Distributed photovoltaic energy storage distribution method, device and equipment based on life prediction

Abstract

The invention provides a distributed photovoltaic energy storage distribution method, device and equipment based on life prediction, and relates to the technical field of power systems. The method comprises the following steps: inputting equipment information and historical use data of each energy storage equipment in an energy storage system of a target area into a pre-trained deep learning model to obtain a life prediction result of each energy storage equipment; performing prediction according to the historical power grid load data in the target area and the historical power generation data of the distributed photovoltaic system to obtain predicted load data and predicted power generation data; determining a first energy storage capacity distribution scheme of the target area; and adjusting the energy storage capacity distribution information of each energy storage device in the first energy storage capacity distribution scheme according to the life prediction results of all the energy storage devices to obtain a second energy storage capacity distribution scheme of the target area. According to the invention, the occurrence probability of failure of the energy storage system can be effectively reduced, and the feasibility of the second energy storage capacity allocation scheme is ensured.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a distributed photovoltaic energy storage allocation method, apparatus and equipment based on lifetime prediction. Background Technology

[0002] Distributed photovoltaic (PV) power output is highly volatile and intermittent due to variations in sunlight and weather, easily leading to issues such as curtailment and voltage fluctuations in the distribution network. Energy storage capacity allocation, as a core component of the system, directly determines operational performance. Rational allocation of energy storage capacity can smooth out PV output fluctuations, reduce curtailment rates, and improve the absorption capacity of clean energy. Furthermore, it can supplement PV power output when it is insufficient, ensuring the reliability of power supply to users and responding to sudden power outages. It can also leverage peak-valley pricing to achieve off-peak energy storage and peak-peak discharge, reducing electricity costs for users. Simultaneously, it provides peak-shaving and frequency regulation services to the grid, contributing to distribution network stability and serving as a key support for the large-scale development of distributed PV.

[0003] Existing technologies primarily employ load matching methods when allocating energy storage capacity. First, they collect photovoltaic (PV) output data and user electricity load data over a period of time, plotting their time-series changes. Then, by comparing the two curves, they accurately identify periods where PV output is lower than user load, calculating the power deficit for each period, and simultaneously calculating the power surplus during periods when PV output is higher than user load. Finally, based on the power deficit, they determine the minimum discharge capacity required for energy storage, and combine this with the surplus to determine the maximum charging capacity. Ultimately, they select a suitable capacity that covers the deficit and matches the surplus as the energy storage capacity, thus avoiding power outages due to insufficient capacity or equipment idleness and waste due to excess capacity.

[0004] However, the inventors discovered that the lifespan of an energy storage system is affected by factors such as the depth of charge and discharge and the frequency of charge and discharge. However, the existing technology does not take into account the impact of the energy storage system's lifespan decay on capacity requirements, which can easily lead to energy storage system failure. Energy storage system failure can easily lead to insufficient power supply in the later stages. Summary of the Invention

[0005] This invention provides a distributed photovoltaic energy storage allocation method, device, and equipment based on lifetime prediction, to solve the problem that the prior art does not take into account the impact of energy storage system lifetime decay on capacity demand, which easily leads to energy storage system failure, and the failure of energy storage system can easily lead to insufficient power supply in the later stage.

[0006] In a first aspect, embodiments of the present invention provide a distributed photovoltaic energy storage allocation method based on lifetime prediction, comprising: The equipment information and historical usage data of each energy storage device in the energy storage system in the target area are input into a pre-trained deep learning model to obtain the lifespan prediction results of each energy storage device. Based on the historical power grid load data and the historical power generation data of distributed photovoltaic power in the target area, the predicted load data and predicted power generation data of the target area in the target time period are obtained. Based on multiple preset constraints, predicted load data, and predicted power generation data, the first energy storage capacity allocation scheme for the target area within the target time period is determined. Based on the life prediction results of all energy storage devices, the energy storage capacity allocation information of each energy storage device in the first energy storage capacity allocation scheme is adjusted, and based on all the adjustment results, the second energy storage capacity allocation scheme for the target area is obtained.

