Optimal power flow method and device based on ai prediction

Abstract

This invention relates to the field of power system distribution technology and discloses a decision-making method for power distribution strategy of a photovoltaic-storage-DC-flexible system based on AI prediction. The method includes: acquiring real-time charging and discharging parameters of the batteries and supercapacitors within the photovoltaic system using an information acquisition module; calculating sampled values ​​based on captured sampling pulse signals; transmitting the calculation results to a decision center; the decision center outputting a decision through a fuzzy controller; calculating the battery's charge using a battery charge calculation module; and determining the supercapacitor's intervention point based on a comparison between the battery's charge value and the output current value under safe power limits. This AI-predictive power distribution strategy decision-making method and apparatus for a photovoltaic-storage-DC-flexible system can change the power supply of the photovoltaic-storage-DC-flexible system according to the electricity demand of the grid, avoiding the impact on the grid's power supply by directly limiting the battery's power supply.

Description

Technical Field

[0001] This invention relates to the field of power system distribution technology, specifically to a method and apparatus for decision-making on distribution strategies for photovoltaic-storage-DC-flexible systems based on AI prediction. Background Technology

[0002] Photovoltaic-storage-DC-flexible power distribution system is an abbreviation for four technologies: solar photovoltaic, energy storage, DC power distribution, and flexible interconnection. In this system, photovoltaic power generation installed on the building roof is connected to the DC bus via a DC-to-DC converter. The DC bus is connected to battery banks located in one or more places via a DC / DC converter. These battery banks can be individual energy storage batteries or batteries installed in some equipment to power the equipment. The charging / discharging process is called "photovoltaic"-"storage". "DC" refers to the realization of DC power supply. The DC power supply can include 375V DC for power and charging equipment, and a 48V DC branch obtained through DC / DC conversion for use by small-power appliances. The 375V DC bus is connected to the 380V AC power grid through an AC / DC converter to input electricity from the grid to meet the building's power demand. "Flexible" means that this system is not a rigid load where the power supply must equal the power consumption of the load side. Instead, the power taken from the grid can be adjusted within a large range according to the grid's supply and demand relationship. This power system becomes a flexible load of the grid. Therefore, in the photovoltaic-storage-DC-flexible system, the main distribution strategy is: to prioritize the generation of electricity by the photovoltaic system for self-consumption, then discharge the energy storage system, and finally supply energy through the grid. Excess photovoltaic power can also be "grid-connected" to maximize the benefits of photovoltaic power generation.

[0003] Currently, the power distribution strategy of photovoltaic-storage-DC-flexible systems is relatively simple. When photovoltaic power generation is insufficient, battery power is used. When the battery's stored capacity is insufficient, the system switches to grid power. Although the battery's discharge current can be limited to protect it, this measure leads to unstable power supply to the grid and fails to meet the grid's power demand. Since the power supply cannot be adjusted according to the grid's power demand, the battery is easily damaged by instantaneous high-current discharge. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides a method and apparatus for decision-making on power distribution strategies for photovoltaic-storage-DC-flexible systems based on AI prediction. This method can change the power supply of the photovoltaic-storage-DC-flexible system according to the power demand of the grid, avoiding the impact on the power supply of the grid caused by directly limiting the power supply of the battery. It solves the problems of not being able to change the power supply according to the power demand of the grid, which can easily cause damage to the battery due to instantaneous high current discharge. At the same time, this measure can lead to unstable power supply to the grid and failure to meet the power demand of the grid.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution: a power distribution strategy decision-making method for photovoltaic-storage-DC-flexible systems based on AI prediction, the method comprising:

[0008] The information acquisition module is used to obtain real-time charging and discharging parameters of the batteries and supercapacitors in the photovoltaic system.

[0009] The sampled values ​​are calculated based on the captured sampled pulse signals, and the calculation results are transmitted to the decision center, which then makes a decision output through a fuzzy controller.

[0010] The battery charge is calculated using the battery charge calculation module.

