before verification, it is necessary to obtain the carbon oxide generation characteristics of the insulating paper used in the transformer. This is usually done by thermal aging test, measuring the cumulative gas production of CO+CO2, and fitting the gas production C per unit decrease in degree of polymerization.

[0103] ; where kg Here, F is the gas production coefficient, and C is the current aging factor. b This represents the baseline gas concentration during normal transformer operation.

[0104] Establish the percentage of transformer insulation life loss The correlation model between aggregation degree DP(t):

[0105] ,in, The fitting parameters were calibrated using historical transformer operating data.

[0106] During commissioning, input the initial degree of polymerization (DP0) of the transformer insulation paper and the initial concentrations of CO and CO2 in the oil into the edge processor. Set a sliding verification window (e.g., 2 months). At the start and end times of the verification window, calculate the percentage of insulation life loss based on the correlation model. Calculate the degree of polymerization value of insulation, and calculate the total amount of gas produced within the sliding verification window based on the amount of gas produced C per unit decrease in degree of polymerization;

[0107] Meanwhile, within the sliding verification window, the net increase in CO+CO2 production is detected based on the online gas monitoring device installed in the transformer oil.

[0108] The verification pass conditions are set as follows: within three consecutive sliding verification windows, the deviation between the total gas production and the net increase in gas production is less than a preset fixed threshold; the trend of the CO2 / CO ratio change is consistent with the equivalent aging temperature calculated based on the hot spot temperature calculation model; there is no sudden deviation (excluding single gas abnormal events, such as acetylene misjudgment caused by tap changer leakage).

[0109] Specifically, if cross-validation fails, the parameters of the hotspot temperature calculation model need to be corrected. This involves adjusting environmental correction coefficients in the hotspot temperature model (such as the absorption coefficient in solar radiation correction and the fitting coefficient in wind speed correction) to bring the temperature trend output by the model closer to the aging temperature indicated by the DGA. After correction, the calculation needs to be recalculated. Then compare it with historical DGA data until the deviation meets the requirements.

[0110] Specifically, in the process of assessing the probability of transformer insulation breakdown, insulation breakdown is a complex physical process involving multiple stress couplings and time-varying accumulation. The insulation aging probability reflects the intrinsic damage to the insulation material due to long-term thermal aging; however, whether an insulation breakdown fault occurs also depends on the current thermal stress, electric field stress, and duration. Therefore, the breakdown probability D is calculated based on the insulation aging probability. Specifically, this includes:

[0111] ;

[0112] Where f(T) is the thermodynamic correction function; f(E) is the electric field stress correction function; and f(t) is the time accumulation correction function.

[0113] Specifically, insulation breakdown is essentially the process by which charge carriers (electrons and ions) within the insulating material gain sufficient energy under the influence of an electric field, triggering an avalanche ionization. Temperature has a dual accelerating effect on this process: firstly, it reduces the breakdown field strength; secondly, it accelerates the development of partial discharge. Since the relationship between insulation breakdown and temperature conforms to the theory of chemical reaction rates, the thermodynamic correction function f(T) is set as follows:

[0114] Where T is the winding hot spot temperature; 110 is the IEEE standard recommended reference temperature of 110°C. 0 C; 10 is an empirical constant, meaning that for every 10 degrees Celsius increase in temperature... 0 C, the molecular thermal motion energy doubles, and the carrier transition probability increases exponentially;

[0115] Specifically, insulation breakdown requires two conditions to be met simultaneously: the existence of weak points in the insulating material and a sufficiently high electric field strength. The presence of weak points in the insulating material is characterized by the insulation aging probability. When the insulation aging probability is low, i.e., the insulation is in good condition, an extremely high electric field strength may directly lead to breakdown. Conversely, even if the insulation aging probability is high, breakdown will not occur if the electric field strength is low (e.g., equipment shutdown). The relationship between the electrical lifetime of the insulation and the electric field strength follows an inverse power law model; therefore, the electric field stress correction function f(E) is set as follows:

[0116] E f E represents the actual local electric field strength. r denoted as the rated electric field strength; z represents the insulation material properties, with values ​​of 9-11 for oil-impregnated paper insulation.