[0007] Secondly, embodiments of the present invention provide a distributed photovoltaic energy storage distribution device based on lifetime prediction, comprising: The prediction module is used to input the equipment information and historical usage data of each energy storage device in the energy storage system in the target area into a pre-trained deep learning model to obtain the life prediction result of each energy storage device. The prediction module is used to make predictions based on historical grid load data and historical power generation data of distributed photovoltaic power in the target area, and to obtain the predicted load data and predicted power generation data of the target area within the target time period. The determination module is used to determine the first energy storage capacity allocation scheme for the target area within the target time period based on multiple preset constraints, predicted load data, and predicted power generation data. The adjustment module is used to adjust the energy storage capacity allocation information of each energy storage device in the first energy storage capacity allocation scheme based on the life prediction results of all energy storage devices, and obtain the second energy storage capacity allocation scheme for the target area based on all the adjustment results.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0009] In this embodiment of the invention, the lifespan prediction result of each energy storage device is obtained through the device information and historical usage data of the energy storage devices. Predicted load data and predicted power generation data are obtained through historical grid load data and historical power generation data of distributed photovoltaic systems. A first energy storage capacity allocation scheme for the target area is determined, ensuring the matching degree between the first energy storage capacity allocation scheme and the load and power generation data. The energy storage capacity allocation information of each energy storage device within the first energy storage capacity allocation scheme is adjusted based on the lifespan prediction results to obtain a second energy storage capacity allocation scheme. This effectively matches the supply and demand balance between grid load and distributed photovoltaic power generation, taking into account the lifespan factor of each energy storage device. This ensures that the second energy storage capacity allocation scheme will not be affected by the lifespan degradation of the energy storage devices during practice, effectively reducing the probability of energy storage system failure and ensuring the feasibility of the second energy storage capacity allocation scheme. This achieves the adaptability of energy storage capacity allocation to each energy storage device in the energy storage system. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the implementation of the distributed photovoltaic energy storage allocation method based on lifetime prediction provided in this embodiment of the invention. Figure 2 This is a flowchart illustrating the implementation of step S140 of the distributed photovoltaic energy storage allocation method based on lifetime prediction provided in this embodiment of the invention. Figure 3 This is a schematic diagram of the structure of the distributed photovoltaic energy storage distribution device based on lifetime prediction provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0012] See Figure 1 The document illustrates a flowchart of the implementation of the distributed photovoltaic energy storage allocation method based on lifetime prediction provided by an embodiment of the present invention, which is described in detail below: Step S110: Input the equipment information and historical usage data of each energy storage device in the energy storage system of the target area into the pre-trained deep learning model to obtain the life prediction result of each energy storage device.

[0013] In some embodiments, the target area refers to a specific geographical area or grid jurisdiction where energy storage capacity allocation is required, and is the spatial boundary of the entire energy storage allocation scheme. An energy storage system refers to the overall system within the target area used to realize energy storage, dispatch, and charge / discharge management. It is the core carrier for balancing regional energy supply and demand, and typically consists of multiple independent energy storage devices, energy management and control units, charge / discharge interfaces, and supporting monitoring equipment. For example, an energy storage system in a community might consist of 10 lithium battery energy storage cabinets, a central energy dispatch system, and related cables and protection devices. Energy storage devices are independent units within the energy storage system that specifically perform the function of energy storage. They are the basic operational objects for lifetime prediction and capacity allocation. Common types include lithium battery energy storage battery packs, lead-acid batteries, and flow batteries, such as a single 2MWh lithium battery energy storage cabinet in an industrial park's energy storage system. Lifetime prediction results refer to the remaining lifetime of the energy storage device and the corresponding capacity range. Equipment information refers to information describing the inherent attributes and basic state of the energy storage device, and is one of the key inputs for deep learning models to predict device lifetime. This mainly includes information such as the device model and installation time. This information determines the initial performance ceiling of the device, directly affecting the accuracy of lifespan prediction.

[0014] It should be noted that this solution is mainly applied to areas with multiple distributed photovoltaic (PV) systems and multiple energy storage devices. When the power generation of distributed PV systems is large, the unconsumable electricity needs to be stored in these energy storage devices. At this time, it is necessary to consider the impact of the lifespan of the energy storage devices on their energy storage capacity, and to avoid situations where the energy storage devices cannot store the energy storage capacity allocated to them. Therefore, the lifespan of the energy storage devices needs to be predicted.

