Control method of energy storage photovoltaic energy system, photovoltaic energy storage equipment and energy storage photovoltaic energy system

Abstract

The invention discloses a control method of an energy storage photovoltaic energy system, photovoltaic energy storage equipment and the energy storage photovoltaic energy system. According to the control method, the electricity consumption and the electricity generation amount in a period of time in the future are predicted, a curve graph of electricity price along with time in the period of time is established, all wave crest and wave trough information in the curve graph is found out, and a wave crest and wave trough object set is established; selecting a to-be-processed wave peak and a to-be-processed wave trough, if the electricity price of the to-be-processed wave peak is greater than that of the to-be-processed wave trough, carrying out wave trough processing logic, and updating the electricity purchasing quantity and the wave peak net generating capacity of the entity corresponding to the index of the to-be-processed wave peak in the wave peak and wave trough object set; until all the wave crests are processed. According to the control method, the electricity demand of the user is moved to the position with the low electricity price step by step to be stored in advance, the electricity demand of the high-order electricity price is moved to the low-order electricity price through multi-iteration peak clipping and valley filling and electricity demand transfer, and economical efficiency maximization is achieved.

Description

Technical Field

[0001] The present invention relates to a control method for an energy storage photovoltaic energy system, in particular to an intelligent control method based on a "three monks" AI large model, as well as a photovoltaic energy storage device and an energy storage photovoltaic energy system adopting the control method. Background Art

[0002] In recent years, the rapid development of artificial intelligence (AI) technology is profoundly changing all walks of life, especially in the field of energy management, showing unprecedented potential. Especially in the inverter market, with the growing demand for green energy and sustainable development, intelligent management has become the core driving force for improving system efficiency and optimizing energy configuration. For the energy storage application market segment, users are increasingly demanding the intelligent and efficient use of energy storage resources to reduce energy consumption costs.

[0003] At present, for energy storage photovoltaic energy systems, the algorithms and applications used in the industry are mainly focused on simple control based on the maximum and minimum values ​​of electricity prices, directly charging in the minimum value range and discharging in the maximum value range, without maximizing economic benefits. When the electricity price curve fluctuates multiple times or is not a simple sine function, there will be a relatively large economic difference from the optimal solution. Existing control methods have not yet fully utilized the deep learning capabilities of big data and AI algorithms. For example, complex factors such as the specific electricity price policy in the user's area, the impact of geographical location on solar power generation, and the historical electricity consumption behavior patterns of the surrounding environment may be ignored. This limitation leads to low prediction accuracy and insufficient refinement of energy storage resource allocation, which cannot fully meet the user's optimal cost saving goals in a specific period of time.

[0004] The above information disclosed in the Background section is only for enhancement of understanding of the background of the present application and therefore it may contain information that does not constitute the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0005] The present invention provides a control method for an energy storage photovoltaic energy system to solve the deficiencies of energy storage inverter users in efficiently utilizing energy storage resources, ensure that users maximize the potential of the energy storage photovoltaic energy system, efficiently and optimally cope with the uncertainty and severe fluctuations in electricity prices, and move electricity demand from high electricity prices to low electricity prices, thereby maximizing economic efficiency.

[0006] The present invention also provides a photovoltaic energy storage device and an energy storage photovoltaic energy system adopting the control method.

[0007] In order to achieve the above object, the present invention adopts the following technical scheme:

[0008] A control method for an energy storage photovoltaic energy system comprises the following steps:

[0009] S100, predicting the power consumption and power generation in a future period of time, establishing a curve chart of the electricity price over time in the period of time, finding all the peak and trough information in the curve chart, and establishing a peak and trough object set;

[0010] S200, setting the peak to be processed, and setting the last peak as the peak to be processed during the first processing;

[0011] S300, taking the trough that is smaller in time than the to-be-processed peak and closest to it as the to-be-processed trough;

[0012] S400, if the electricity price of the peak to be processed is greater than the electricity price of the trough to be processed, perform trough processing logic, and update the purchased electricity and peak net power generation of the peak and trough object information corresponding to the index of the peak to be processed in the peak and trough object set;

[0013] S500, determining whether the net power generation of the peak to be processed is 0;

[0014] If the result is yes, the peak to be processed is marked as used, and the peaks with higher electricity prices on the left and right sides of the peak are taken as new peaks to be processed, and the following step S600 is executed;

[0015] If the result is no, the left and right sides of the trough to be processed with lower electricity prices are taken as new troughs to be processed; if the index of the new trough to be processed in the peak and trough object set is not equal to the index of the peak to be processed in the peak and trough object set, repeat step S400; if the index of the new trough to be processed is equal to the index of the peak to be processed, execute the following step S600;

[0016] S600, determining whether the new peak to be processed is the first peak;

[0017] If the result is yes, then end the program;

[0018] If the answer is no, then step S200 is repeated.

[0019] In a preferred embodiment, in step S100, the current power of the energy storage photovoltaic energy system is put into the earliest time data in the peak and trough object set as the initial power.

[0020] In a preferred embodiment, in step S100, information is initialized according to the predicted data, and the initialized information is sorted according to time, and the initialized information includes the expected power generation, power consumption, net power generation, maximum hourly charge capacity of the battery, purchasable power, excess net power generation, dischargeable capacity, initial power, initial purchased power and battery capacity for each time interval; wherein, if the current power is less than or equal to 0, the initial power is 0.

[0021] In a more preferred embodiment, in step S100, the peak and trough object sets use different memory addresses, and the peak and trough object sets include multiple ones of power station ID, inverter serial number, capacity, time, electricity price, expected power generation, expected power consumption, net power generation, out-of-range power generation, purchasable power, dischargeable power, initial battery power SOC, purchasable power, expected battery power, maximum charge per hour, charge and discharge plan (still, charging, discharging), cost, and initial cost; find the indexes of all peaks and troughs according to the peak and trough object sets, and put them into the peak index set and the trough index set.

