Energy management methods based on AI algorithms and large models

Abstract

This invention provides an energy management method based on AI algorithms and large models, belonging to the field of energy management; it solves the problem of low energy management efficiency; specifically as follows: It calculates the expected electricity consumption rate of rigid and adjustable loads in each municipal district, designs a greedy algorithm to adjust the power supply from the power supply end to each municipal district; it calculates the equivalent receiving voltage and expected transmission loss of each municipal district, and, combined with the actual transmission loss, introduces distributed new energy power generation equipment to the target area, and adjusts the operation status of the power grid; it continuously updates the expected electricity consumption rate of rigid and adjustable loads in each municipal district, and adjusts the operation status of the power supply end and the power grid; this invention improves the stability of energy supply in the target area by fitting and mathematically modeling the load of each municipal district through load matching and mathematical modeling.

Description

Technical Field

[0001] This invention relates to an energy management method based on AI algorithms and large models, and pertains to the field of energy management. Background Technology

[0002] Existing methods for energy management have the following shortcomings:

[0003] Lack of load management: With the rapid growth of high-power loads such as electric vehicles and data centers, the peak-valley difference of the power grid will further widen, leading to a surge in peak-shaving pressure on the generation side and highlighting the lack of flexibility of traditional power sources such as coal and gas power. In extreme cases, the lack of power supply will lead to frequent peak electricity demand (i.e., a surge in electricity consumption in multiple regions during a certain period of time), and may even trigger regional power shortage crises. Without scientific load management methods, the power grid may be forced to rely on higher-cost energy storage or demand response resources to maintain balance.

[0004] Poor power dispatching performance: In urban areas with high power density, the increased cable coverage leads to increased charging power. If the reactive power distribution cannot be accurately calculated, it may cause safety hazards such as local overvoltage or resonant overvoltage. With the development of AC / DC hybrid power grids (such as the increase in flexible DC transmission projects), the traditional method of unilaterally adjusting the active power of the power supply end based on the power demand of a certain area is becoming increasingly unsuitable. It cannot simulate the impact of power electronic equipment on the system, which will lead to planning errors.

[0005] The failure to consider reactive power compensation technology: Some existing energy management methods ignore the impact of reactive power on voltage stability when dispatching power resources. If reactive power compensation technology is not adopted in the power dispatching process, the power quality of the power grid will deteriorate in the future, inducing harmonic pollution and reactive power fluctuations in the power grid, and causing the traditional fixed capacitor bank compensation method to become completely ineffective. Summary of the Invention

[0006] In view of the shortcomings of existing technologies, the purpose of this invention is to provide an energy management method based on AI algorithms and large models, which aims to solve the problem of low energy management efficiency.

[0007] To achieve the above objectives, the present invention provides an energy management method based on AI algorithms and large models, comprising:

[0008] Obtain the quantity and historical power consumption of each municipal district in the target area, and calculate the expected power consumption rate of rigid load and adjustable load in each municipal district; obtain and determine whether the power supply power of the power supply end in the target area is sufficient; if sufficient, adjust the power supply power of the power supply end to each municipal district based on the expected power consumption rate of rigid load and adjustable load in each municipal district, and design a greedy algorithm; if insufficient, obtain the reactive power of the power supply end and adjust the power supply end.

[0009] Based on the expected electricity consumption rate of rigid and adjustable loads in each municipality, the equivalent receiving voltage of each municipality is calculated, and the expected transmission loss of each municipality is calculated based on the equivalent receiving voltage of each municipality. The actual receiving voltage of each municipality is obtained to calculate the actual transmission loss. Based on the expected transmission loss and the actual transmission loss, distributed new energy power generation equipment is introduced to the target area, and the operation status of the power grid is adjusted.

[0010] Continuously update the expected electricity consumption rate of rigid and adjustable loads in each municipal district, and adjust the operation status of the power supply and the power grid.

[0011] Furthermore, the specific steps to determine whether the active power is sufficient are as follows:

[0012] Obtain the expected electricity consumption rate (pl) of the rigid load in the first municipal district from 1 to 24 hours. （1，1） ~pl （1，24） The expected power consumption rate pd of adjustable load （1,1） ~pd （1,24） ;

[0013] Similarly, the expected electricity consumption rate pl of rigid loads in mn municipal districts （mn，1） ~pl （mn，24） The expected power consumption rate pd of adjustable load （mn,1） ~pd （mn,24） ;

[0014] Obtain the power supply Pu at the power supply end of the target area; calculate pl （1，1） ~pl （mn，24） and PD （1，1） ~pd （mn，24） and Pld;

[0015] Calculate the equivalent total electricity consumption rate Pb for the first municipal district. （1） Pb （1） =Pld / 24;

[0016] Calculate the equivalent total electricity consumption rate Pb for the 2nd to mnth municipal districts. （2） ~Pb （mn） ;

[0017] Calculate Pb （1） ~Pb （mn） By comparing the values ​​of Pu and aPp, we can determine whether the active power at the power supply end is sufficient.

[0018] If Pu≥aPp, it means that the active power at the power supply end is sufficient. Based on the expected power consumption rate of the rigid load and adjustable load of each municipal district, the peak-shifting adjustment is carried out for the first to the mnth municipal districts.

[0019] If Pu < aPp, it means that the active power at the power supply end is insufficient. Obtain the reactive power Qu at the power supply end and adjust the power supply and reactive power at the power supply end.

[0020] Adjust the power supply at the power supply end to aPp;

[0021] Calculate the reactive power aQu after adjustment at the power supply end:

[0022] .

[0023] Furthermore, the specific steps for staggering peak hours in the first municipal district are as follows:

[0024] Calculate the total electricity consumption rate zp of the first municipal district at time 1. （1） zp （1） =pl （1，1） +pd （1,1） ;

[0025] Total electricity consumption at 2 o'clock zp （2） zp （2） =pl （1，2） +pd （1,2） ;

[0026] And so on, the total electricity consumption rate zp at 24 hours （24） zp （24） =pl （1，24） +pd （1,24） ;

[0027] Define the steps of a greedy algorithm:

[0028] Extract zp （1） ~zp （24） The maximum value zpa in （max） Extract zpa （max） The expected power consumption rate pda corresponding to the adjustable load;

[0029] Calculate the power distribution aP in the first distribution （1） :aP （1） = pda / 24;

[0030] Extract zpa （max） time period tt （1） Among them, tt （1） ∈[1~24];

[0031] The first municipal district will be located in tt （1） The total electricity consumption rate decreased by 23 × aP （1） The first municipal district will be located between 1 and (tt) （1） -1) and (tt) （1） +1) Total electricity consumption increased by aP up to 24 hours.（1） The primary power consumption rate is obtained as zp. （1，1） ~zp （1，24） ;

[0032] Extract zp （1，1） ~zp （1，24） The maximum value zp in （1，max） and minimum value zp （1，min） Calculate the greedy electricity consumption difference Δzp. （1） : ;

[0033] Define the processing steps for a quadratic greedy algorithm.