[0015] In a possible implementation, the specific processing method of step S110 is as follows: Collect the device information and historical usage data of each energy storage device in the energy storage system of the target area. The device information includes the device model and the device installation time; the historical usage data includes charge-discharge historical behavior data and historical fault data; for each energy storage device, extract all the charge start times, the SOC at all the charge start times, all the charge end times, and the SOC at all the charge end times from the charge-discharge historical behavior data of the energy storage device, and calculate the cycle count of the energy storage device based on all the charge start times, the SOC at all the charge start times, all the charge end times, and the SOC at all the charge end times of the energy storage device; input the device model, device installation time, historical fault data, and cycle count of each energy storage device into a pre-trained deep learning model to obtain the life prediction result of each energy storage device; the pre-trained deep learning model is trained with the device models, device installation times, fault data, cycle counts, and life data of multiple energy storage devices; the life data includes the life intervals corresponding to all the remaining lives of the energy storage device and the capacity intervals corresponding to the remaining lives of each life interval.

[0016] In some embodiments, through data acquisition devices such as current sensors, voltage monitoring modules, time recorders, and device ledger systems supporting the energy storage system, the device information and historical usage data of each energy storage device in the energy storage system of the target area can be obtained. The charge-discharge historical behavior data refers to the detailed operation record data related to the charging and discharging of electric energy during the past operation of the energy storage device, which is used to analyze the device usage intensity and calculate the cycle count. Its core content includes all the charge start times, the SOC at the start of charging, the charge end times, and the SOC at the end of charging. To improve the solution, only by covering both the charging and discharging links can it be accurately judged whether a complete energy cycle is completed. The cycle count is the core index to measure the life of the energy storage device, which refers to a complete "100% power" usage cycle, and it is calculated by accumulating multiple charge-discharges. For example, if you use half and charge half each time, two times才算一个循环。随着循环次数的增加,电池内部的化学物质会逐渐老化降解,导致其最大容量不可避免地下滑,使设备续航时间缩短。寿命数据指用于训练深度学习模型的样本数据,是模型学习设备参数和寿命规律的映射关系的训练样本,某储能设备的剩余寿命与容量区间的映射规律如表1所示,表1中剩余寿命的单位为年,容量区间指的是剩余寿命对应的容量占最大储能容量的百分比。<{

[0017] Table 1 Mapping relationship between the remaining life and capacity interval of a certain energy storage device It should be noted that there is an unclear expression "two times才算一个循环" in the original Chinese text. I have translated it as literally as possible while keeping the context. You may need to check and correct this part according to the actual situation.

[0018] Step S120: Based on the historical grid load data and the historical power generation data of distributed photovoltaic power generation in the target area, predictions are made to obtain predicted load data and predicted power generation data.

[0019] In some embodiments, historical grid load data refers to the actual recorded data of electricity load of users in the target area over a past period. For example, the total daily load data of industrial, residential, and commercial electricity consumption in a county-level power grid over the past year. This data reflects the historical variation pattern of regional electricity load and serves as the basis for subsequent prediction of future load data, helping to determine when energy storage discharge is needed to supplement power supply. Distributed photovoltaic (PV) refers to relatively small-scale photovoltaic power generation systems deployed on the user side or near load centers. They are usually directly connected to the distribution network and do not require long-distance power transmission. Common forms include photovoltaic panels installed on residential rooftops, photovoltaic arrays on the roofs of factory buildings in industrial parks, and photovoltaic modules on agricultural greenhouses. Their power generation output is significantly affected by natural factors, exhibiting fluctuations and intermittent characteristics, and is an important source of clean power generation in the target area. Historical power generation data refers to the actual power generation records generated by the distributed photovoltaic system during its past operation, reflecting the power generation capacity and output pattern of the photovoltaic system. Predicted load data refers to the electricity load data of the target area for a future period calculated based on historical grid load data using a specialized load prediction model. Predicted power generation data refers to the power generation data of distributed photovoltaic (PV) systems in a target area over a future period, calculated using a PV power generation prediction model based on historical power generation data and predicted environmental data. PV power generation prediction models can be regression models based on meteorological data, machine learning models, etc.