[0022] In a preferred embodiment, step S400 includes the following steps:

[0023] S401, judging whether the electricity price of the peak to be processed is greater than the electricity price of the trough to be processed, and whether the trough state is normal; if the result is no, re-finding the peak and trough to be processed; if the result is yes, executing the following steps S410 to S411;

[0024] S410, finding the closest unused peaks on both sides of the peak to be processed;

[0025] S411, determining whether the net power generation of the peak to be processed is greater than or equal to 0;

[0026] If the result is yes, mark the peak as used, and compare the peaks on both sides found in step S410 to see which has the highest electricity price. Set the peak with the highest electricity price as the new current processing peak, and repeat step S401; if the result is no, perform the trough processing logic.

[0027] In a more preferred embodiment, step S400 includes the following steps:

[0028] S420: whether to perform trough processing; if trough processing is performed, execute the following step S421;

[0029] S421, determine whether the net power generation of the trough currently processed is greater than 0; if the result is yes, execute the following step S422; if the result is no, jump to the following step S424;

[0030] S422, determining whether the net power generation of the to-be-processed trough is greater than the net power generation required by the to-be-processed peak; if the result is yes, executing steps S423 to S424;

[0031] S423, updating the data of the valley net power generation and the peak power generation to be processed, and the peak net power generation is 0;

[0032] S424: If the peak to be processed still needs electricity, determine whether there is available capacity for purchase in the trough to be processed; if the result is yes, execute the following step S425;

[0033] S425, determining whether the purchasable capacity of the to-be-processed trough can meet the power demand of the to-be-processed peak;

[0034] If the result is yes, then execute step S426: set the net power generation of the peak to be processed to 0, and update the data of the available trough power, the net trough power, the peak power purchase and the peak power to be processed;

[0035] If the result is no, execute step S427: set the purchasable electricity amount of the to-be-processed trough to 0, and update the data of the peak net power generation, trough net power generation, trough purchased electricity and the to-be-processed peak processed electricity.

[0036] In a more preferred embodiment, step S400 includes the following steps after step S426 or step S427:

[0037] S430, determine whether the conditions are met: the power generation is negative, there is power purchase and it is a peak; if the result is yes, execute the following step S431; if the result is no, execute the following step 435;

[0038] S431, determining whether the peak net power generation has changed, if it has changed, executing the following step S432, if it has not changed, executing the following step S434;

[0039] S432, determine whether the peak change value is greater than the peak purchase amount; if the result is yes, execute the following step S433; if the result is no, update the sum of the peak purchase amount and the peak change value to the new peak purchase amount, and execute the following step S434;

[0040] S433, clearing the peak purchase amount;

[0041] S434, updating the purchase quantity of the entity corresponding to the peak index in the peak and trough object collection;

[0042] S435, updating the peak and trough object sets, trough net power generation, trough purchasable power, trough purchasable power and peak net power generation;

[0043] S436. If the trough purchase amount is not 0, it is recorded as the net power generation value to be processed for peak processing.

[0044] In a further preferred embodiment, step S400 includes the following steps after step S436:

[0045] S440, updating initial battery power and expected battery power information;

[0046] S441, passing in the current peak and trough object set and the current trough index;

[0047] S442, if the index of the currently processed trough is greater than or equal to 0, cyclically update the data starting from the trough index;

[0048] S443, determine whether the purchased power is greater than 0; if greater than 0, execute the following step S444; if equal to 0, update the expected battery power = expected battery power = current battery power + purchased power + expected power generation - expected purchased power, and then execute the following step S450;

[0049] S444. If the difference between the expected power generation and the expected purchase power is greater than 0, then update the expected battery power = current battery power + purchase power + expected power generation - expected purchase power, and then execute step S450; if it is not greater than 0, then update the expected battery power = current battery power + purchase power, and then execute step S450;

[0050] S450, determining whether the expected battery power is greater than the battery capacity;

[0051] If the result is no, update the expected battery capacity to 0;

[0052] If the result is yes, and the processed data is the data at the trough or next to the trough currently being processed, then mark this round of beam processing as completed, and set the expected battery capacity equal to the battery capacity;

[0053] S451, determine whether the current battery power + the available power is greater than the battery capacity; if the result is yes, execute the following step S452; if the result is no, jump to step S457;

[0054] S452, set excess capacity = current battery power + available power - battery capacity;

[0055] S453, determine whether the excess capacity is greater than the available power; if the result is yes, clear the available power and jump to step S457; if the result is no, execute the following step S454;

[0056] S454, subtracting the excess capacity from the available electricity to update the available electricity;

[0057] S455, determine whether the original purchased power is greater than the battery capacity minus the current battery power; if the result is positive, execute the following step S456; if the result is negative, jump to step S457;

[0058] S456, update the purchased power=battery capacity-current battery power;

[0059] S457: Update the expected power, and the initial power of the object with the current index + 1 is the expected power of the current data.

[0060] In a more preferred embodiment, in step S422, if the result is no, the data is updated, the net trough power generation after the update is 0, the net peak power generation after the update = the net peak power generation before the update + the net trough power generation, the peak power processing amount before the update = the peak power processing amount after the update - the net trough power generation, and then jump to step S424.

[0061] In a more preferred embodiment, in step S423, the net trough power generation after update = the net trough power generation before update + the net peak power generation; the peak power processing amount after update = the peak power processing amount before update + the net peak power generation.

[0062] In a preferred embodiment, in step S400, the expected power is updated and the expected power is used as the initial power of the peak and valley object information of the current index +1.

[0063] In a preferred embodiment, before step S600, the trough processing logic is firstly jumped out and the next peak processing logic is entered.

[0064] In a preferred embodiment, the future period is 1 to 7 days in the future, and the future period is divided into multiple time segments, each of which has a duration of 10 to 60 minutes. More preferably, each time segment has a duration of 10 to 30 minutes. In a specific and preferred embodiment, each time segment has a duration of 15 minutes.

[0065] Another technical solution adopted by the present invention is as follows:

[0066] A photovoltaic energy storage device, comprising:

[0067] The controller is used to receive the prediction data generated by the server according to the control method, and control the charging and discharging time and charging and discharging power of the battery according to the prediction data.