[0034] Furthermore, the steps for handling the second-order greedy algorithm are as follows:

[0035] Extract zp （1，1） ~zp （1，24） The maximum value zpb （max） Extract zpb （max） The expected electricity consumption rate (plb) corresponding to rigid loads;

[0036] Calculate the secondary distribution power aP （2） : ;

[0037] Extract zpb （max） time period tt （2） Among them, tt （2） ∈[1~24];

[0038] The first municipal district will be located in tt （2） The total electricity consumption rate decreased by 23 × aP （2） The first municipal district will be located between 1 and (tt) （2） -1) and (tt) （2） +1) Total electricity consumption increased by aP up to 24 hours. （2） The secondary power consumption rate is obtained as zp （2，1） ~zp （2，24） ;

[0039] Extract zp （2，1） ~zp （2，24） The maximum value zp in （2，max） and minimum value zp （2，min） Calculate the power consumption difference Δzp of the quadratic greedy algorithm. （2） : ;

[0040] Compare Δzp （1） With Δzp （2） The size of the value determines the termination condition of the greedy algorithm;

[0041] If Δzp （1）≥Δzp （2） Then the greedy algorithm ends, according to zp （1，1） ~zp （1，24） Adjust the power supply from the power supply terminal to the first municipal district;

[0042] If Δzp （1） <Δzp （2） If the greedy algorithm is repeated twice, the greedy algorithm will continue to be executed until the difference in power consumption rate between the two greedy algorithms is less than or equal to the difference in power consumption rate between the two greedy algorithms.

[0043] Furthermore, the specific steps for calculating the expected transmission loss are as follows:

[0044] Obtain the expected electricity consumption rate of rigid load and adjustable load for each municipal district, and calculate the equivalent total electricity consumption rate po for the 1st to mnth municipal districts. （1） ~po （mn） ;

[0045] Obtain the apparent power Se and rated voltage Ue at the power supply end; let the equivalent total electricity consumption rate of the i-th municipality be po. （i） The equivalent received voltage is uo （i） Construct equation A-1:

[0046] ;

[0047] Calculate the equivalent received voltage uo corresponding to the 1st to mnth municipal districts. （1） ~uo （mn） ;

[0048] Let Yo be the equivalent admittance of the i-th municipal district. （i） Construct formula A-2:

[0049] ;

[0050] Let Io be the equivalent transmission current of the i-th municipal district. （i） Construct formula A-3:

[0051] ;

[0052] Let the equivalent impedance of the i-th municipality be Zo. （i） Construct formula A-4:

[0053] ;

[0054] Calculate the equivalent transmission current Io for the first to the mnth municipal districts. （1） ~Io （mn） Equivalent impedance Zo （1） ~Zo （mn） ;

[0055] Extract Zo （1） ~Zo （mn） The real and imaginary parts are used to obtain the equivalent resistance Ro of the 1st to mnth municipalities. （1） ~Ro （mn） and equivalent reactance Xo （1） ~Xo （mn） ;

[0056] Calculate the expected active power loss Plo of the first municipal district. （1） : ;

[0057] Expected reactive power loss Qlo （1） : ;

[0058] Similarly, the expected active power loss Plo of the mn-th municipal district （mn） :

[0059] ;

[0060] Expected reactive power loss Qlo （mn） : ;

[0061] Calculate the power transmission capacity when the power supply terminal supplies power to the 1st to the mnth municipal districts.

[0062] Furthermore, the specific steps for calculating transmission power are as follows:

[0063] When calculating the power supply from the power supply end to the first municipal district, the transmission power is Pv （1） :

[0064] Where j represents the imaginary unit;

[0065] Similarly, when supplying power to the mn-th municipal district, the transmission power is Pv. （mn） :

[0066] ;

[0067] The transmission voltage when the power supply terminal supplies power to the 1st to the mnth municipal districts is uniformly set to Ue, and the transmission power when the power supply terminal supplies power to the 1st to the mnth municipal districts is set to Pv. （1） ~Pv （mn） Power is supplied to the first to the mnth municipal districts, and the actual received power Pt of the first to the mnth municipal districts is recorded. （1） ~Pt （mn） ;

[0068] According to Pv （1） ~Pv （mn）and Pt （1） ~Pt （mn） Calculate the actual active power loss ΔPl for the first to the mnth municipal districts. （1） ~ΔPl （mn） And actual reactive power loss ΔQl （1） ~ΔQl （mn） ;

[0069] Determine if only ΔQl exists （1） ~ΔQl （mn） Negative values ​​exist in it;

[0070] If ΔPl （1） ~ΔPl （mn） and ΔQl （1） ~ΔQl （mn） If there are no negative values ​​in the middle, then no action is taken;

[0071] If only ΔQl （1） ~ΔQl （mn） If there are negative values, the municipal districts with negative actual reactive power loss are marked as District I, and reactive power compensation is carried out for District I.

[0072] If ΔPl （1） ~ΔPl （mn） If there are negative values, then new energy power generation equipment is introduced to the power supply end for grid connection of new energy, and adjustments are made for frequency stability, voltage stability and power angle stability.

[0073] Furthermore, the specific steps for reactive power compensation are as follows:

[0074] Count the number of zones I, nl, and obtain the actual reactive power loss Qle of zones I from the 1st to the nlth. （1） ~Qle （nl） Expected reactive power loss Qls （1） ~Qls （nl） ;

[0075] Obtain the excitation potential Ef and power factor angle θ at the power supply end, and perform reactive power compensation for the first I zone;

[0076] Obtain the equivalent reactance Xl of the first region I, and let the new transmission voltage from the power supply end to the first region I be Ul. Construct equation B-1:

[0077] ;

[0078] Obtain the equivalent impedance Zl and equivalent admittance Yl of the first region I. Let the new power transmitted from the power supply end to the first region I be Ptl. Construct equation B-2:

[0079] ;

[0080] Calculate the values ​​of Ul and Ptl; keep the power supply voltage to the first zone I constant, and adjust the power supply power to the first zone I to prt;

[0081] Reactive power compensation is performed on the 2nd to nth I zones.