[0020] In one possible implementation, step S120 is specifically processed as follows: inputting historical grid load data within the target area into the load prediction model of the target area to obtain the predicted load data of the target area within the target time period; inputting historical power generation data and predicted environmental data of distributed photovoltaic power into the photovoltaic power generation prediction model to obtain the predicted power generation data of the target area within the target time period.

[0021] In some embodiments, historical grid load data refers to historical data generated by the overall electricity load of the target area's power grid over a past period. The target area load prediction model refers to an algorithm model specifically designed or trained for the electricity consumption characteristics of the target area, used to output future electricity load data for that area. The model type needs to be selected based on the regional load characteristics. For example, if the target area is an industrial park, where the load is greatly affected by the production cycle and the fluctuation pattern is relatively fixed, a time series model can be used; if the area is a mixed community containing residents, businesses, and small industries, with complex load fluctuations, a deep learning model, such as an LSTM model, can be used. Predictive environmental data refers to environmental state parameters of the target area within the target time period, including data such as light intensity, weather conditions, ambient temperature, and wind speed. The photovoltaic power generation prediction model refers to an algorithm model used to predict the future power generation of distributed photovoltaic systems. By inputting the predicted environmental data within the target time period, it outputs the predicted power generation data of distributed photovoltaic systems in the target area. The model type needs to consider both environmental factors and equipment characteristics, and can use a meteorological factor regression model or a CNN-LSTM hybrid model.

[0022] Step S130: Based on multiple preset constraints, predicted load data, and predicted power generation data, determine the first energy storage capacity allocation scheme for the target area within the target time period.

[0023] In some embodiments, multiple preset constraints refer to the limiting conditions that must be followed when determining the first energy storage capacity allocation scheme to ensure the safe operation of the power grid and the stable operation of the energy storage equipment. These constraints represent the safety red line for the feasibility of the scheme. They mainly include five categories: power flow constraints, branch power flow constraints, node voltage constraints, energy storage capacity constraints, and energy storage charging and discharging power constraints. The energy storage capacity constraint here refers to the constraint on the rated capacity of the energy storage equipment, without considering the impact of equipment lifespan. These constraints ensure that the first scheme, while meeting supply and demand balance, will not pose a safety risk to the power grid and equipment. The first energy storage capacity allocation scheme refers to the preliminary target area energy storage capacity allocation scheme formulated based on predicted load data, predicted generation data, and multiple preset constraints. The target time period refers to the time period during which energy storage capacity allocation needs to be carried out.

[0024] In one possible implementation, step S130 is specifically processed as follows: with the objectives of maximizing the marginal benefit of energy storage operation, minimizing line loss, and minimizing voltage deviation, an objective function is constructed, and with power flow constraints, branch power flow constraints, node voltage constraints, energy storage capacity constraints, and energy storage charging and discharging power constraints as preset constraints, an energy storage capacity allocation model is constructed; the predicted load data and predicted power generation data are input into the energy storage capacity allocation model to obtain the first energy storage capacity allocation scheme for the target area.