[0011] Based on the comparison between the battery's charge value and the output current value under the safe power limit, the supercapacitor's intervention point is determined. The output current value under the safe power limit is set to a maximum and a minimum value. Within the range of the maximum and minimum values, the photovoltaic system's battery supplies full power to the grid.

[0012] The charge of the supercapacitor is calculated using the supercapacitor charge calculation module.

[0013] The auxiliary current required for the supercapacitor is calculated using the supercapacitor power matching module.

[0014] Based on the charge and output of the supercapacitor, the supercapacitor is replenished with electrical energy in a timely manner.

[0015] When the charge of the supercapacitor is lower than the minimum set value, the mains power ratio module controls the power output control module to directly output the full power of the mains to supply the household power grid.

[0016] The photovoltaic system charges the battery, and then charges the supercapacitor. When the battery's charge reaches the discharge standard, the mains power is cut off and the battery resumes power supply.

[0017] Preferably, the supercapacitor and the intervention node use the battery charge and the power demand of the grid as key parameters of the control strategy.

[0018] The battery charge calculation module uses a fusion algorithm to calculate the battery's state of charge (SOC). The basic parameter used in the calculation is the battery's rated capacity Q. r Remaining battery capacity Q e The amount of electricity Q charged into or output by the battery u The specific formula is as follows;

[0019]

[0020] Among them, the amount of electricity Q that the battery charges or releases within time t. b When the charge / discharge current is equal to the integral of the charging / discharging current over time t, the charge value Q is calculated using the following formula. b ;

[0021]

[0022] Where η is the discharge efficiency;

[0023] When the initial state of charge of the battery is SOC i And when Q is not 1, at time t b The formula is as follows;

[0024]

[0025] Based on the above formula, the formula for battery SOC is as follows:

[0026]

[0027] When the battery's SOC value and the output current value are within the range of the safe power limit, the battery supplies the grid at full power. When the battery's SOC value is higher than the maximum output current value, the excess electrical energy is fed into the grid by the supercapacitor. When the battery's SOC value is lower than the minimum output current value, the electrical energy is fed into the grid at full power by the supercapacitor.

[0028] Preferably, when the battery discharges with current I within the output current range under the safe power limit, after a certain time t, the battery voltage changes from U1 to U2. At this time, the formula for the charging and discharging energy of the battery is as follows.

[0029]

[0030] C represents the battery capacity, and E represents the energy stored in the battery.

[0031] The energy consumed by the equivalent internal resistance R of the battery is E. R The formula is as follows;

[0032]

[0033] From the two formulas above, we can see that the battery discharge efficiency η dis and charging efficiency η cha The formula is as follows

[0034]

[0035] In the formula, τ=RC is the time constant, β=U1 / U2 during charging or β=U2 / U1 during discharging, and the depth of discharge DOD=1-β is defined.

[0036] Preferably, the formula for calculating the charge of the supercapacitor is as follows;

[0037]

[0038] In the formula, E is the energy stored in the supercapacitor, U is the voltage, and C is the capacitance.

[0039] Based on the above formula, the formula for calculating the state of charge (SOC) of a supercapacitor is as follows;

[0040]

[0041] In the formula, the operating voltage range of the supercapacitor is [U max U min ], U max U is the highest value. min The lowest value;

[0042] The energy corresponding to the terminal voltage U of a supercapacitor and its rated voltage U C The corresponding energy ratio (SOE) is:

[0043]

[0044] Preferably, when the supercapacitor receives the auxiliary current, a bidirectional DC / DC power converter is required. Assuming the auxiliary current is D, the supercapacitor's output voltage U... out With input voltage U in The ratio formula is:

[0045]

[0046] (1) is the input state, and (2) is the output state;

[0047] The auxiliary current supplied to the supercapacitor is calculated using the following formula.

[0048] I out U out =η DC I in Uin

[0049] In the formula, I in with I out η represents the input current and output current of the supercapacitor, respectively. DC This indicates the efficiency of the bidirectional DC / DC power converter.