[0082] Furthermore, the specific steps for frequency stabilization adjustment are as follows:

[0083] The municipal districts with negative actual active power loss are designated as District II, and the actual active power loss Pls of District II is obtained. （1） ~Pls （nr） Where nr represents the number of zones II;

[0084] Obtain the actual reactive power loss Qls of the first to nrth II zones. （1） ~Qls （nr） ;

[0085] Calculate the compensated transmission power ΔPp in Zone 1 II. （1） :

[0086] ;

[0087] Calculate the compensated transmission power ΔPp for the 2nd to nrth II zones. （2） ~ΔPp （nr） ;

[0088] Calculate ΔPp （1） ~ΔPp （nr） The sum of absolute values, Pts, represents the new energy power generation equipment with a power generation capacity of Pts introduced to the power supply end;

[0089] The virtual inertia coefficient of the simulated synchronous generator of the new energy power generation equipment is obtained, and the compensation power is generated according to the grid frequency change rate and virtual inertia coefficient of the target area to quickly suppress RoCoF in the early stage of frequency fluctuation.

[0090] Calculate the frequency deviation Δf of the power grid based on the proportional coefficient Kp of the primary frequency regulation. （1） Calculate the adjustment power Pfc: Pfc=Δf×Kp （1） Primary frequency regulation is achieved by using PFC to correct the static deviation of power grid frequency fluctuations.

[0091] By introducing the AGC mechanism, the steady-state error of power grid frequency fluctuations is eliminated, frequency recovery is achieved, and the power grid frequency is stabilized and adjusted.

[0092] Furthermore, the adjustment steps for voltage stabilization and power angle stabilization are as follows:

[0093] Calculate the additional admittance ΔPp of the 1st, 2nd to nrth regions II. （1） ΔPp （2）~ΔPp （nr） ;

[0094] The municipal districts where the actual active power loss is not negative are designated as District III, and the equivalent admittance Ym of District III is obtained. （1） ~Ym （ns） Where ns represents the number of regions III;

[0095] Calculate ΔYp （1） ~ΔYp （nr） and Yb （1） ~Yb （nr） and Ym （1） ~Ym （ns） Sum the total Yj; construct the admittance matrix Yg;

[0096] Obtain the maximum transmission power PE of the power station （max） Let the power factor angle of the power supply station be υ, and let the equivalent power factor angle of the 1st, 2nd, up to the nrth II zone be φ. （1） φ （2） ~φ （nr） ;

[0097] Let ψ be the equivalent power factor angle of the 1st to the nsth III regions. （1） ~ψ （ns） ;

[0098] Obtain the transmission power Pwr of the first to nrth II zones. （1） ~Pwr （nr） ;

[0099] Obtain the transmission power Pws of the first to the nsth III zones. （1） ~Pws （nr） ;

[0100] Construct the voltage matrix Ug and the power matrix Pg;

[0101] Define formula C: Pg = Yg × Ug;

[0102] Calculate φ using Newton-Raphson's algorithm. （1） φ （2） ~φ （nr） and ψ （1） ~ψ （ns） The value of .

[0103] Furthermore, the adjustment steps for voltage stabilization and power angle stabilization also include:

[0104] Calculate φ （1） ~φ （nr） and ψ （1） ~ψ （ns） The ratio of the corresponding cosine values ​​yields xr （1） ~xr （nr） xs（1） ~xs （ns） ;

[0105] Calculate xr （1） ~xr （nr） and XS （1） ~xs （ns） and axx;

[0106] When the power supply terminal transmits power to the first Zone II, the transmission power is adjusted as follows:

[0107] ;

[0108] Similarly, when the power supply end transmits power to the nrth II zone, the transmission power is adjusted as follows:

[0109] ;

[0110] When the power supply terminal transmits power to Zone 1 (III), the transmission power is adjusted as follows:

[0111] ;

[0112] Similarly, when the power supply terminal transmits power to the ns-th zone III, the transmission power is adjusted as follows:

[0113] .

[0114] Compared with the prior art, the beneficial effects of the present invention are:

[0115] Peak load adjustment: This invention uses peak power dispatching to optimize the time distribution of load, shifting the electricity demand during peak hours to off-peak hours, significantly reducing the peak load of the power grid, reducing the generation side's dependence on peak-shaving capacity, delaying the expansion investment of power equipment, and helping users reduce electricity expenses (such as reducing demand charges).

[0116] Refined power distribution: This invention provides a scientific basis for the refined distribution of power systems based on power flow calculations. By simulating the voltage, power flow and line load of each node in the power grid, this invention optimizes the power distribution scheme in different areas, reduces line losses, improves the utilization rate of transformers and transmission equipment, and avoids local voltage drops or over-limit problems caused by uneven load.

[0117] Stabilizing voltage and power: This invention improves voltage stability by regulating the flow of reactive power in the power grid, reducing the additional losses caused by reactive current, mitigating voltage fluctuations caused by the grid connection of intermittent power sources such as new energy power generation equipment, and enhancing the power grid's ability to absorb renewable energy. Attached Figure Description

[0118] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0119] Figure 1 This is a schematic diagram of the method of the present invention;

[0120] Figure 2 This is a schematic diagram of the processing flow of the present invention;

[0121] Figure 3 This is a schematic diagram of the processing flow of the present invention. Detailed Implementation

[0122] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0123] Please see Figure 1 and Figure 2 Energy management methods based on AI algorithms and large models include:

[0124] Step S1: Obtain the quantity and historical power consumption of each municipal district in the target area, and calculate the expected power consumption rate of rigid load and adjustable load in each municipal district; obtain and determine whether the power supply power of the power supply end in the target area is sufficient; if sufficient, adjust the power supply power of the power supply end to each municipal district based on the expected power consumption rate of rigid load and adjustable load in each municipal district, and design a greedy algorithm; if insufficient, obtain the reactive power of the power supply end and adjust the power supply end.

[0125] It should be noted that, in this invention, "target area" refers to a city-level region where energy management is carried out using this invention (an energy management method based on AI algorithms and large models);

[0126] The specific steps of step S1 are as follows:

[0127] Rigid load: refers to electrical equipment or institutions whose power consumption time or power consumption cannot be adjusted, such as hospitals, government buildings, traffic lights, etc.

[0128] Adjustable load: refers to power-consuming equipment or mechanisms whose power consumption time or power consumption can be adjusted, such as street lighting, industrial production lines, electric vehicle charging stations, etc.