[0025] In some embodiments, maximizing the marginal benefit of energy storage operation refers to maximizing the additional revenue generated by each additional unit of operational investment in the energy storage system when constructing an energy storage capacity allocation scheme. This is the core objective for measuring the economic rationality of energy storage allocation. Minimizing line loss refers to minimizing the energy loss caused by factors such as line resistance and equipment losses during power grid transmission and distribution when constructing an energy storage capacity allocation scheme. This is a key technical objective for ensuring grid operating efficiency. Minimizing voltage deviation refers to minimizing the difference between the actual operating voltage and the rated voltage of each node in the grid when constructing energy storage capacity allocation objectives. This is a core technical objective for ensuring power grid supply quality and equipment safety. The objective function refers to the core function used to guide the adjustment of energy storage capacity allocation after quantifying the three objectives of maximizing the marginal benefit of energy storage operation, minimizing line loss, and minimizing voltage deviation into mathematical expressions. Since the three objectives may conflict to some extent, the objective function is usually integrated by weighted summation, ultimately transforming into a single-objective adjustment problem. For example, if the target area focuses more on economic efficiency, a higher weight will be assigned to maximizing the marginal benefit of energy storage operation, while a lower weight will be assigned to minimizing line loss and minimizing voltage deviation. Power flow constraints refer to the constraints set when constructing an energy storage capacity allocation model to ensure that the flow of electrical energy within the power grid conforms to the basic laws of circuits. These are fundamental constraints for ensuring the safe and stable operation of the power grid. Branch power flow constraints, based on power flow constraints, are upper limits of power or current set for specific distribution circuits, transformer connection lines, and other branches within the power grid. These are key constraints to prevent overload damage to branch equipment. Each branch has its rated carrying capacity; if the power flow in a branch exceeds the rated value, it can lead to overheating, insulation aging, and even short-circuit faults. Node voltage constraints refer to the actual allowable voltage fluctuation range set for each node in the power grid, based on the goal of minimizing voltage deviation. These are hard constraints to ensure power quality. Different types of nodes have different allowable voltage deviation ranges. For example, the allowable voltage deviation for residential power nodes is ±10% of the rated voltage, while the allowable deviation for industrial precision equipment nodes is ±5%. Node voltage constraints clearly define the upper and lower limits of the voltage for each node, ensuring that the voltage of all nodes does not exceed this range after energy storage capacity is allocated. Energy storage capacity constraints refer to the constraints set on the physical capacity limitations of energy storage devices. Specifically, the actual stored electrical energy of an energy storage device must not exceed its rated total capacity, nor be lower than its minimum allowable discharge capacity. These are fundamental constraints for protecting the safety of energy storage devices and extending their service life. Energy storage charge / discharge power constraints refer to the constraints set on the charge / discharge rate of energy storage devices. Specifically, the actual charge / discharge power of an energy storage device must not exceed its rated charge / discharge power. These are key constraints to prevent overheating and current surges caused by excessive power. The energy storage capacity allocation model is a mathematical model constructed by integrating the objective function and preset constraints. It is used to calculate the first energy storage capacity allocation scheme for the target area.The first energy storage capacity allocation scheme refers to the initial energy storage capacity allocation scheme output by the model after inputting predicted load data and predicted power generation data. It serves as the basis for subsequent adjustments based on the lifespan of the energy storage devices. This scheme satisfies all preset constraints and objective functions, but it does not yet consider the impact of energy storage device lifespan degradation on capacity demand.

[0026] Step S140: Based on the life prediction results of all energy storage devices, adjust the energy storage capacity allocation information of each energy storage device in the first energy storage capacity allocation scheme, and based on all the adjustment results, obtain the second energy storage capacity allocation scheme for the target area.

[0027] In some embodiments, energy storage capacity allocation information is a key parameter describing the details of the capacity allocation of target energy storage devices, directly determining the operating status of the devices and their ability to support the power grid. This mainly includes the initial energy storage capacity, the final energy storage capacity, and the energy storage rate. The second energy storage capacity allocation scheme refers to the final scheme obtained by adjusting the energy storage capacity allocation information of each target energy storage device in the first scheme based on the lifetime prediction results of all target energy storage devices. During the adjustment process, the allocated capacity of each target energy storage device is adjusted according to its lifetime prediction results. The final scheme not only satisfies the advantages of supply and demand balance and compliance with grid constraints, but also takes lifetime factors into account, ensuring that each energy storage device operates under reasonable load, effectively reducing the probability of device failure, and guaranteeing the long-term feasibility of the scheme.

[0028] Step S1401: Extract the energy storage capacity allocation information of each energy storage device from the first energy storage capacity allocation scheme; wherein, the energy storage capacity allocation information of the energy storage device includes the initial energy storage capacity, the final energy storage capacity, and the energy storage rate.

[0029] In some embodiments, the initial energy storage capacity is the amount of electrical energy stored by the energy storage device at the beginning of an operating cycle, such as a device with an initial energy capacity of 1.2 MWh before energy storage. The final energy storage capacity is the amount of electrical energy stored by the energy storage device after storing the energy required in the energy storage capacity allocation scheme, such as a device with a final energy capacity of 2.6 MWh after completing energy storage. The energy storage rate is the rate of change of electrical energy by the energy storage device per unit time, reflecting how fast the charging and discharging is. For example, a device with a charging rate of 400 kWh / hour means that it can add 400 kWh of energy per hour.

[0030] Step S1402: Calculate the first capacity difference between the final energy storage capacity and the initial energy storage capacity of each energy storage device.