[0050] Preferably, the fuzzy controller uses three input variables and a single output variable to perform fuzzy inference to calculate the ratio of the battery output power to the required power. The three input variables are the required power P. r State of charge (SOC) of supercapacitors u and battery state of charge (SOC) b The single output variable is the initial allocation coefficient of the battery output power.

[0051] Among them, the required power P r The fuzzy subset is divided into 7 levels, and the state of charge (SOC) of the supercapacitor is... u The fuzzy subset is divided into 3 levels, and the initial allocation coefficient of the battery output power is... The fuzzy subsets are divided into 5 levels.

[0052] Preferably, the supercapacitor has a state of charge (SOC). u With battery state of charge (SOC) b The actual universe of discourse and the fuzzy universe of discourse are the same, therefore the fuzzification factor K for both is the same. u and K b Both are constants of 1, and the required power P r The actual domain of discourse is [P] r-min P r-max The fuzzy universe of discourse is [-3,3]; the required power P r The actual domain of discourse is transformed into the symmetric interval [-P] of the equivalent demand power P. e P e The conversion formula is as follows;

[0053]

[0054] The fuzzification factor K for the equivalent demand power P is then... p 3 / P e The initial distribution factor of battery output power The actual universe of discourse is the same as the fuzzy universe of discourse, therefore its scaling factor for refinement is P. r / 1.

[0055] According to another aspect of the present invention, this technical solution also provides a power distribution strategy decision-making device for a photovoltaic-storage-DC-flexible system based on AI prediction, the device comprising:

[0056] The information acquisition module is used to obtain real-time charging and discharging parameters of the batteries and supercapacitors in the photovoltaic system;

[0057] The decision center calculates the sampled values ​​based on the captured sampled pulse signals, transmits the calculation results to the decision center, and outputs the decision through a fuzzy controller;

[0058] The battery charge calculation module is used to calculate the battery charge and control the output power through the battery power matching module.

[0059] The power output control module, controlled by the decision center, determines the intervention point of the supercapacitor based on the comparison between the battery's charge value and the output current value under the safe power limit. The output current value under the safe power limit is set with a maximum and a minimum value. Within the range of the maximum and minimum values, the photovoltaic system's battery supplies full power to the grid. When the supercapacitor's charge value is lower than the minimum set value, the mains power matching module controls the power output control module to directly output full mains power to supply the grid. The photovoltaic system charges the battery and then charges the supercapacitor. When the battery's charge value reaches the discharge standard, the mains power is cut off and the battery resumes power supply.

[0060] The supercapacitor charge calculation module is used to calculate the charge of the supercapacitor.

[0061] The supercapacitor power matching module is used to calculate the auxiliary current value that the supercapacitor needs to be supplied.

[0062] Based on the supercapacitor's charge and output, the supercapacitor is replenished with electrical energy in a timely manner.

[0063] (III) Beneficial Effects

[0064] Compared with existing technologies, the present invention provides a method and apparatus for decision-making on power distribution strategies for photovoltaic-storage-DC-flexible systems based on AI prediction, which has the following advantages:

[0065] In use, the decision-making method of this invention acquires real-time charging and discharging parameters of the batteries and supercapacitors within the photovoltaic system using an information acquisition module. This allows for real-time monitoring of the battery's discharge and charging parameters. The battery's charge level and the power demand of the grid are used as key parameters for the control strategy, serving as the intervention point for the supercapacitor. This ensures that within the maximum and minimum output current value range set under safe power limits, the photovoltaic system's batteries supply full power to the grid. Furthermore, when the output current value exceeds the safe power limit, the system calculates the sampled value based on the captured sampling pulse signal. The calculation result... The data is transmitted to the decision center, which uses a fuzzy controller to output decisions to determine the intervention point and amount of the supercapacitor. This enables the battery and supercapacitor to supply power together, ensuring that when power demand suddenly increases, the supercapacitor's intervention does not cause a large current surge to the battery. Furthermore, when the supercapacitor's charge level falls below the minimum set value, the mains power distribution module controls the power output control module to directly output the full power of the mains to supply the household grid. When the battery's charge level reaches the discharge standard, the mains power is cut off and the battery resumes power supply, thus ensuring a stable power supply to the household grid. Attached Figure Description