[0129] Calculate the expected electricity consumption rate for rigid and adjustable loads in each municipal district:

[0130] The historical electricity consumption data for rigid and adjustable loads per hour over the past three days for each municipal district is obtained. Time series analysis of the historical electricity consumption is performed using the ARIMA model to calculate the difference order, autoregressive parameters, and moving average parameters for the corresponding rigid and adjustable loads of each municipal district. Based on the difference order, autoregressive parameters, and moving average parameters for the corresponding rigid and adjustable loads of each municipal district, the autocorrelation coefficient and moving average coefficient for each municipal district are calculated. Based on the autocorrelation coefficient and moving average coefficient for each municipal district, the expected total electricity consumption for rigid and adjustable loads per hour in the next day is estimated using the ARIMA model. Then, the expected total electricity consumption for rigid and adjustable loads per hour in each municipal district is divided by one hour to obtain the expected electricity consumption rate for rigid and adjustable loads per hour in each municipal district.

[0131] Obtain the number of municipal districts (mn); at 1:00, 2:00, and up to 24:00, obtain the expected electricity consumption rate (pl) of the rigid load of the first municipal district. （1，1） pl （1，2） ~pl （1，24） The expected power consumption rate pd of adjustable load （1,1） ,pd （1,2） ~pd （1,24） ;

[0132] Expected electricity consumption rate (pl) of rigid load in 2 municipal districts （2，1） pl （2，2） ~pl （2，24） The expected power consumption rate pd of adjustable load （2,1） ,pd （2,2） ~pd （2,24） ;

[0133] Similarly, the expected electricity consumption rate pl for rigid loads in mn municipal districts （mn，1） pl （mn，2） ~pl （mn，24） The expected power consumption rate pd of adjustable load （mn,1） ,pd （mn,2） ~pd （mn,24） ;

[0134] Obtain the power supply Pu at the power supply end of the target area; calculate pl （1，1） ~pl （mn，24） and PD （1，1） ~pd （mn，24） and Pld;

[0135] Calculate the equivalent total electricity consumption rate Pb for the first municipal district. （1） Pb （1） =Pld / 24;

[0136] Repeat Pb （1）The calculation process involves calculating the equivalent total electricity consumption rate Pb for the 2nd to mnth municipal districts. （2） ~Pb （mn） ;

[0137] Calculate Pb （1） ~Pb （mn） By comparing the values ​​of Pu and aPp, we can determine whether the active power at the power supply end is sufficient.

[0138] If Pu≥aPp, it means that the active power at the power supply end is sufficient. Based on the expected power consumption rate of the rigid load and adjustable load of each municipal district, peak-shifting adjustments are made for each municipal district.

[0139] Staggered peak hours will be implemented for the first municipal district;

[0140] Calculate the total electricity consumption rate zp of the first municipal district at time 1. （1） zp （1） =pl （1，1） +pd （1,1） ;

[0141] Total electricity consumption at 2 o'clock zp （2） zp （2） =pl （1，2） +pd （1,2） ;

[0142] And so on, the total electricity consumption rate zp at 24 hours （24） zp （24） =pl （1，24） +pd （1,24） ;

[0143] Define the steps of a greedy algorithm:

[0144] Extract zp （1） ~zp （24） The maximum value zpa in （max） Extract zpa （max） The expected power consumption rate pda corresponding to the adjustable load; where pda∈{pd} （1,1） ~pd （1,24）};

[0145] Calculate the power distribution aP in the first distribution （1） :aP （1） = pda / 24;

[0146] Extract zpa （max） time period tt （1） Among them, tt （1） ∈[1~24];

[0147] The first municipal district will be located in tt （1）The total electricity consumption rate decreased by 23 × aP （1） The first municipal district will be located between 1 and (tt) （1） -1) and (tt) （1） +1) Total electricity consumption increased by aP up to 24 hours. （1） The primary power consumption rate is obtained as zp. （1，1） ~zp （1，24） ;

[0148] Extract zp （1，1） ~zp （1，24） The maximum value zp in （1，max） and minimum value zp （1，min） Calculate the greedy electricity consumption difference Δzp. （1） : ;

[0149] Define the processing steps for a quadratic greedy algorithm:

[0150] Extract zp （1，1） ~zp （1，24） The maximum value zpb （max） Extract zpb （max） The expected electricity consumption rate plb corresponding to rigid load; where plb∈{pl （1，1） ~pl （1，24）};

[0151] Calculate the secondary distribution power aP （2） : ;

[0152] Extract zpb （max） time period tt （2） Among them, tt （2） ∈[1~24];

[0153] The first municipal district will be located in tt （2） The total electricity consumption rate decreased by 23 × aP （2） The first municipal district will be located between 1 and (tt) （2） -1) and (tt) （2） +1) Total electricity consumption increased by aP up to 24 hours. （2） The secondary power consumption rate is obtained as zp （2，1） ~zp （2，24） ;

[0154] Extract zp （2，1） ~zp （2，24） The maximum value zp in （2，max） and minimum value zp （2，min） Calculate the power consumption difference Δzp of the quadratic greedy algorithm. （2） : ;

[0155] Compare Δzp （1） With Δzp （2） The size of the value determines the termination condition of the greedy algorithm;

[0156] If Δzp （1） ≥Δzp （2） Then the greedy algorithm ends, according to zp （1，1） ~zp （1，24） Adjust the power supply from the power supply terminal to the first municipal district;

[0157] If Δzp （1） <Δzp （2） If the greedy algorithm is repeated twice, the greedy algorithm will continue to be executed until the difference in power consumption rate of the second greedy algorithm is less than or equal to the difference in power consumption rate of the first greedy algorithm.

[0158] Repeat the same steps for staggered peak hours adjustment for the first municipal district, and then apply the same steps for staggered peak hours adjustment for the second to the mnth municipal districts.

[0159] If Pu < aPp, it means that the active power at the power supply end is insufficient. Obtain the reactive power Qu at the power supply end and adjust the power supply and reactive power at the power supply end.

[0160] Adjust the power supply at the power supply end to aPp;

[0161] Calculate the reactive power aQu after adjustment at the power supply end:

[0162] .