[0068] In a preferred embodiment, a command to turn on or off the intelligent control is input through a user interface;

[0069] After receiving the instruction, the controller of the user terminal or photovoltaic energy storage device generates an instruction data packet sent to the server, and the instruction data packet includes a user ID, a device ID, and an enabled or disabled state of the intelligent control.

[0070] Another technical solution adopted by the present invention is as follows:

[0071] A photovoltaic energy storage system, comprising:

[0072] A server, which is used to receive a command data packet sent by a photovoltaic energy storage device, wherein the command data packet includes a user ID, a device ID, and an enable or disable status of the intelligent control;

[0073] The server also obtains instructions from the instruction data packet, and after confirming that the intelligent control is turned on, generates prediction data according to the device ID based on the device's geographical location, historical power generation and power consumption data, and regional electricity price policy, and sends it to the photovoltaic energy storage device by the control method as described. The prediction data includes the power generation and power consumption, charging and discharging time, and charging and discharging power in the future period of time.

[0074] The above solution adopted by the present invention has the following advantages:

[0075] The control method of the energy storage photovoltaic energy system of the present invention utilizes the "carrying water" mode in the algorithm to move the user's electricity demand step by step to a location with low electricity prices for advance storage. After multiple iterations of peak shaving and valley filling and electricity demand transfer, it can efficiently and optimally cope with the uncertainty and violent jitter of the electricity price model fluctuation, smoothly cope with multiple peaks and valleys one by one, and move the electricity demand of high electricity prices to low electricity prices to maximize economic efficiency. This control method realizes intelligent electricity fee optimization and significant cost savings; realizes accurate prediction and efficient energy storage management; realizes personalized strategies and strong adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] In order to more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0077] Figures 1 to 8 A flow chart of a control method according to an embodiment of the present invention is shown.

[0078] Fig. 9 The electricity fee statistics diagram of the control method according to the embodiment of the present invention and the traditional method are shown. DETAILED DESCRIPTION

[0079] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0080] This embodiment is dedicated to building a highly intelligent energy storage photovoltaic energy management system, which includes photovoltaic energy storage equipment or controls one or more photovoltaic energy storage equipment. Photovoltaic energy storage equipment can be power equipment such as photovoltaic inverters and converters, in which the batteries have energy storage functions, can store electricity from photovoltaic panels, purchase electricity from the power grid when needed, and store the purchased electricity in the battery for use. The energy storage photovoltaic energy management system also has a server for predicting the power generation and power consumption of photovoltaic energy storage equipment in the future based on the "Three Monks" AI large model, and customizing control strategies. The control strategies include but are not limited to optimal charging and discharging time, charging and discharging power, etc.

[0081] The control method of the energy storage photovoltaic energy management system integrates the electricity price policy of the user's country, the precise latitude and longitude information of the deployed equipment, and the historical electricity consumption and power generation big data of the surrounding environment. Through the machine learning model, these multivariate data are comprehensively analyzed to achieve high-precision prediction of electricity demand in the next 1-7 days. On this basis, it further provides users with personalized electricity consumption and energy storage strategy recommendations to ensure that the energy storage power is used most efficiently at the most critical moment, thereby achieving significant cost savings in subtle ways and improving user experience and satisfaction.

[0082] This control method utilizes the "water-carrying" mode in the algorithm to move the user's electricity demand step by step to a location with low electricity prices for advance storage. After multiple iterations of peak shaving and valley filling and electricity demand transfer, it can efficiently and optimally respond to the uncertainty and severe jitter of electricity price model fluctuations, smoothly respond to multiple peaks and valleys one by one, and move electricity demand at high electricity prices to low electricity prices, thereby maximizing economic efficiency.

[0083] Initial information used in this control method (one record per hour):

[0084] Estimated power generation: The large model calculates the estimated power generation * discharge loss rate;

[0085] Power consumption: large model calculates power consumption;

[0086] Net power generation: estimated power generation - estimated power consumption;

[0087] Maximum charge per hour: If it is less than or equal to 0, it is considered to be 0;

[0088] The amount of electricity that can be purchased in the interval: when the maximum charging amount per hour is greater than 0, (maximum charging amount per hour * charging loss rate) - net power generation (if the net power generation is less than or equal to 0, the net power generation will not be reduced); when the maximum charging amount per hour is 0, the amount of electricity that can be purchased in the interval is 0;

[0089] Exceeding net power generation: the net power generation is greater than the maximum charge per hour, and the excess value will be included in the excess net power generation;

[0090] Dischargeable capacity: maximum power consumption per hour * discharge loss rate;

[0091] Initial power: 0;

[0092] Initial purchase amount: 0;

[0093] Battery capacity: Find the result according to the model.

[0094] Prediction logic: Based on the electricity price policy of the user's country, the precise latitude and longitude information of the deployed equipment, and the historical electricity consumption and power generation big data of the surrounding environment, the system can comprehensively analyze these multivariate data through machine learning models to achieve high-precision prediction of electricity demand for the next 1-7 days (one record every 15 minutes).

[0095] 1. Create two dimensions of peaks and troughs based on the forecast data (peak: higher than the electricity prices on both sides is the peak; trough: lower than the electricity prices on both sides is the trough); find all the peak and trough information based on the forecast data. Put the current power obtained by the heavy equipment as the initial power into the earliest time data in the peak and trough object collection.

[0096] 2. Set the peak information of this processing as the last one in the time range, and obtain the closest trough with a smaller time than the peak as the trough of this round of processing.

[0097] 2.1 Determine whether the peak price is greater than the trough price, and whether the trough state is normal:

[0098] a) If the conditions are not met, find the peak and trough to be processed again, that is, the peak closest to the left of the current peak and the trough closest to the peak time smaller than the peak time are the troughs to be processed in this round.