[0117] Specifically, transformer insulation breakdown is a process of energy accumulation and damage development, which takes time. A short voltage pulse may not be enough to complete the penetration, but a sustained high electric field will gradually advance this process. The electric field stress correction function f(t) is set as follows:

[0118] ;t c This represents the duration of the high-temperature state when the current ambient temperature exceeds a preset threshold. The characteristic time constant is the time that an insulating material can withstand under high stress.

[0119] Specifically, the joint dispatch center, based on the insulation breakdown probability uploaded by all substations in the region whose ambient temperature exceeds a preset threshold, performs fault prediction using a knowledge graph, and implements coordinated dispatch based on the fault prediction results, specifically including:

[0120] After receiving the insulation breakdown probability uploaded by the substation, the joint dispatch center determines whether the insulation breakdown probability is greater than a first threshold. If it is greater, it uses the substation as the center and a preset distance as the radius to deduce the insulation breakdown fault evolution link based on the substation-related nodes in the knowledge graph. It then uses a multi-objective optimization algorithm to obtain the deduction and dispatch results and issues dispatch instructions based on the deduction and dispatch results.

[0121] The multi-objective optimization algorithm adopts NSGA-III, which is based on the network topology relationship in the scheduling area knowledge graph, combined with power generation prediction, load prediction and insulation breakdown probability of multiple substations. It aims to minimize the total loss of fault-derived links, the fault impact range, and the total system scheduling cost, calculate the optimal scheduling solution and deduce the scheduling results.

[0122] The minimum total loss of the fault evolution link is the minimum transformer loss due to insulation breakdown and the minimum equipment failure loss in the set of downstream nodes that are topologically related to the transformer.

[0123] The minimum fault impact range means that the transformer with insulation breakdown has the smallest fault evolution link in the knowledge graph.

[0124] The minimum total scheduling cost of the system is equivalent to the minimum economic loss from scheduling.

[0125] The constraints of the multi-objective optimization algorithm include power balance, total transformer capacity, total generator output, and non-negative load shedding.

[0126] The optimal scheduling solution can be obtained by weighted sum method or fuzzy satisfaction function calculation. Then, the optimal scheduling solution is analyzed by the joint dispatch center, and corresponding dispatch instructions are generated and sent to each substation for execution.

[0127] After receiving the insulation breakdown probability uploaded by the substation, the joint dispatch center determines whether the insulation breakdown probability is greater than a first threshold. If it is, it uses the substation as the center and a preset distance as the radius to deduce the insulation breakdown fault evolution link based on the substation-related nodes in the knowledge graph, thereby realizing local area dispatching for substation breakdown faults. If multiple transformers are in a high-risk state in the fault evolution link, i.e., the insulation breakdown probability is greater than the first threshold, it calculates and infers the cascade failure probability. If the cascade failure probability is greater than a second threshold, it triggers a system-level emergency alarm and initiates unified dispatching across the entire network.

[0128] To address the aforementioned technical problems, this invention also provides an intelligent linkage decision-making system for power grid dispatching in extreme weather conditions, such as... Figure 2 As shown, the system includes: a joint dispatch center and multiple substation substations;

[0129] The substation substation includes an environmental information acquisition module, an edge processor, and a communication module; the substation edge processor acquires environmental parameters collected in real time by the environmental information acquisition module, including the ambient temperature, sunshine, and wind speed at the location of the substation.

[0130] When the ambient temperature exceeds a preset threshold, the substation edge processor calculates the transformer winding hot spot temperature based on the collected environmental parameters; and calculates the insulation breakdown probability based on the calculated winding hot spot temperature, and sends its own insulation breakdown probability to the joint dispatch center through the communication module.