[0163] Step S2: Based on the expected electricity consumption rate of rigid load and adjustable load in each municipality, calculate the equivalent receiving voltage of each municipality, and calculate the expected transmission loss (i.e., expected active power loss and expected reactive power loss) of each municipality according to the equivalent receiving voltage; obtain the actual receiving voltage of each municipality and calculate the actual transmission loss (i.e., actual active power loss and actual reactive power loss); introduce distributed new energy power generation equipment for the target area according to the expected transmission loss and actual transmission loss, and adjust the operation status of the power grid.

[0164] Please see Figure 3 The specific steps of step S2 are as follows:

[0165] Obtain the expected electricity consumption rate of rigid load and adjustable load for each municipal district, calculate the equivalent total electricity consumption rate for each municipal district, and obtain the equivalent total electricity consumption rate po for the 1st, 2nd, up to the mnth municipal district. （1） ,po （2） ~po （mn） ; (po （1） ~po （mn） For the calculation process, refer to the calculation process of aPb above.

[0166] Obtain the apparent power Se and rated voltage Ue at the power supply end; let the equivalent total electricity consumption rate of the i-th municipality be po. （i） The equivalent received voltage is uo （i） Construct equation A-1:

[0167] (Equation A-1); where the value of i ranges from 1 to mn;

[0168] po （1） ,po （2） ~po （mn） As a po （i） Substituting into equation A-1, calculate the equivalent received voltage uo corresponding to the 1st, 2nd, and up to the mnth municipal district. （1） uo （2） ~uo （mn） ;

[0169] Let Yo be the equivalent admittance of the i-th municipal district. （i） Construct formula A-2:

[0170] ;

[0171] Let Io be the equivalent transmission current of the i-th municipal district. （i） Construct formula A-3:

[0172] ;

[0173] Let the equivalent impedance of the i-th municipality be Zo. （i） Construct formula A-4:

[0174] ;in, Indicates conjugate;

[0175] (Based on formulas A-2 to A-4) Calculate the equivalent transmission current Io for the 1st, 2nd, and up to the mnth municipal districts. （1） Io （2） ~Io （mn） Equivalent impedance Zo （1） Zo （2） ~Zo （mn） ;

[0176] Extract Zo （1） Zo （2） ~Zo （mn） By taking the real and imaginary parts, we obtain the equivalent resistance Ro of the 1st, 2nd, and so on up to the mnth municipality. （1） Ro （2） ~Ro （mn） and equivalent reactance Xo（1） Xo （2） ~Xo （mn） (Where, the real part of the impedance represents the equivalent resistance, and the imaginary part represents the equivalent reactance.)

[0177] Calculate the expected active power loss Plo of the first municipal district. （1） : ;

[0178] Expected reactive power loss Qlo （1） : ;

[0179] The expected active loss Plo in the second municipal district （2） : ;

[0180] Expected reactive power loss Qlo （2） : ;

[0181] And so on, the expected active power loss Plo of the mn-th municipality. （mn） :

[0182] ;

[0183] Expected reactive power loss Qlo （mn） : ;

[0184] When calculating the power supply from the power supply end to the first municipal district, the transmission power is Pv （1） :

[0185] Where j represents the imaginary unit;

[0186] When supplying power to the second municipal district, the transmission power is Pv （2） :

[0187] ;

[0188] Similarly, when supplying power to the mn-th municipal district, the transmission power is Pv. （mn） :

[0189] ;

[0190] The transmission voltage when the power supply terminal supplies power to the 1st to the mnth municipal districts is uniformly set to Ue, and the transmission power when the power supply terminal supplies power to the 1st to the mnth municipal districts is set to Pv. （1） ~Pv （mn） ;

[0191] Power is supplied to the first to the mnth municipal districts, and the actual received power Pt of the first, second, and so on, is recorded for the mnth municipal districts. （1） Pt （2） ~Pt （mn） ;

[0192] Extracting Pt （1） Reduce po （1） The actual active power loss ΔPl of the first municipal district was obtained. （1） Pt （1） Reduce po （1） The imaginary part is used to obtain the actual reactive power loss ΔQl of the first municipal district. （1） ;

[0193] Extracting Pt （2） Reduce po （2） The actual active power loss ΔPl of the second municipal district was obtained. （2） Pt （2） Reduce po （2） The imaginary part is used to obtain the actual reactive power loss ΔQl in the second municipal district. （2） ;

[0194] And so on, extract Pt （mn） Reduce po （mn） The actual active power loss ΔPl of the mn-th municipal district is obtained from the real part. （mn） Pt （mn） Reduce po （mn） The imaginary part is used to obtain the actual reactive power loss ΔQl of the mn-th municipal district. （mn） ;

[0195] Determine if only ΔQl exists （1） ~ΔQl （mn） Negative values ​​exist in it;

[0196] If ΔPl （1） ~ΔPl （mn） and ΔQl （1） ~ΔQl （mn） If there are no negative values ​​in the middle, then no action is taken;

[0197] If only ΔQl （1） ~ΔQl （mn） If there are negative values, the municipal districts with negative actual reactive power loss are marked as District I, and reactive power compensation is carried out for District I.

[0198] If ΔPl （1） ~ΔPl （mn） If there are negative values, then new energy power generation equipment is introduced to the power supply end (if there is no new energy power generation equipment in the target area, then distributed power generation equipment is introduced to the power supply end) to connect new energy to the grid, and adjustments are made for frequency stability, voltage stability and power angle stability.

[0199] Count the number of zones I, nl, and obtain the actual reactive power loss Qle of zones I from the 1st to the nlth. （1） ~Qle （nl） Expected reactive power loss Qls （1） ~Qls （nl） ;

[0200] Among them, {Qle （1） ~Qle （nl）}⊊{ΔQl （1） ~ΔQl （mn）}, {Qls （1） ~Qls （nl）}⊊{Qlo （1） ~Qlo （nl）};

[0201] Obtain the excitation potential Ef and power factor angle θ at the power supply end, and perform reactive power compensation for the first I zone;

[0202] Obtain the equivalent reactance Xl of the first region I, and let the new transmission voltage from the power supply end to the first region I be Ul. Construct equation B-1:

[0203] Where, Xl∈{Xo （1） ~Xo （mn）};

[0204] Calculate the value of Ul based on equation B-1;

[0205] Obtain the equivalent impedance Zl and equivalent admittance Yl of the first region I. Let the new power transmitted from the power supply end to the first region I be Ptl. Construct equation B-2:

[0206] ;

[0207] Calculate the value of Ptl based on equation B-2;

[0208] Maintain the power supply voltage to the first zone I unchanged, and adjust the power supply to the first zone I to prt (to achieve reactive power compensation).