[0099] b) Find the forecast data closest to the peak in time and unused (all are unused by default)

[0100] c) Determine whether the current peak net power generation is greater than or equal to 0. If it is greater than or equal to 0, mark the peak as used, and compare the unused peaks on both sides to see which has a higher electricity price. Obtain the corresponding detailed information and set it as the new peak. At the same time, re-trigger 2.1. If it is less than 0, enter the valley processing logic

[0101] 2.2 Get whether to enter the trough processing process:

[0102] a) If the trough processing logic is not entered, the next peak and trough are searched again, that is, the closest peak to the left of the current peak and the closest trough with a smaller time than the peak are taken as the trough of this round of processing and the logic of 2.0 is re-entered.

[0103] b) Enter the trough data processing process

[0104] The current trough net power generation is greater than 0:

[0105] i. Whether the net power generation at the trough is greater than the net power generation required at the peak:

[0106] Yes, update data:

[0107] Net power generation at trough = Net power generation at trough + Net power generation at peak

[0108] This is the peak power processing = This is the peak power processing + peak net power generation

[0109] Peak net power generation = 0

[0110] No, update the data and:

[0111] Peak net power generation = peak net power generation + trough net power generation;

[0112] The current peak power generation = the current peak power generation - the current valley net power generation

[0113] Net power generation at trough = 0

[0114] d) If the current trough net power generation is less than or equal to 0 or the updated trough net power generation is greater than 0, it is determined whether there is purchasable capacity in the trough and whether electricity is still needed in the peak:

[0115] Yes, and the capacity that can be purchased during the trough period can meet the demand during the peak period. Update data:

[0116] Valley electricity purchase amount = Valley electricity purchase amount + Peak net power generation

[0117] Net power generation at trough = Net power generation at trough + Net power generation at peak

[0118] Peak purchase amount = Peak purchase amount + | Peak net power generation |

[0119] The current peak power processing = the current peak power processing + the peak net power generation = 0;

[0120] No, and the capacity that can be purchased during the trough period cannot solve the demand during the peak period. Update data:

[0121] Peak net power generation = peak net power generation + trough power available for purchase

[0122] Net power generation at trough = Net power generation at trough - Power available for purchase at trough

[0123] Power purchased at trough = Power purchased at trough + |Power available for purchase at trough|

[0124] The current peak power processing = the current peak power processing - the trough power available for purchase

[0125] The amount of electricity available for purchase at the trough = 0.00.

[0126] e) The scenario where there is available capacity for purchase during the trough period and electricity is still needed during the peak period, or there is available capacity for purchase after processing and electricity is still needed during the peak period:

[0127] The net power generation is negative and there is power purchase and it is a peak:

[0128] Yes, whether the peak net power generation has changed:

[0129] The peak change value is greater than the peak purchase power:

[0130] Yes: Peak purchase power is cleared

[0131] No: Peak purchase amount = Peak purchase amount (negative number) + Peak change value No, whether the peak net power generation has changed. Or the net power generation is negative and there is purchase amount and the peak processing is completed:

[0132] Update the purchased electricity of the entity corresponding to the peak index in the peak and trough collections, update the peak and trough collections, the trough net power generation, the trough purchase power, the trough purchase power peak net power generation.

[0133] If the amount of electricity purchased for the peak is not 0, record in the trough information how much net electricity was processed for that peak.

[0134] f) Update information such as initial power, expected battery power, etc.

[0135] Get the current peak and trough object collection and the current trough index (just update the trough index and all subsequent initial power information)

[0136] i. Whether the trough index is greater than or equal to 0: Re-trough index starts to update data in a loop

[0137] ii. Is the purchased electricity greater than 0?

[0138] Greater than 0, the expected power generation minus the expected power purchase is greater than 0;

[0139] Expected battery power = current battery power + purchased power + expected power generation - expected purchased power.

[0140] Greater than 0, the expected power generation minus the expected power purchase is less than or equal to 0;

[0141] Expected battery power = current battery power + purchased power.

[0142] iii. The amount of electricity purchased is equal to 0:

[0143] Expected battery power = current battery power + purchased power + expected power generation - expected purchased power

[0144] g) Whether the expected battery power is greater than the battery capacity and the expected battery capacity is less than 0:

[0145] Update expected battery capacity = 0;

[0146] Expected battery charge to be greater than battery capacity:

[0147] If the processed data happens to be the current trough or the data next to the trough, the processing of this round of trough is marked as completed;

[0148] Set expected battery capacity = battery capacity.

[0149] h) Current battery power + available power is greater than battery capacity

[0150] Exceeded capacity = current battery power + available power - battery capacity

[0151] Excess capacity is greater than the amount of electricity that can be purchased:

[0152] Yes: Clear available power

[0153] No: Available electricity = Available electricity - Exceeded capacity

[0154] The original purchase power is greater than the battery capacity minus the current battery power:

[0155] Yes: Purchase power = battery capacity - current battery power.

[0156] i) Update the expected power, and the initial power of the object with the current index + 1 is the expected power of the current data. j) If the net power generation of the trough is less than or equal to 0 and the available power is less than or equal to 0, mark the trough as used. k) If the net power generation of the peak is 0,

[0157] 0: Mark the peak as used, compare the indexes on the left and right sides that have been found before, see which one has a higher electricity price as the new peak index, and update the peak index to the one with a higher price

[0158] Others: Find unused indexes on the left and right sides of the trough, find the one with a lower price as the next trough, and update the trough index to the one with a lower price.

[0159] i) If there is a new trough, re-enter logic 2.2;

[0160] j) If the new trough index is equal to the peak index, jump out of the trough logic, enter the next peak, update the peak and trough information and re-enter logic 2;

[0161] In other cases, the trough logic is jumped out and the next peak is entered. The peak and trough information is updated and the 2 logic is re-entered until all peaks are processed.