[0066] Figure 1 This invention provides a framework for the AI-based prediction-based power distribution strategy decision-making method for photovoltaic-storage-DC-flexible systems. Figure 1 ;

[0067] Figure 2 This invention provides a framework for the AI-based prediction-based power distribution strategy decision-making method for photovoltaic-storage-DC-flexible systems. Figure 2 ;

[0068] Figure 3 This is a block diagram of the power distribution strategy decision-making device for a photovoltaic-storage-DC-flexible system based on AI prediction proposed in this invention. Detailed Implementation

[0069] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] Example 1:

[0071] See attached document Figure 1-2A decision-making method for power distribution strategy of photovoltaic-storage-DC-flexible system based on AI prediction. This method includes: using an information acquisition module to obtain real-time charging and discharging parameters of batteries and supercapacitors in the photovoltaic system, and using the battery charge and the power demand of the grid as key parameters for the control strategy of the supercapacitor and the intervention node.

[0072] The sampled values ​​are calculated based on the captured sampled pulse signals, and the calculation results are transmitted to the decision center. The decision center then makes a decision output through the fuzzy controller.

[0073] The fuzzy controller uses three input variables and a single output variable to perform fuzzy inference to calculate the ratio of battery output power to demand power. The three input variables are the demand power P. r State of charge (SOC) of supercapacitors u and battery state of charge (SOC) b The single output variable is the initial distribution coefficient of the battery output power.

[0074] Among them, the required power P r The fuzzy subset is divided into 7 levels, and the state of charge (SOC) of the supercapacitor is... u The fuzzy subset is divided into 3 levels, and the initial allocation coefficient of the battery output power is... The fuzzy subsets are divided into 5 levels;

[0075] Supercapacitor State of Charge (SOC) u With battery state of charge (SOC) b The actual universe of discourse and the fuzzy universe of discourse are the same, therefore the fuzzification factor K for both is the same. u and K b Both are constants of 1, and the required power P r The actual domain of discourse is [P] r-min P r-max The fuzzy universe of discourse is [-3,3]; the required power P r The actual domain of discourse is transformed into the symmetric interval [-P] of the equivalent demand power P. e P e The conversion formula is as follows;

[0076]

[0077] The fuzzification factor K for the equivalent demand power P is then... p 3 / P e The initial distribution factor of battery output power The actual universe of discourse is the same as the fuzzy universe of discourse, therefore its scaling factor for refinement is P. r / 1;

[0078] The battery charge calculation module calculates the battery's charge capacity. This module uses a fusion algorithm to calculate the battery's State of Charge (SOC), with the battery's rated capacity (Q) as the basic parameter used in the calculation. r Remaining battery capacity Q w The amount of electricity Q charged into or output by the battery u The specific formula is as follows;

[0079]

[0080] Among them, the amount of electricity Q that the battery charges or releases within time t. b When the charge / discharge current is equal to the integral of the charging / discharging current over time t, the charge value Q is calculated using the following formula. b ;

[0081]

[0082] Where η is the discharge efficiency;

[0083] When the initial state of charge of the battery is SOC i And when Q is not 1, at time t b The formula is as follows;

[0084]

[0085] Based on the above formula, the formula for battery SOC is as follows:

[0086]

[0087] When the battery's SOC value and the output current value are within the range of the safe power limit, the battery supplies the grid at full power. When the battery's SOC value is higher than the maximum output current value, the excess electrical energy is fed into the grid by the supercapacitor. When the battery's SOC value is lower than the minimum output current value, the electrical energy is fed into the grid at full power by the supercapacitor.

[0088] When a battery discharges with current I within the output current range under safe power limits, after a certain time t, the battery voltage changes from U1 to U2. The formula for the charging and discharging energy of the battery at this time is as follows.

[0089]

[0090] C represents the battery capacity, and E represents the energy stored in the battery.