[0031] In some embodiments, the first capacity difference refers to a quantitative indicator obtained by calculating the difference between the final energy storage capacity and the initial energy storage capacity of a single target energy storage device, used to reflect the charging demand scale that the device needs to achieve in the first energy storage capacity allocation scheme. Since this scheme is mainly applied to the working scenario of distributed photovoltaic power generation needing to be stored in energy storage devices, the first capacity difference must be greater than 0.

[0032] Step S1403: Based on the equipment information and life prediction results of each energy storage device, determine the corresponding capacity for the remaining life of each energy storage device.

[0033] In some embodiments, the capacity corresponding to the remaining lifetime refers to the maximum effective storage capacity that the target energy storage device can safely carry during its current lifespan, determined based on the device information and lifetime prediction results. It is a core lifetime constraint indicator limiting energy storage capacity allocation. The device model can be extracted from the device information, and the corresponding lifetime prediction result can be determined based on the device model. The lifetime prediction result includes the capacity range corresponding to the remaining lifetime interval of the energy storage device.

[0034] In one possible implementation, step S1403 is specifically processed as follows: extract the device usage time of each energy storage device from the device information of each energy storage device; extract the capacity percentage range to which the remaining lifespan of each energy storage device belongs from the lifespan prediction results based on the device usage time of each energy storage device; and determine the lower bound of the capacity range to which the remaining lifespan of each energy storage device belongs as the capacity corresponding to the remaining lifespan of each energy storage device.

[0035] In some embodiments, device usage time refers to the actual continuous operating time of the target energy storage device from the time it is installed and put into normal operation until the start of the target time period. This is the core time dimension data extracted from the device information and a key index for matching lifetime prediction results. Device usage time directly reflects the degree of device aging; the longer the operating time, the more charge-discharge cycles and environmental impacts the device experiences, and the more significant the lifespan degradation usually becomes. The remaining lifetime capacity percentage range refers to the percentage range of energy storage capacity corresponding to the current remaining lifetime of the device relative to the initial total capacity, matched from the lifetime prediction results based on the device usage time. This is a quantitative description of the current lifespan status of the device. The lower bound of the remaining lifetime capacity range refers to the smaller endpoint of the remaining lifetime capacity percentage range. It is a conservative value selected when determining the capacity corresponding to the remaining lifetime, and its core function is to reserve a safety margin for device capacity allocation, avoiding the risk of device overload caused by lifetime prediction errors. Since the lifetime prediction results output a range, selecting the lower bound of the range rather than the upper bound or the middle value ensures that even if the actual lifespan degradation rate of the device is faster than predicted, the allocated capacity will not exceed the device's tolerance range, avoiding accelerated device aging or failure. For example, if the remaining lifespan of a target energy storage device falls within the capacity percentage range of 65%-70%, then the lower bound of this range is 65%. If the initial total capacity of the device is a fixed value, then the capacity corresponding to its remaining lifespan is the initial total capacity multiplied by 65%. By selecting a lower bound, the capacity allocation scheme becomes more reliable, especially suitable for energy storage systems with high safety requirements.

[0036] Step S1404: Use the K-means clustering algorithm to cluster the first capacity difference, energy storage rate, energy storage information and remaining lifetime capacity of all energy storage devices to obtain multiple cluster categories.

[0037] In some embodiments, a cluster category refers to a group of devices grouped according to the similarity of all target energy storage devices based on four dimensions: first capacity difference, energy storage rate, energy storage information, and remaining lifetime capacity, using the K-means clustering algorithm. Devices in the same cluster category have similar characteristics in the above four dimensions, so a unified adjustment strategy can be adopted, eliminating the need to adjust each device individually and significantly improving adjustment efficiency.

[0038] Step S1405: For each cluster category, based on the remaining lifetime capacity of all energy storage devices within that cluster category, adjust the first capacity difference of all energy storage devices within that cluster category to obtain the adjustment result.

[0039] In some embodiments, adjusting the first capacity difference for all energy storage devices refers to adjusting the first capacity difference based on the remaining lifespan of each energy storage device, so that the adjustment result meets the constraints imposed by the remaining lifespan capacity. The adjustment result must simultaneously meet two objectives: first, ensuring that the device allocation does not exceed its own lifespan constraints; and second, ensuring a balance between regional power supply and demand.