[0162] Combination Figures 1 to 8 As shown, the control method of the energy storage photovoltaic energy system of this embodiment includes the following steps:

[0163] S100, predicting the power consumption and power generation in a future period of time, establishing a curve chart of the electricity price over time in the period of time, finding all the peak and trough information in the curve chart, and establishing a peak and trough object set;

[0164] S200, setting the peak to be processed, and setting the last peak as the peak to be processed during the first processing;

[0165] S300, taking the trough that is smaller in time than the to-be-processed peak and closest to it as the to-be-processed trough;

[0166] S400, if the electricity price of the peak to be processed is greater than the electricity price of the trough to be processed, perform trough processing logic, and update the purchased electricity and peak net power generation of the peak and trough object information corresponding to the index of the peak to be processed in the peak and trough object set;

[0167] S500, determining whether the net power generation of the peak to be processed is 0;

[0168] If the result is yes, the peak to be processed is marked as used, and the peaks with higher electricity prices on the left and right sides of the peak are taken as new peaks to be processed, and the following step S600 is executed;

[0169] If the result is no, the left and right sides of the trough to be processed with lower electricity prices are taken as new troughs to be processed; if the index of the new trough to be processed in the peak and trough object set is not equal to the index of the peak to be processed in the peak and trough object set, repeat step S400; if the index of the new trough to be processed is equal to the index of the peak to be processed, execute the following step S600;

[0170] S600, determining whether the new peak to be processed is the first peak;

[0171] If the result is yes, then end the program;

[0172] If the answer is no, then step S200 is repeated.

[0173] The specific process of step S100 is as follows: determine the initial battery power of the photovoltaic energy storage device. If it is less than or equal to 0, modify the input initial power to 0; if it is greater than 0, initialize other information according to the prediction data set, and store the data according to time. The prediction data set includes: expected power generation, power consumption, net power generation, maximum charge per hour, interval purchase power, excess net power generation, dischargeable power, initial power, initial purchase power and battery capacity.

[0174] Create peak and trough object sets based on the forecast data set, and use different memory addresses. The elements of the peak and trough object sets include: power station ID, inverter serial number, capacity, time, electricity price, expected power generation, expected power consumption, net power generation, out-of-range power generation, available power, available discharge, initial battery power SOC, available power, expected battery power, maximum charge per hour, charge and discharge plan (no change, charge, discharge), cost, and initial cost.

[0175] According to the peak and trough object sets, find all peak and trough indexes, put them into the peak index set and trough index set, corresponding to the positions of peaks and troughs in the peak and trough object sets respectively. The elements of the peak index set specifically include: the power station ID of the peak, the inverter serial number, the capacity, the time, the electricity price, the expected power generation, the expected power consumption, the net power generation, the power generation beyond the range, the power that can be purchased, the power that can be discharged, the initial battery power SOC, the power that can be purchased, the expected battery power, the maximum charge per hour, the charge and discharge plan (unchanged, charging, discharging), the cost, and the initial cost; the elements of the trough index set specifically include: the power station ID of the trough, the inverter serial number, the capacity, the time, the electricity price, the expected power generation, the expected power consumption, the net power generation, the power generation beyond the range, the power that can be purchased, the power that can be discharged, the initial battery power SOC, the power that can be purchased, the expected battery power, the maximum charge per hour, the charge and discharge plan (unchanged, charging, discharging), the cost, and the initial cost.

[0176] Put the initial power into the earliest time data in the peak and trough object collections.

[0177] In step S200, obtain the peak and trough object set, the peak index in the current processing peak index set, the peak index set, and the trough index combination; confirm whether the incoming peak index exists, if not, end the program; if it exists, jump to step S300. In the first processing, the index of the last peak is passed in as the index of the peak to be processed.

[0178] In step S300, according to the peak index passed in, the data of the peak in the peak and trough object set is found, the nearest trough index on the left is found according to the peak index, and then the data of the trough is found in the peak and trough object set according to the trough index. The peak index and trough index to be processed this time are initialized to the indexes corresponding to the peaks and troughs found above, and the process jumps to step S400.

[0179] Step S400 specifically includes:

[0180] S401, determine whether the electricity price of the peak to be processed (currently processed peak) is greater than the trough to be processed (currently processed trough) and whether the trough state is normal. If the result is no, the index of the current index set minus 1 is passed as the index of the peak to be processed, and then jump to step S200.

[0181] If the result is yes, find the unused indexes on both sides of the peak to be processed (all indexes are unused by default).

[0182] Step S411, determine whether the net power generation of the current processing peak is greater than 0; if the result is yes, mark the peak as used, and compare which of the unused peaks on both sides has a higher electricity price, and obtain its corresponding index to set as the new peak index to be processed, and jump to step S401 after updating the peak. If the result is no, it is processed by default and enters the valley loop processing logic, that is, jump to step S420.

[0183] S420, determine whether to cycle the trough. If not, pass the index of the current index set minus 1 as the index of the peak to be processed, and then jump to step S200. If yes, jump to step S421.

[0184] S421, determine whether the net power of the current processing valley is greater than 0; if the result is yes, execute the following step S422; if the result is no, jump to the following step S424;

[0185] S422, determining whether the net power generation of the to-be-processed trough is greater than the net power generation required by the to-be-processed peak;

[0186] S423. If the result is yes, update the data as follows:

[0187] Net power generation at trough = Net power generation at trough + Net power generation at peak

[0188] This is the peak power processing = This is the peak power processing + peak net power generation

[0189] Peak net power generation = 0

[0190] If the result is no, update the data as follows:

[0191] Peak net power generation = peak net power generation + trough net power generation

[0192] The current peak power generation = the current peak power generation - the current valley net power generation

[0193] Net power generation at trough = 0.

[0194] S424: If the peak to be processed still needs electricity, determine whether there is available capacity for purchase in the trough to be processed; if the result is yes, execute the following step S425;

[0195] S425: Determine whether the purchasable capacity of the to-be-processed trough can meet the electricity demand of the to-be-processed peak.

[0196] If the result is yes, then step S426 is executed, that is, the capacity available for purchase during the trough period can meet the demand during the peak period, and the updated data is as follows:

[0197] Valley electricity purchase amount = Valley electricity purchase amount + Peak net power generation

[0198] Net power generation at trough = Net power generation at trough + Net power generation at peak

[0199] Peak purchase amount = Peak purchase amount + | Peak net power generation |

[0200] This is the peak power processing = This is the peak power processing + peak net power generation

[0201] Peak net power generation = 0.