[0209] Repeat the reactive power compensation process for the first zone I, adjust the power transmission power from the power supply end to the second to nlth zones I, and perform reactive power compensation for the second to nlth zones I;

[0210] The municipal districts with negative actual active power loss are designated as District II, and the actual active power loss Pls of District II is obtained. （1） ~Pls （nr） Where nr represents the number of zones II;

[0211] Obtain the actual reactive power loss Qls of the first to nrth II zones. （1） ~Qls （nr） ;

[0212] Calculate the compensated transmission power ΔPp in Zone 1 II. （1） :

[0213] ;

[0214] Repeat ΔPp （1） The calculation steps are as follows: calculate the compensated transmission power ΔPp from the 2nd to the nrth II zone. （2） ~ΔPp （nr） ;

[0215] Calculate ΔPp （1） ~ΔPp （nr） The sum of absolute values, Pts, represents the new energy power generation equipment with a power generation capacity of Pts introduced to the power supply end;

[0216] The virtual inertia coefficient of the simulated synchronous generator of the new energy power station is obtained. Through virtual inertia control, that is, the compensation power is generated based on the grid frequency change rate of the power grid from the power supply end of the target area to the power grid of the entire municipal area and the virtual inertia coefficient, so as to quickly suppress RoCoF (frequency change rate) in the early stage of frequency fluctuation.

[0217] Calculate the frequency deviation Δf of the power grid (between two consecutive unit times) based on the primary frequency regulation proportional coefficient Kp. （1） Calculate the adjustment power Pfc: Pfc=Δf×Kp （1） Primary frequency regulation is achieved by using PFC to correct the static deviation of power grid frequency fluctuations.

[0218] By introducing an AGC (Automatic Generation Control) mechanism, the steady-state error of grid frequency fluctuations is eliminated, frequency recovery is achieved, and grid frequency stabilization adjustment is completed.

[0219] ΔPp （1） ΔPp （2） ~ΔPp （nr） As a po （i） Substituting into formula A-2, calculate the additional admittance ΔYp of the 1st, 2nd to nrth II regions. （1） ΔYp （2） ~ΔYp （nr） ;

[0220] Obtain the equivalent admittance Yb ​​of the 1st, 2nd, and up to the nrth II region. （1） ~Yb （nr） ;

[0221] The municipal districts with non-negative actual active power loss are designated as District III, and the equivalent admittance Ym of the first to the nsth District III is obtained.（1） ~Ym （ns） Where ns represents the number of regions III;

[0222] Calculate ΔYp （1） ~ΔYp （nr） and Yb （1） ~Yb （nr） and Ym （1） ~Ym （ns） The sum of Yj; construct the admittance matrix Yg of ((mn+1)×(mn+1)):

[0223] ;

[0224] The admittance matrix Yg has non-zero elements only in the diagonal, first row, and first column; all other elements are zero.

[0225] Obtain the maximum transmission power PE of the power station （max） Let the power factor angle of the power supply station be υ, and let the equivalent power factor angle of the 1st, 2nd, up to the nrth II zone be φ. （1） φ （2） ~φ （nr） ;

[0226] Let ψ be the equivalent power factor angle of the 1st to the nsth III regions. （1） ~ψ （ns） ;

[0227] Obtain the transmission power Pwr of the 1st, 2nd, up to the nrth II zone. （1） ,Pwr （2） ~Pwr （nr） ;

[0228] Obtain the transmission power Pws of the first to the nsth III zones. （1） Pws （2） ~Pws （nr） ;

[0229] Construct a voltage matrix Ug of ((mn+1)×1):

[0230] ;

[0231] Construct a power matrix Pg of ((mn+1)×1):

[0232] ;

[0233] Define formula C: Pg = Yg × Ug; where "×" in formula C represents matrix multiplication;

[0234] Calculate φ using Newton-Raphson's algorithm. （1） φ （2）~φ （nr） and ψ （1） ~ψ （ns） The value;

[0235] Calculate φ （1） φ （2） ~φ （nr） and ψ （1） ~ψ （ns） The ratio of the corresponding cosine values ​​yields: xr （1） :xr （2） :~:xr （nr） :xs （1） :~:xs （ns） ;

[0236] Calculate xr （1） xr （2） ~xr （nr） and XS （1） ~xs （ns） and axx;

[0237] When the power supply terminal transmits power to the first Zone II, the transmission voltage is maintained at Uz, and the transmission power is adjusted as follows:

[0238] ;

[0239] When the power supply terminal transmits power to the second II zone, the transmission voltage is maintained at Uz, and the transmission power is adjusted as follows:

[0240] ;

[0241] Similarly, when the power supply end transmits power to the nrth II zone, the transmission voltage is maintained at Uz, and the transmission power is adjusted as follows:

[0242] ;

[0243] When the power supply terminal transmits power to the first Zone III, the transmission voltage is maintained at Uz, and the transmission power is adjusted as follows:

[0244] ;

[0245] Similarly, when the power supply end transmits power to the ns-th zone III, the transmission voltage is maintained at Uz, and the transmission power is adjusted as follows:

[0246] .

[0247] Step S3: Continuously update the expected electricity consumption rate of rigid load and adjustable load in each municipal district, and adjust the operation status of the power supply end and the power grid.

[0248] The solutions from steps S1 to S3 are structured and stored in a database or vector knowledge base (such as Elasticsearch or Pinecone). Retrieval-based AI (such as customer service robots) directly matches the corresponding questions (i.e., peak-shaving scheduling, reactive power compensation, and new energy grid connection issues) through semantic search. For generative AI (such as GPT-like models), the solutions are used as training data or real-time retrieval content through fine-tuning or retrieval-enhanced generation (RAG) techniques to ensure the accuracy of the answers. Simultaneously, an automated update mechanism is established, triggering iterations of the model or knowledge base when solutions are revised, and continuously optimizing the output effect through test feedback, enabling AI to intelligently call and dynamically adapt to technical solutions.

[0249] The above formulas are all dimensionless calculations. The formulas are derived from software simulations using a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. For example, there are weighting coefficients and proportional coefficients. The values ​​set are to quantify each parameter to obtain a specific value, which is convenient for subsequent comparison. The values ​​of the weighting coefficients and proportional coefficients are only required to not affect the proportional relationship between the parameters and the quantified values.