[0040] In one possible implementation, step S1405 is specifically processed as follows: The energy storage device with the shortest distance to the cluster center of the cluster is determined as the central energy storage device for that cluster; the product of the remaining lifetime capacity of the central energy storage device and its storage margin is determined as the upper limit of the central energy storage device's energy storage capacity; the difference between the upper limit of the central energy storage device's energy storage capacity and the final energy storage capacity is calculated; if the difference is greater than or equal to 0, then it is not necessary to adjust the energy storage capacity allocation information of all energy storage devices within the cluster, and the difference is compared with the energy storage capacity allocation information of all energy storage devices within the cluster. The product of the number of devices is used to determine the remaining energy storage space for the cluster category; if the difference is less than 0, the upper limit of the energy storage capacity of the central energy storage device is determined as the final energy storage capacity of all energy storage devices in the cluster category, and the product of the difference and the number of energy storage devices in the cluster category is determined as the remaining energy storage capacity of the cluster category; the difference between the corresponding final energy storage capacity of the central energy storage device and the initial energy storage capacity is determined as the second capacity difference of all energy storage devices in the cluster category; the second capacity difference of all energy storage devices is determined as the adjustment result of the cluster category.

[0041] In some embodiments, cluster centers are virtual reference points used in the K-means clustering algorithm to characterize the core features of each cluster category. Their parameters are calculated from the mean or median of key indicators of all target energy storage devices within that category, serving as the core benchmark for measuring the similarity between a device and its category. By calculating the distance (e.g., Euclidean distance) between each device and each cluster center, devices can be assigned to the closest category. The closest device then becomes the central energy storage device for that category, ensuring that subsequent adjustments are based on the device that best fits the category characteristics. A central energy storage device refers to the target energy storage device with the shortest distance to the cluster center within a single cluster category; it is the physical representative of the characteristics of that cluster category. Since devices within the same category are highly similar in dimensions such as charging demand, operating rate, and lifespan after K-means clustering, the parameters of the central energy storage device can reflect the common characteristics of the entire category to the greatest extent. Therefore, subsequent capacity allocation adjustments for that cluster category will prioritize calculating adjustment parameters based on the central energy storage device, and then generalize the results to all devices within the category, avoiding individual calculations for each device and significantly simplifying the adjustment process. For example, if a cluster contains 10 lithium-ion battery energy storage devices, and device A is closest to the cluster center, then device A is the central energy storage device for that cluster. Subsequent calculations of the energy storage capacity limit and adjustments to the energy storage rate for this cluster are based on the parameters of device A. Storage margin is a safety factor set to mitigate lifetime prediction errors and ensure safe operation of the equipment. It is typically a fixed percentage less than 1. Based on the remaining lifetime capacity of the central energy storage device, the upper limit of the energy storage capacity is further reduced to prevent capacity allocation from approaching the device's lifespan limit due to deviations in lifetime prediction, thus preventing accelerated device degradation. Furthermore, storage margin can also be determined based on the characteristics of the energy storage device. For instance, if the damage to a certain energy storage device is minimal when its charge level is between 0.2 and 0.8, then 0.8 can be set as the storage margin for that device during charging. The upper limit of energy storage capacity, relative to the central energy storage device, is obtained by multiplying the remaining lifetime capacity by the storage margin. It represents the maximum energy storage capacity that the device can safely handle during its current lifespan and is the highest standard limiting the capacity allocation of all devices within that cluster. Since the central energy storage device represents the common characteristics of the cluster category, its upper limit of energy storage capacity also applies to other devices within the category, ensuring that the capacity allocation of the entire category meets the lifetime constraints. The second capacity difference refers to the difference between the final energy storage capacity of the central energy storage device after adjustment and its initial energy storage capacity. It is a benchmark difference used to unify the capacity allocation of all target energy storage devices within the cluster category. Because the device characteristics of the same cluster category are highly similar, the adjustment result of the central energy storage device can be directly generalized to the entire category. Therefore, using its adjusted capacity difference as the second capacity difference for all devices ensures that the charging and discharging needs of devices within the category are consistent with the lifetime constraints, avoiding supply and demand imbalances within the category caused by individual adjustments.For example, after the central energy storage device is adjusted, the final energy storage capacity is adjusted to T, while the initial energy storage capacity remains U. Then the second capacity difference is (TU). All other devices in this cluster use (TU) as their own capacity difference and adjust their final energy storage capacity synchronously to ensure that the entire cluster responds in a coordinated and unified manner to the regional power supply and demand.