[0202] If the result is no, then step S427 is executed, i.e. the capacity available for purchase during the trough period cannot meet the demand during the peak period, and the data is updated as follows:

[0203] Peak net power generation = peak net power generation + trough power available for purchase

[0204] Net power generation at trough = Net power generation at trough - Power available for purchase at trough

[0205] Power purchased at trough = Power purchased at trough + |Power available for purchase at trough|

[0206] The current peak power processing = the current peak power processing - the trough power available for purchase

[0207] The amount of electricity available for purchase at the trough = 0.00.

[0208] After step S421, step S426 or step S427, perform the following steps:

[0209] S430, judging whether the condition is met: the net power generation in the peak index set processed this time is negative, there is purchased power and it is a peak; if the result is yes, execute the following step S431; if the result is no, execute the following step 435;

[0210] S431, determining whether the peak net power generation has changed, if it has changed, executing the following step S432, if it has not changed, executing the following step S434;

[0211] S432, determine whether the peak change value is greater than the peak purchase amount; if the result is yes, execute the following step S433; if the result is no, update the sum of the peak purchase amount (negative number) and the peak change value to the new peak purchase amount, and execute the following step S434;

[0212] S433, clearing the peak purchase amount;

[0213] S434, updating the purchase quantity of the information corresponding to the peak index in the peak and trough object set;

[0214] S435, updating the peak and trough object sets, trough net power generation, trough purchasable power, trough purchasable power and peak net power generation;

[0215] S436. If the trough purchase amount is not 0, it is recorded as the net power generation value to be processed for peak processing.

[0216] If the amount of electricity purchased for the peak is not 0, the net electricity generated by the peak is recorded in the valley information. This is only for the purpose of reserving storage information and is not used yet.

[0217] After that, update the initial power, expected battery power and other information as follows:

[0218] S440: Update initial battery power and expected battery power information.

[0219] S441, passing in the current peak and trough object collection and the current trough index; only the trough index and all subsequent initial power information need to be updated.

[0220] S442, if the trough index to be processed is greater than or equal to 0, cyclically update the data starting from the trough index;

[0221] S443, determine whether the purchased power is greater than 0; if greater than 0, execute the following step S444; if equal to 0, update the expected battery power = expected battery power = current battery power + purchased power + expected power generation - expected purchased power, and then execute the following step S450;

[0222] S444. If the difference between the expected power generation and the expected purchase power is greater than 0, then update the expected battery power = current battery power + purchase power + expected power generation - expected purchase power, and then execute step S450; if it is not greater than 0, then update the expected battery power = current battery power + purchase power, and then execute step S450;

[0223] S450, determining whether the expected battery power is greater than the battery capacity;

[0224] If the result is no, update the expected battery capacity to 0;

[0225] If the result is yes, and the processed data is the data at the trough or next to the trough currently being processed, then mark the current round of beam processing as completed, and set the expected battery capacity equal to the battery capacity;

[0226] S451, determine whether the current battery power + the available power is greater than the battery capacity; if the result is yes, execute the following step S452; if the result is no, jump to step S457;

[0227] S452, set excess capacity = current battery power + available power - battery capacity;

[0228] S453, determine whether the excess capacity is greater than the available power; if the result is yes, clear the available power and jump to step S457; if the result is no, execute the following step S454;

[0229] S454, subtracting the excess capacity from the available electricity to update the available electricity;

[0230] S455, determine whether the original purchased power is greater than the battery capacity minus the current battery power; if the result is positive, execute the following step S456; if the result is negative, jump to step S457;

[0231] S456, update the purchased power=battery capacity-current battery power;

[0232] S457, update the expected power in the information corresponding to the peak and valley index currently being processed, that is, the expected power obtained by the logical processing of S443, S444, etc., and the initial power of the object of the current index + 1 are the expected power of the current data.

[0233] In step S600, if the judgment result is no, jump to step S401 first. When the peak price is not greater than the trough or the trough state is abnormal, the index of the current encapsulated index set is passed in minus 1, and then enter step S200.

[0234] The specific operation process of the above energy storage photovoltaic energy system is as follows:

[0235] Step 1: User Preferences Configuration Receiver

[0236] User interface: The user chooses whether to enable the “Three Monks” AI large model intelligent control algorithm through the user interface, such as the settings interface of the APP. This operation generates a command data packet containing the user ID, device ID, and algorithm enable / disable status.

[0237] Processing process: The user interface collects the user's selections, encrypts the instruction data packet, and sends it to the user preference processing module on the server through a secure channel.

[0238] Step 2: Server-side processing and prediction task startup

[0239] The encrypted instruction data packet received by the server's user preference processing module & prediction algorithm module

[0240] Processing process: The user preference processing module decrypts the data packet, verifies the user's identity and extracts the command content. After confirming the start of the intelligent control algorithm, the device identification and related parameters are forwarded to the prediction algorithm module. The prediction algorithm module starts the power generation and power consumption forecast for 1 to 7 days based on multiple data such as the device's geographical location, historical power generation and power consumption data, and regional electricity price policies.

[0241] Step 3: Data analysis and strategy generation

[0242] The server's prediction algorithm module & intelligent optimization module, the predicted data after analysis and processing (future power generation and power consumption forecast values)

[0243] Processing: The future power supply and demand forecast data generated by the prediction algorithm module is sent to the intelligent optimization module. This module optimizes the energy storage charging and discharging strategy based on the prediction results, combined with the current energy storage status, equipment characteristics and user preferences. The strategy includes but is not limited to the optimal charging and discharging time, charging and discharging power, etc.

[0244] The controller of the photovoltaic energy storage device receives the above energy storage charging and discharging strategy, and controls the charging and discharging time and charging and discharging power of the battery according to the predicted data.