[0250] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, 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 covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An energy management method based on AI algorithms and large-scale models, characterized in that: The method includes: Obtain the quantity and historical power consumption of each municipal district in the target area, and calculate the expected power consumption rate of rigid load and adjustable load in each municipal district; obtain and determine whether the power supply power of the power supply end in the target area is sufficient; if sufficient, adjust the power supply power of the power supply end to each municipal district based on the expected power consumption rate of rigid load and adjustable load in each municipal district, and design a greedy algorithm; if insufficient, obtain the reactive power of the power supply end and adjust the power supply end. The specific steps to determine whether the active power is sufficient are as follows: Obtain the expected electricity consumption rate (pl) of the rigid load in the first municipal district. （1，1） ~pl （1，24） The expected power consumption rate pd of adjustable load （1,1） ~pd （1,24） ; Obtain the expected electricity consumption rate pl of rigid loads in mn municipal districts. （mn，1） ~pl （mn，24） The expected power consumption rate pd of adjustable load （mn,1） ~pd （mn,24） ; Obtain the power supply Pu at the power supply end of the target area; calculate pl （1，1） ~pl （mn，24） and PD （1，1） ~pd （mn，24） and Pld; Calculate the equivalent total electricity consumption rate Pb for the first municipal district. （1） Pb （1） =Pld / 24; Calculate Pb （2） ~Pb （mn） ; Calculate Pb （1） ~Pb （mn） By comparing the values ​​of Pu and aPp, we can determine whether the active power at the power supply end is sufficient. If Pu≥aPp, it indicates that the active power at the power supply end is sufficient. Based on the expected power consumption rate of rigid load and adjustable load, peak-shifting adjustments are made for mn municipal districts. If Pu < aPp, it means that the active power at the power supply end is insufficient. Obtain the reactive power Qu at the power supply end and adjust the power supply and reactive power at the power supply end. Adjust the power supply at the power supply end to aPp; Calculate the reactive power aQu after adjustment at the power supply end: ； Based on the expected electricity consumption rate of rigid and adjustable loads in each municipality, the equivalent receiving voltage of each municipality is calculated, and the expected transmission loss of each municipality is calculated based on the equivalent receiving voltage of each municipality. The actual receiving voltage of each municipality is obtained to calculate the actual transmission loss. Based on the expected transmission loss and the actual transmission loss, distributed new energy power generation equipment is introduced to the target area, and the operation status of the power grid is adjusted. Continuously update the expected electricity consumption rate of rigid and adjustable loads in each municipal district, and adjust the operation status of the power supply and the power grid.

2. The energy management method based on AI algorithms and large models according to claim 1, characterized in that, The specific steps for staggering peak hours in the first municipal district are as follows: Calculate the total electricity consumption rate zp of the first municipal district at time 1. （1） zp （1） =pl （1，1） +pd （1,1） ; Total electricity consumption rate zp at 24 hours （24） zp （24） =pl （1，24） +pd （1,24） ; Define the steps of a greedy algorithm: Extract zp （1） ~zp （24） The maximum value zpa in （max） Extract zpa （max） The expected power consumption rate pda corresponding to the adjustable load; Calculate the power distribution aP in the first distribution （1） :aP （1） = pda / 24; Extract zpa （max） time period tt （1） ; Among them, tt （1） ∈[1~24]; The first municipal district will be located in tt （1） The total electricity consumption rate decreased by 23×aP. （1） The first municipal district will be located between 1 and (tt) （1） -1) and (tt) （1） +1) Total electricity consumption increased by aP up to 24 hours. （1） The primary power consumption rate is obtained as zp. （1，1） ~zp （1，24） ; Extract zp （1，1） ~zp （1，24） The maximum value zp in （1，max） and minimum value zp （1，min） Calculate the greedy electricity consumption difference Δzp. （1） : ; Define the processing steps for a quadratic greedy algorithm.

3. The energy management method based on AI algorithms and large models according to claim 2, characterized in that, The steps for handling the second-order greedy algorithm are as follows: Extract zp （1，1） ~zp （1，24） The maximum value zpb （max） Extract zpb （max） The expected electricity consumption rate (plb) corresponding to rigid loads; Calculate the secondary distribution power aP （2） : ; Extract zpb （max） time period tt （2） ; The first municipal district will be located in tt （2） The total electricity consumption rate decreased by 23×aP. （2） The first municipal district will be located between 1 and (tt) （2） -1) and (tt) （2） +1) Total electricity consumption increased by aP up to 24 hours. （2） The secondary power consumption rate is obtained as zp （2，1） ~zp （2，24） ; Extract zp （2，1） ~zp （2，24） The maximum value zp in （2，max） and minimum value zp （2，min） Calculate the power consumption difference Δzp of the quadratic greedy algorithm. （2） : ; Compare Δzp （1） With Δzp （2） The size of the value determines the termination condition of the greedy algorithm; If Δzp （1） ≥Δzp （2） Then the greedy algorithm ends, according to zp （1，1） ~zp （1，24） Adjust the power supply from the power supply terminal to the first municipal district; If Δzp （1） <Δzp （2） If the greedy algorithm is repeated twice, the greedy algorithm will continue to be executed until the difference in power consumption rate between the two greedy algorithms is less than or equal to the difference in power consumption rate between the two greedy algorithms.

4. The energy management method based on AI algorithms and large models according to claim 1, characterized in that, The specific steps for calculating the expected transmission loss are as follows: Obtain the expected electricity consumption rates of rigid and adjustable loads, and calculate the equivalent total electricity consumption rate po for mn municipal districts. （1） ~po （mn） ; Obtain the apparent power Se and rated voltage Ue at the power supply end; let the equivalent total electricity consumption rate of the i-th municipality be po. （i） The equivalent received voltage is uo （i） Construct equation A-1: ； Calculate the equivalent received voltage uo corresponding to the 1st to mnth municipal districts. （1） ~uo （mn） ; Let Yo be the equivalent admittance of the i-th municipal district. （i） Construct formula A-2: ； Let Io be the equivalent transmission current of the i-th municipal district. （i） Construct formula A-3: ； Let the equivalent impedance of the i-th municipality be Zo. （i） Construct formula A-4: ； Calculate the equivalent transmission current Io for the first to the mnth municipal districts. （1） ~Io （mn） Equivalent impedance Zo （1） ~Zo （mn） ; Extract Zo （1） ~Zo （mn） The real and imaginary parts are used to obtain the equivalent resistance Ro of the 1st to mnth municipalities. （1） ~Ro （mn） and equivalent reactance Xo （1） ~Xo （mn） ; Calculate the expected active power loss Plo of the first municipal district. （1） : ; Expected reactive power loss Qlo （1） : ; Similarly, calculate the expected active power loss Plo of the mn-th municipal district. （mn） And expected reactive power loss Qlo （mn） ; Calculate the power transmission capacity when the power supply terminal supplies power to the 1st to the mnth municipal districts.