[0091] The energy consumed by the equivalent internal resistance R of the battery is E. R The formula is as follows;

[0092]

[0093] From the two formulas above, we can see that the battery discharge efficiency η dis and charging efficiency η cha The formula is as follows

[0094]

[0095] In the formula, τ=RC is the time constant, β=U1 / U2 during charging or β=U2 / U1 during discharging, and the depth of discharge DOD=1-β is defined at the same time;

[0096] Based on the comparison between the battery's charge value and the output current value under the safe power limit, the supercapacitor's intervention point is determined. The output current value under the safe power limit is set to a maximum and a minimum value. Within the range of the maximum and minimum values, the photovoltaic system's battery supplies full power to the grid.

[0097] The supercapacitor charge calculation module is used to calculate the supercapacitor charge, and the supercapacitor power ratio module is used to calculate the auxiliary current value required for the supercapacitor. The supercapacitor charge calculation formula is as follows.

[0098]

[0099] In the formula, E is the energy stored in the supercapacitor, U is the voltage, and C is the capacitance.

[0100] Based on the above formula, the formula for calculating the state of charge (SOC) of a supercapacitor is as follows;

[0101]

[0102] In the formula, the operating voltage range of the supercapacitor is [U max U min ], U max U is the highest value. min The lowest value;

[0103] The energy corresponding to the terminal voltage U of a supercapacitor and its rated voltage U C The corresponding energy ratio SOE is:

[0104]

[0105] When the supercapacitor is supplied with auxiliary current, a bidirectional DC / DC power converter is required. Assuming the supercapacitor current is D, the supercapacitor's output voltage U... out With input voltage U in The ratio formula is:

[0106]

[0107] 1 represents the input state, and 2 represents the output state;

[0108] The auxiliary current supplied to the supercapacitor is calculated using the following formula.

[0109] I out U out =η DC I in U in

[0110] In the formula, I in with I out η represents the input current and output current of the supercapacitor, respectively. DC Indicates the efficiency of a bidirectional DC / DC power converter;

[0111] Based on the supercapacitor's charge and output, the supercapacitor is replenished with power in a timely manner. When the supercapacitor's charge is lower than the minimum set value, the mains power output control module directly outputs the full power of the mains to supply the grid. The photovoltaic system charges the battery, and then the supercapacitor is charged. When the battery's charge reaches the discharge standard, the mains power is cut off and the battery resumes power supply.

[0112] In use, the decision-making method of this invention acquires real-time charging and discharging parameters of the batteries and supercapacitors within the photovoltaic system using an information acquisition module. This allows for real-time monitoring of the battery's discharge and charging parameters. The battery's charge level and the power demand of the grid are used as key parameters for the control strategy, serving as the intervention point for the supercapacitor. This ensures that within the maximum and minimum output current value range set under safe power limits, the photovoltaic system's batteries supply full power to the grid. Furthermore, when the output current value exceeds the safe power limit, the system calculates the sampled value based on the captured sampling pulse signal. The calculation result... The data is transmitted to the decision center, which uses a fuzzy controller to output decisions to determine the intervention point and amount of the supercapacitor. This enables the battery and supercapacitor to supply power together, ensuring that when power demand suddenly increases, the supercapacitor's intervention does not cause a large current surge to the battery. Furthermore, when the supercapacitor's charge level falls below the minimum set value, the mains power distribution module controls the power output control module to directly output the full power of the mains to supply the household grid. When the battery's charge level reaches the discharge standard, the mains power is cut off and the battery resumes power supply, thus ensuring a stable power supply to the household grid.

[0113] Example 2:

[0114] See attached document Figure 3 A decision-making device for power distribution strategy based on AI prediction of photovoltaic-storage-DC-flexible system includes:

[0115] The information acquisition module is used to obtain real-time charging and discharging parameters of the batteries and supercapacitors in the photovoltaic system;

[0116] The decision center calculates the sampled values ​​based on the captured sampled pulse signals, transmits the calculation results to the decision center, and outputs the decision through a fuzzy controller;

[0117] The battery charge calculation module is used to calculate the battery charge and control the output power through the battery power matching module.