[0131] The joint dispatch center includes a communication unit, a cloud server, and a database;

[0132] The cloud server performs fault deduction based on the insulation breakdown probability uploaded by all substations in the dispatch area whose ambient temperature exceeds a preset threshold, which is received by the communication unit. Based on the knowledge graph stored in the database, the cloud server generates dispatch instructions based on the fault deduction results and sends them to the corresponding substation substations through the communication unit.

[0133] Specifically, embodiments of the present invention also include a computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of the intelligent linkage decision-making method for power grid dispatch in extreme weather.

[0134] The processor can be a CPU, DSP, FPGA, or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0135] The memory may be ROM or other type of static storage device that can store static information and instructions, RAM or other type of dynamic storage device that can store information and instructions, or EEPROM or other optical disc storage, disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0136] The memory stores application code that executes the present invention, and its execution is controlled by a processor. The processor executes the application code stored in the memory to implement the content shown in the foregoing method embodiments.

[0137] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept.

Claims

1. A smart linkage decision-making method for power grid dispatching in extreme weather, characterized in that: The method includes: The substation edge processor acquires real-time environmental parameters, including ambient temperature, sunshine and wind speed at the location of the substation. When the ambient temperature exceeds a preset threshold, the substation edge processor calculates the transformer winding hot spot temperature based on the acquired environmental parameters. The substation edge processor calculates the insulation breakdown probability based on the winding hot spot temperature and sends its own insulation breakdown probability to the joint dispatch center. The joint dispatch center performs fault deduction based on a knowledge graph by receiving insulation breakdown probabilities uploaded by all substations in the dispatch area whose ambient temperature exceeds a preset threshold, and performs coordinated dispatch based on the fault deduction results.

2. The intelligent linkage decision-making method for power grid dispatching in extreme weather as described in claim 1, characterized in that: The substation edge processor calculates the transformer winding hot spot temperature based on the collected environmental parameters, specifically including: The substation edge processor calculates the transformer winding hot spot temperature T based on the collected environmental parameters. The specific calculation formula is as follows: ； Among them, t a For ambient temperature, Temperature correction for sunlight, For the top oil temperature rise relative to the environment, This represents the temperature difference between the hot spot of the winding and the top layer oil.

3. The intelligent linkage decision-making method for power grid dispatching in extreme weather as described in claim 2, characterized in that: The The formula for calculating the solar radiation correction temperature is as follows: ; in, denoted as the solar radiation absorption coefficient of the transformer surface; S is the collected solar radiation intensity; A is the effective radiation-receiving area; G is the solar radiation absorption coefficient of the transformer surface; S is the solar radiation absorption coefficient of the transformer surface; G is the solar radiation absorption coefficient of the transformer surface; S is the solar radiation intensity of the transformer surface; A is the effective radiation-receiving area; G is the solar radiation absorption coefficient of the transformer surface ... th The total heat dissipation thermal conductivity of the transformer; in, ; P0 is the no-load loss; P k This is the rated load loss; This is the rated top oil temperature rise under windless standard conditions.

4. The intelligent linkage decision-making method for power grid dispatching in extreme weather as described in claim 2, characterized in that: The The specific calculation method for the top-layer oil temperature rise relative to the environment includes: The edge processor collects the real-time current of the transformer and calculates the load factor K under the current ambient temperature. ； Among them, I 实 For the real-time current of the transformer, I 额 This refers to the rated current on the transformer's nameplate. The top oil temperature rise relative to the environment The calculation formula is: ; in, R represents the oil temperature rise under wind speed correction; R is the ratio of rated load loss to no-load loss; x is the oil index of the oil-immersed transformer. Oil temperature rise under wind speed correction The calculation formula is: ； in ν is the rated top layer oil temperature rise under windless standard conditions; v is the normalized wind speed value; a and b are fitting coefficients.