[0245] Through the above steps, from user setting trigger, through data encryption transmission, cloud-based intelligent analysis and processing, to device-side policy implementation and user feedback, a closed-loop high-efficiency management process is formed, ensuring the security of user data and the intelligence of device operation.

[0246] Simulation Example

[0247] According to the above control method of the embodiment, the energy storage charging and discharging strategy of the photovoltaic energy storage device is specified, and the data of 24 hours of operation is shown in Table 1 below.

[0248] Table 1

[0249]

[0250] Among them, the traditional control strategy is a simple control based on the maximum and minimum values ​​of electricity prices, directly charging in the minimum value range and discharging in the maximum value range. Fig. 9As shown, the curve of the initial mode represents the electricity cost and electricity consumption statistics of the traditional control strategy, and the curve of the smart mode represents the electricity cost and electricity consumption statistics of the control strategy of the embodiment.

[0251] The control method of this embodiment can optimize electricity charges and significantly save costs; it can make accurate predictions and achieve efficient energy storage management; it can customize control strategies for photovoltaic energy storage equipment and has strong adaptability.

[0252] The current systems on the market fail to fully utilize the potential of big data and AI algorithms, especially ignoring the electricity price policies in specific regions, the specific impact of geographical location on solar power generation, and the historical electricity consumption behavior patterns in the surrounding environment. This means that the key to improvement lies in more in-depth integration and analysis of these complex factors. The control method of this embodiment improves the prediction accuracy: By introducing more advanced machine learning models, the new system aims to significantly improve the accuracy of future electricity demand forecasts, covering short-term forecasts of 1-7 days. This improvement is directly related to the efficient planning and allocation of energy storage resources. The control method of this embodiment can realize personalized strategy formulation: the system will provide more personalized electricity and energy storage strategy recommendations based on the unique needs of each user and the specific conditions of the environment in which they are located, so as to achieve the optimal use of energy storage power at critical moments, which is an important breakthrough in the traditional "one-size-fits-all" strategy. The control method of this embodiment realizes refined resource allocation: the improved system is committed to realizing the refined management of energy storage resources, ensuring the maximum cost savings during the time period when users need it most, thereby improving the overall energy utilization efficiency and economic benefits. The control method of this embodiment improves user experience and satisfaction: Through the above improvement measures, the ultimate goal is to improve the user experience and satisfaction in energy saving and fee reduction, so that users can intuitively feel the cost savings while enhancing their trust and reliance on the intelligent energy management system. In summary, the control method of this embodiment focuses on deepening data application, improving prediction accuracy, implementing personalized strategies, achieving refined management, and ultimately improving user satisfaction.

[0253] As shown in this specification and claims, the terms "comprises" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements. The term "and / or" as used herein includes any combination of one or more of the relevant listed items.

[0254] It should be noted that, unless otherwise specified, when a feature is referred to as being "fixed" or "connected" to another feature, it may be directly fixed or connected to the other feature, or it may be indirectly fixed or connected to the other feature. In addition, the descriptions of up, down, left, right, etc. used in the present invention are only relative to the relative positional relationship of the components of the present invention in the drawings.

[0255] The above embodiment is only for illustrating the technical concept and features of the present invention, and is a preferred embodiment. Its purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and it cannot be used to limit the protection scope of the present invention. Any equivalent transformation or modification made according to the principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A control method for an energy storage photovoltaic energy system, characterized in that: The steps include: S100, predicting the power consumption and power generation in a future period of time, establishing a curve chart of the electricity price over time in the period of time, finding all the peak and trough information in the curve chart, and establishing a peak and trough object set; S200, setting the peak to be processed, and setting the last peak as the peak to be processed during the first processing; S300, taking the trough that is smaller in time than the to-be-processed peak and closest to it as the to-be-processed trough; S400, if the electricity price of the peak to be processed is greater than the electricity price of the trough to be processed, perform trough processing logic, and update the purchase amount and peak net power generation of the peak and trough objects corresponding to the index of the peak to be processed in the peak and trough object set; S500, determining whether the net power generation of the peak to be processed is 0; If the result is yes, the peak to be processed is marked as used, and the peaks with higher electricity prices on the left and right sides of the peak are taken as new peaks to be processed, and the following step S600 is executed; If the result is no, the left and right sides of the trough to be processed with lower electricity prices are taken as new troughs to be processed; if the index of the new trough to be processed in the peak and trough object set is not equal to the index of the peak to be processed in the peak and trough object set, repeat step S400; if the index of the new trough to be processed is equal to the index of the peak to be processed, execute the following step S600; S600, determining whether the new peak to be processed is the first peak; If the result is yes, then end the program; If the answer is no, then step S200 is repeated.

2. The control method according to claim 1, characterized in that: In step S100, the current power of the energy storage photovoltaic energy system is put into the earliest time data in the peak and trough object set as the initial power.

3. The control method according to claim 2, characterized in that: In step S100, information is initialized according to the predicted data and sorted according to time. The initialized information includes the expected power generation, power consumption, net power generation, maximum hourly charge capacity of the battery, purchasable power, excess net power generation, dischargeable capacity, initial power, initial purchased power and battery capacity for each time interval; wherein, if the current power is less than or equal to 0, the initial power is 0.

4. The control method according to claim 3, characterized in that: In step S100, the peak and trough object sets use different memory addresses, and the peak and trough object sets include multiple ones of power station ID, inverter serial number, capacity, time, electricity price, expected power generation, expected power consumption, net power generation, out-of-range power generation, purchasable power, dischargeable power, initial battery power SOC, purchasable power, expected battery power, maximum charge per hour, charge and discharge plan, cost, and initial cost; find the indexes of all peaks and troughs according to the peak and trough object sets, and put them into the peak index set and the trough index set.

5. The control method according to claim 1, characterized in that: Step S400 includes the following steps: S401, determining whether the electricity price of the peak to be processed is greater than the electricity price of the trough to be processed, and whether the trough state is normal; If the result is no, then find the peaks and troughs to be processed again; If the result is yes, then execute the following steps S410 to S411; S410, finding the closest unused peaks on both sides of the peak to be processed; S411, determining whether the net power generation of the peak to be processed is greater than or equal to 0; If the result is yes, mark the peak as used, and compare the peaks on both sides found in step S410 to see which has the highest electricity price. Set the peak with the highest electricity price as the new current processing peak, and repeat step S401; if the result is no, perform the trough processing logic.