[0118] The power output control module, controlled by the decision center, determines the intervention point of the supercapacitor based on the comparison between the battery's charge value and the output current value under the safe power limit. The output current value under the safe power limit is set with a maximum and a minimum value. Within the range of the maximum and minimum values, the photovoltaic system's battery supplies full power to the grid. When the supercapacitor's charge value is lower than the minimum set value, the mains power matching module controls the power output control module to directly output full mains power to supply the grid. The photovoltaic system charges the battery and then charges the supercapacitor. When the battery's charge value reaches the discharge standard, the mains power is cut off and the battery resumes power supply.

[0119] The supercapacitor charge calculation module is used to calculate the charge of the supercapacitor.

[0120] The supercapacitor power matching module is used to calculate the auxiliary current value that the supercapacitor needs to be supplied.

[0121] Based on the supercapacitor's charge and output, the supercapacitor is replenished with electrical energy in a timely manner.

[0122] In use, this invention acquires real-time charging and discharging parameters of the battery and supercapacitor within the photovoltaic system via an information acquisition module. Based on a comparison between the battery's charge value and the output current value under the safety power limit, the decision center determines the supercapacitor's intervention point. The output current value under the safety power limit is set to a maximum and a minimum value. Within the range of the maximum and minimum values, the photovoltaic system's battery supplies full power to the grid. When the supercapacitor's charge value is lower than the minimum set value, the mains power distribution module controls the power output control module to directly output full mains power to supply the grid, allowing the photovoltaic system to charge the battery, and then charge the supercapacitor. When the battery's charge value reaches the discharge standard, the mains power is cut off, and the battery resumes power supply.

[0123] It should be noted that the term "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0124] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An AI prediction-based power distribution strategy decision method for a light storage direct flexible system, characterized in that, The method includes: The information acquisition module is used to obtain real-time charging and discharging parameters of the batteries and supercapacitors in the photovoltaic system. The sampled values ​​are calculated based on the captured sampled pulse signals, and the calculation results are transmitted to the decision center, which then makes a decision output through a fuzzy controller. The fuzzy controller uses three input variables and a single output variable to perform fuzzy inference to calculate the ratio of battery output power to demand power. The three input variables are the demand power... State of charge of supercapacitors and battery state of charge The single output variable is the initial allocation coefficient of the battery output power. ; Among them, the required power The fuzzy subset is divided into 7 levels, and the state of charge of the supercapacitor is... The fuzzy subset is divided into 3 levels, and the initial allocation coefficient of the battery output power is... The fuzzy subsets are divided into 5 levels; The state of charge of the supercapacitor With battery state of charge The actual domain and the fuzzy domain are the same, therefore the quantization factor of both is fuzzy. and Both are constants of 1, and the required power is... The actual domain of discourse is [ , The fuzzy domain is [-3,3]; the required power The actual domain of discourse is converted into equivalent demand power. The symmetric interval [- , The conversion formula is as follows; Then the equivalent power demand fuzzy quantization factor for The initial distribution coefficient of battery output power The actual universe of discourse is the same as the fuzzy universe of discourse, therefore its refinement scaling factor is ; The battery charge is calculated using the battery charge calculation module. Based on the comparison between the battery's charge value and the output current value under the safe power limit, the supercapacitor's intervention point is determined. The output current value under the safe power limit is set to a maximum and a minimum value. Within the range of the maximum and minimum values, the photovoltaic system's battery supplies full power to the grid. The charge of the supercapacitor is calculated using the supercapacitor charge calculation module. The auxiliary current required for the supercapacitor is calculated using the supercapacitor power matching module. When the supercapacitor is supplied with auxiliary current, a bidirectional DC / DC power converter is required. Assuming the excess current is... Output voltage of supercapacitor With input voltage The ratio formula is: (1) is the input state, and (2) is the output state; The auxiliary current supplied to the supercapacitor is calculated using the following formula. In the formula, These represent the input current and output current of the supercapacitor, respectively. Indicates the efficiency of a bidirectional DC / DC power converter; Based on the charge and output of the supercapacitor, the supercapacitor is replenished with electrical energy in a timely manner. When the charge of the supercapacitor is lower than the minimum set value, the mains power ratio module controls the power output control module to directly output the full power of the mains to supply the household power grid. The photovoltaic system charges the battery, and then charges the supercapacitor. When the battery's charge reaches the discharge standard, the mains power is cut off and the battery resumes power supply.