5. The intelligent linkage decision-making method for power grid dispatching in extreme weather as described in claim 2, characterized in that: The temperature difference between the winding hot spot and the top layer oil The calculation formula is: ； in y is the hotspot gradient; K is the load factor at the current ambient temperature; y is the winding index.

6. The intelligent linkage decision-making method for power grid dispatching in extreme weather as described in claim 1, characterized in that: The substation edge processor calculates the insulation breakdown probability based on the winding hot spot temperature, specifically including: Calculate the transformer insulation aging factor F according to IEEE standards: ； Where 15000 is the aging rate constant recommended by the IEEE standard, and 383 is the reference temperature of 110°C. 0 The absolute temperature value at C, where T is the winding hot spot temperature; Based on the transformer insulation aging factor F, the cumulative average equivalent aging factor F of the transformer can be calculated. j : ； Where n is the sampling order, n is a natural number, and N is the total number of samples; The duration of the nth sampling; The aging factor is calculated based on the average winding hot spot temperature during the sampling time at the nth sampling time. The percentage of transformer insulation life loss The calculation formula is: ; Where t is the transformer's operating time; L is the total test insulation life of the transformer.

7. The intelligent linkage decision-making method for power grid dispatching in extreme weather as described in claim 6, characterized in that: Based on the percentage of transformer insulation life loss Calculate the probability of insulation aging. : ， in, The shape parameter characterizes the steepness of the aging process; is a scale parameter that characterizes the percentage of insulation life loss when the aging probability reaches a preset threshold. The breakdown probability D is obtained based on the insulation aging probability calculation: ； Where f(T) is the thermodynamic correction function; f(E) is the electric field stress correction function; and f(t) is the time accumulation correction function.

8. The intelligent linkage decision-making method for power grid dispatch in extreme weather as described in claim 7, characterized in that: The formula for calculating the thermodynamic correction function f(T) is as follows: ; Where T is the winding hot spot temperature; 110 is the IEEE standard recommended reference temperature of 110°C. 0 C; 10 is an empirical constant; The formula for calculating the thermodynamic correction function f(E) is as follows: ; Among them, E f E represents the actual local electric field strength. r Z represents the rated electric field strength; z represents the properties of the insulating material. The formula for calculating the thermodynamic correction function f(T) is as follows: ; Where t c This represents the duration of the high-temperature state when the current ambient temperature exceeds a preset threshold. This is the characteristic time constant.

9. The intelligent linkage decision-making method for power grid dispatching in extreme weather as described in claim 1, characterized in that: The joint dispatch center implements coordinated dispatching, specifically including: After receiving the insulation breakdown probability uploaded by the substation, the joint dispatch center determines whether the insulation breakdown probability is greater than a first threshold. If it is greater, it uses the substation as the center and a preset distance as the radius to deduce the insulation breakdown fault evolution link based on the substation-related nodes in the knowledge graph. It then uses a multi-objective optimization algorithm to obtain the deduction and dispatch results and issues dispatch instructions based on the deduction and dispatch results.

10. A smart grid dispatching and decision-making system for extreme weather, characterized in that: The system includes: a joint dispatch center and multiple substation substations; The substation substation includes an environmental information acquisition module, an edge processor, and a communication module; the substation edge processor acquires environmental parameters collected in real time by the environmental information acquisition module, including the ambient temperature, sunshine, and wind speed at the location of the substation. When the ambient temperature exceeds a preset threshold, the substation edge processor calculates the transformer winding hot spot temperature based on the collected environmental parameters; and calculates the insulation breakdown probability based on the calculated winding hot spot temperature, and sends its own insulation breakdown probability to the joint dispatch center through the communication module. The joint dispatch center includes a communication unit, a cloud server, and a database; The cloud server performs fault deduction based on the insulation breakdown probability uploaded by all substations in the dispatch area whose ambient temperature exceeds a preset threshold, which is received by the communication unit. Based on the knowledge graph stored in the database, the cloud server generates dispatch instructions based on the fault deduction results and sends them to the corresponding substation substations through the communication unit.

Images