6. The control method according to claim 1 or 5, characterized in that: Step S400 includes the following steps: S420: whether to perform trough processing; if trough processing is performed, execute the following step S421; S421, determining whether the net power generation of the trough currently processed is greater than 0; If the result is yes, execute the following step S422; if the result is no, jump to the following step S424; S422, determining whether the net power generation of the to-be-processed trough is greater than the net power generation required by the to-be-processed peak; If the result is yes, then execute steps S423 to S424; S423, updating the data of the valley net power generation and the peak power generation to be processed, and the peak net power generation is 0; S424, when the peak to be processed still needs electricity, determine whether there is purchasable capacity in the trough to be processed; If the result is yes, then execute the following step S425; S425, determining whether the purchasable capacity of the to-be-processed trough can meet the power demand of the to-be-processed peak; If the result is yes, then execute step S426: set the net power generation of the peak to be processed to 0, and update the data of the trough power available for purchase, the trough net power generation, the peak power purchase and the peak power to be processed; If the result is no, execute step S427: set the purchasable electricity amount of the to-be-processed trough to 0, and update the data of the peak net power generation, trough net power generation, trough purchased electricity and the to-be-processed peak processed electricity.

7. The control method according to claim 6, characterized in that: Step S400 includes the following steps after step S426 or step S427: S430, determine whether the conditions are met: the power generation is negative, there is power purchase and it is a peak; if the result is yes, execute the following step S431; if the result is no, execute the following step 435; S431, determining whether the peak net power generation has changed, if it has changed, executing the following step S432, if it has not changed, executing the following step S434; S432, determining whether the peak change value is greater than the peak purchase power; If the result is yes, the following step S433 is executed; if the result is no, the sum of the peak purchase amount and the peak change value is updated as the new peak purchase amount, and the following step S434 is executed; S433, clearing the peak purchase amount; S434, updating the purchase quantity of the entity corresponding to the peak index in the peak and trough object collection; S435, updating the peak and trough object sets, trough net power generation, trough purchasable power, trough purchasable power and peak net power generation; S436. If the trough purchase amount is not 0, it is recorded as the net power generation value to be processed for peak processing.

8. The control method according to claim 7, characterized in that: Step S400 includes the following steps after step S436: S440, updating initial battery power and expected battery power information; S441, passing in the current peak and trough object set and the current trough index; S442, if the index of the trough to be processed is greater than or equal to 0, cyclically update the data starting from the trough index; S443, determine whether the purchased power is greater than 0; if greater than 0, execute the following step S444; if equal to 0, update the expected battery power = expected battery power = current battery power + purchased power + expected power generation - expected purchased power, and then execute the following step S450; S444. If the difference between the expected power generation and the expected purchase power is greater than 0, then update the expected battery power = current battery power + purchase power + expected power generation - expected purchase power, and then execute step S450; if it is not greater than 0, then update the expected battery power = current battery power + purchase power, and then execute step S450; S450, determining whether the expected battery power is greater than the battery capacity; If the result is no, update the expected battery capacity to 0; If the result is yes, and the processed data is the data at the trough or next to the trough currently being processed, then mark this round of beam processing as completed, and set the expected battery capacity equal to the battery capacity; S451, determining whether the current battery power + the available power is greater than the battery capacity; If the result is yes, then execute the following step S452; If the result is no, jump to step S457; S452, set excess capacity = current battery power + available power - battery capacity; S453, determining whether the excess capacity is greater than the available power; If the result is yes, clear the available power and jump to step S457; If the result is no, execute the following step S454; S454, subtracting the excess capacity from the available electricity to update the available electricity; S455, determining whether the original purchased power is greater than the battery capacity minus the current battery power; If the result is position, execute the following step S456; If the result is no, jump to step S457; S456, update the purchased power=battery capacity-current battery power; S457: Update the expected power, and the initial power of the object with the current index + 1 is the expected power of the current data.

9. The control method according to claim 6, characterized in that: In step S422, if the result is no, the data is updated, the net trough power generation after the update is 0, the net peak power generation after the update = the net peak power generation before the update + the net trough power generation, the peak power processing amount before the update = the peak power processing amount after the update - the net trough power generation, and then jump to step S424.

10. The control method according to claim 6 or 9, characterized in that: In step S423, the net trough power generation after updating = the net trough power generation before updating + the net peak power generation; the peak power processing amount after updating = the peak power processing amount before updating + the net peak power generation.

11. The control method according to claim 1, characterized in that: In step S400, the expected power is updated, and the expected power is used as the initial power of the peak and valley object with the current index +1.

12. The control method according to claim 1, characterized in that: Before step S600, the trough processing logic is jumped out and the next peak processing logic is entered.

13. The control method according to claim 1, characterized in that: The future period of time is 1 to 7 days in the future, and the future period of time is divided into multiple time segments, and the duration of each time segment is 10 to 60 minutes.

14. A photovoltaic energy storage device, characterized in that: include: A controller, which is used to receive prediction data generated by a server according to the control method described in any one of claims 1 to 13, and control the charge and discharge time and charge and discharge power of a battery according to the prediction data.

15. A photovoltaic energy storage system, characterized in that: include: A server, which is used to receive a user's instruction data packet, wherein the instruction data packet includes a user ID, a device ID, and an enable or disable status of the smart control; The server also obtains instructions from the instruction data packet, and after confirming that the intelligent control is turned on, generates prediction data according to the device ID based on the device's geographical location, historical power generation and power consumption data, and regional electricity price policy, and sends it to the photovoltaic energy storage device by the control method described in any one of claims 1 to 13. The prediction data includes power generation and power consumption, charging and discharging time, and charging and discharging power in a future period of time.

Images