2. The power distribution strategy decision-making method for a photovoltaic-storage-DC-flexible system based on AI prediction as described in claim 1, characterized in that: The supercapacitor and its intervention node use the battery charge and the power demand of the grid as key parameters for the control strategy. The battery charge calculation module uses a fusion algorithm to calculate the battery's charge value. The basic parameter used in the calculation is the rated capacity of the battery. Remaining battery capacity and the amount of electricity charged into or output by the battery The specific formula is as follows; = Among them, the amount of electricity charged or released by the battery within time t. When the charge / discharge current is equal to the integral of the charging / discharging current over time t, the charge value is calculated using the following formula. ; in, For discharge efficiency; When the initial nuclear power state of the battery is And when it is not 1, at time t The formula is as follows; Based on the above formula, the battery can be obtained. The formula is as follows; In the storage battery When the output current value is within the range of numerical and safety-limited power, the battery supplies full power to the household power grid. When the value exceeds the maximum output current value, the excess electrical energy is fed into the grid by the supercapacitor and stored in the battery. When the value is lower than the minimum output current value, the electrical energy is supplied to the grid at full power by the supercapacitor.

3. The power distribution strategy decision-making method for a photovoltaic-storage-DC-flexible system based on AI prediction as described in claim 2, characterized in that: The battery outputs current within the range of safe power limits. During discharge, after a certain time t, the battery voltage changes from... Become The formula for the charging and discharging energy of the battery at this time is as follows; For battery capacity, To store energy in the battery; The energy consumed by the equivalent internal resistance R of the battery The formula is as follows; From the two formulas above, we can see that the battery discharge efficiency and charging efficiency The formula is as follows: In the formula, It is a time constant, during charging. or during discharge Meanwhile, the depth of discharge is defined. .

4. The power distribution strategy decision-making method for a photovoltaic-storage-DC-flexible system based on AI prediction as described in claim 1, characterized in that: The formula for calculating the charge of the supercapacitor is as follows; In the formula, E is the energy stored in the supercapacitor, U is the voltage, and C is the capacitance. Based on the above formula, the formula for calculating the state of charge (SOC) of a supercapacitor is as follows; In the formula, the operating voltage range of the supercapacitor is [ , ], The highest value, The lowest value; The terminal voltage of the supercapacitor is The energy corresponding to the rated voltage The corresponding energy ratio ( )for; 。 5. A power distribution strategy decision-making device for a photovoltaic-storage-DC-flexible system based on AI prediction, used in the method of any one of claims 1-4, characterized in that, The device includes; The information acquisition module is used to obtain real-time charging and discharging parameters of the batteries and supercapacitors in the photovoltaic system. The decision center calculates the sampled values ​​based on the captured sampled pulse signals, transmits the calculation results to the decision center, and outputs the decision through the fuzzy controller; The battery charge calculation module is used to calculate the battery charge and control the output power through the battery power matching module. The power output control module, controlled by the decision center, determines the intervention point of the supercapacitor based on the comparison between the battery's charge value and the output current value under the safe power limit. The output current value under the safe power limit is set with a maximum and a minimum value. Within the range of the maximum and minimum values, the photovoltaic system's battery supplies full power to the grid. When the supercapacitor's charge value is lower than the minimum set value, the mains power matching module controls the power output control module to directly output full mains power to supply the grid. The photovoltaic system charges the battery and then charges the supercapacitor. When the battery's charge value reaches the discharge standard, the mains power is cut off and the battery resumes power supply. The supercapacitor charge calculation module is used to calculate the charge of the supercapacitor. The supercapacitor power matching module is used to calculate the auxiliary current value that the supercapacitor needs to be supplied. Based on the supercapacitor's charge and output, the supercapacitor is replenished with electrical energy in a timely manner.

Images