5. The energy management method based on AI algorithms and large models according to claim 4, characterized in that, The specific steps for calculating transmission power are as follows: When calculating the power supply from the power supply end to the first municipal district, the transmission power is Pv （1） : Where j represents the imaginary unit; Similarly, when supplying power to the mn-th municipal district, the transmission power is Pv. （mn） ; The transmission voltage when the power supply terminal supplies power to the municipal area is uniformly set to Ue, and the transmission power when the power supply terminal supplies power to the municipal area is set to Pv. （1） ~Pv （mn） It supplies power to the city's jurisdiction and records the actual received power Pt in the city's jurisdiction. （1） ~Pt （mn） ; According to Pv （1） ~Pv （mn） and Pt （1） ~Pt （mn） Calculate the actual active power loss ΔPl for the first to the mnth municipal districts. （1） ~ΔPl （mn） And actual reactive power loss ΔQl （1） ~ΔQl （mn） ; Determine if only ΔQl exists （1） ~ΔQl （mn） Negative values ​​exist in it; If ΔPl （1） ~ΔPl （mn） and ΔQl （1） ~ΔQl （mn） If there are no negative values ​​in the middle, then no action is taken; If only ΔQl （1） ~ΔQl （mn） If there are negative values, the municipal districts with negative actual reactive power loss are marked as District I, and reactive power compensation is carried out for District I. If ΔPl （1） ~ΔPl （mn） If a negative value exists, it indicates that new energy power generation equipment is introduced to the power supply end for grid connection of new energy, and adjustments are made to stabilize the frequency, voltage, and power angle.

6. The energy management method based on AI algorithms and large models according to claim 5, characterized in that, The specific steps for reactive power compensation are as follows: Count the number of zones I, nl, and obtain the actual reactive power loss Qle of zones I from the 1st to the nlth. （1） ~Qle （nl） Expected reactive power loss Qls （1） ~Qls （nl） ; Obtain the excitation potential Ef and power factor angle θ at the power supply end, and perform reactive power compensation for the first I zone; Obtain the equivalent reactance Xl of the first region I, and let the new transmission voltage from the power supply end to the first region I be Ul. Construct equation B-1: ； Obtain the equivalent impedance Zl and equivalent admittance Yl of the first region I. Let the new power transmitted from the power supply end to the first region I be Ptl. Construct equation B-2: ； Calculate the values ​​of Ul and Ptl; keep the transmission voltage from the power supply end to the first I zone unchanged, and adjust the transmission power from the power supply end to the first I zone to prt; perform reactive power compensation for nl I zones.

7. The energy management method based on AI algorithms and large models according to claim 6, characterized in that, The specific steps for frequency stabilization adjustment are as follows: The municipal districts with negative actual active power loss are designated as District II, and the actual active power loss Pls of District II is obtained. （1） ~Pls （nr） Where nr represents the number of zones II; Obtain the actual reactive power loss Qls of the first to nrth II zones. （1） ~Qls （nr） ; Calculate the compensated transmission power ΔPp in Zone 1 II. （1） : ； Calculate the compensated transmission power ΔPp for the 2nd to nrth II zones. （2） ~ΔPp （nr） ; Calculate ΔPp （1） ~ΔPp （nr） The sum of absolute values, Pts, represents the new energy power generation equipment with a power generation capacity of Pts introduced to the power supply end; The virtual inertia coefficient of the simulated synchronous generator of the new energy power generation equipment is obtained, and the compensation power is generated according to the grid frequency change rate and virtual inertia coefficient of the target area to quickly suppress RoCoF in the early stage of frequency fluctuation. Calculate the frequency deviation Δf of the power grid based on the proportional coefficient Kp of the primary frequency regulation. （1） Calculate the adjustment power Pfc: Pfc=Δf×Kp （1） Primary frequency regulation is achieved by using PFC to correct the static deviation of power grid frequency fluctuations. By introducing the AGC mechanism, the steady-state error of power grid frequency fluctuations is eliminated, frequency recovery is achieved, and the power grid frequency is stabilized and adjusted.

8. The energy management method based on AI algorithms and large models according to claim 6, characterized in that, The adjustment steps for voltage stabilization and power angle stabilization are as follows: Calculate the additional admittance ΔPp of the 1st, 2nd to nrth regions II. （1） ΔPp （2） ~ΔPp （nr） ; The municipal districts where the actual active power loss is not negative are designated as District III, and the equivalent admittance Ym of District III is obtained. （1） ~Ym （ns） Where ns represents the number of regions III; Calculate ΔYp （1） ~ΔYp （nr） and Yb （1） ~Yb （nr） and Ym （1） ~Ym （ns） Sum the total Yj; construct the admittance matrix Yg; Obtain the maximum transmission power PE of the power station （max） Let the power factor angle of the power supply station be υ, and let the equivalent power factor angle of the first to nrth II zones be φ. （1） ~φ （nr） ; Let ψ be the equivalent power factor angle of the 1st to the nsth III regions. （1） ~ψ （ns） ; Obtain the transmission power Pwr of the first to nrth II zones. （1） ~Pwr （nr） ; Obtain the transmission power Pws of the first to the nsth III zones. （1） ~Pws （nr） ; Construct the voltage matrix Ug and the power matrix Pg; Pg = Yg × Ug; Calculate φ using Newton-Raphson's algorithm. （1） ~φ （nr） and ψ （1） ~ψ （ns） The value of .

9. The energy management method based on AI algorithms and large models according to claim 8, characterized in that, The adjustment steps for voltage stabilization and power angle stabilization also include: Calculate φ （1） ~φ （nr） and ψ （1） ~ψ （ns） The ratio of the corresponding cosine values ​​yields xr （1） ~xr （nr） xs （1） ~xs （ns） ; Calculate xr （1） ~xr （nr） and XS （1） ~xs （ns） and axx; When the power supply terminal transmits power to the first Zone II, the transmission power is adjusted as follows: Where Pts represents the power generation capacity of the new energy power generation equipment; Similarly, when the power supply end transmits power to the nrth II zone, the transmission power is adjusted as follows: ； When the power supply terminal transmits power to Zone 1 (III), the transmission power is adjusted as follows: ； Similarly, when the power supply terminal transmits power to the ns-th zone III, the transmission power is adjusted as follows: 。

Images