[0042] In some embodiments, when an energy storage device is storing energy, its durability can be improved by limiting its energy storage rate. For example, when the energy storage rate of the central energy storage device exceeds a first preset threshold and the remaining capacity of the central energy storage device is less than a second preset threshold, the preset energy storage rate can be determined as the energy storage rate of the central energy storage device. The first preset threshold is a safe operating upper limit set for the energy storage rate, used to determine whether the current energy storage rate of the central energy storage device poses an overload risk. This threshold is typically determined based on the technical characteristics of the equipment type, industry safety standards, and long-term operating experience. If the energy storage rate of the central energy storage device exceeds this threshold, it means that the equipment may be operating under high load, which can easily lead to problems such as overheating and accelerated cell wear. The risk is further amplified, especially when the equipment's remaining lifespan is insufficient. The second preset threshold is a safe lower limit set for the capacity corresponding to the remaining lifespan, used to determine whether the remaining lifespan of the central energy storage device is sufficient. This threshold is also determined based on equipment type, retirement standards, and operational risk assessment. If the remaining lifespan of the central energy storage equipment is lower than this threshold, it indicates that the equipment has entered the mid-to-late stage of its lifespan, and its ability to withstand high-load operation has significantly decreased. It is necessary to extend its effective operating time by reducing the energy storage rate and controlling capacity allocation. The preset energy storage rate is an alternative energy storage rate set to ensure equipment safety when the central energy storage equipment simultaneously meets the conditions of exceeding the first preset threshold and having a remaining lifespan capacity lower than the second preset threshold. It is a verified, low-load safe rate suitable for equipment in the mid-to-late stage of its lifespan. This rate is typically lower than the first preset threshold and matches the current lifespan state of the equipment, minimizing wear and tear on the equipment while meeting basic charging and discharging requirements.

[0043] Step S1406: Determine the second energy storage capacity allocation scheme for the target area based on the adjustment results of all cluster categories.

[0044] In some embodiments, the capacity adjustment results of each energy storage device can be obtained from the adjustment results of all cluster categories, and the energy to be stored can be redistributed according to the capacity adjustment results to obtain a second energy storage capacity allocation scheme.

[0045] In one possible implementation, step S1406 is specifically processed as follows: Extract the remaining energy storage capacity or remaining energy storage space for each cluster type within the target area from the adjustment results; determine the sum of the remaining energy storage capacities of all cluster types within the target area as the total remaining energy storage capacity of the target area; calculate the ratio of the remaining energy storage space of each cluster type to the sum of the remaining energy storage spaces of all cluster types, and determine this ratio as the remaining energy storage capacity allocation ratio for each cluster type; allocate the total remaining energy storage capacity to all energy storage devices in each cluster type according to all remaining energy storage capacity allocation ratios; for each cluster type with remaining energy storage space, determine the ratio of the remaining energy storage capacity allocated to that cluster type to the number of energy storage devices within that cluster type as the third capacity difference for each energy storage device in that cluster type; for each cluster type with remaining energy storage capacity, define 0 as the third capacity difference for all energy storage devices within that cluster type; the second capacity difference, the third capacity difference, and the energy storage rate of all energy storage devices constitute the second energy storage capacity allocation scheme for the target area.

[0046] In some embodiments, the remaining energy storage capacity refers to the total energy storage capacity gap of the entire cluster category when the difference between the upper limit of the central energy storage device's energy storage capacity and the final energy storage capacity is less than 0 during the initial adjustment of the cluster category. It reflects the scale of energy storage capacity that the cluster category needs to supplement to meet normal operation requirements and is one of the core bases for subsequent regional-level overall allocation of remaining capacity. For example, if a cluster category contains several target energy storage devices, and the difference between the upper limit of the central energy storage device's energy storage capacity and the final energy storage capacity is negative (i.e., insufficient energy storage capacity), multiplying this negative difference by the total number of devices in the category yields the remaining energy storage capacity of that cluster category. This means that all devices in the category need to supplement this scale of energy storage capacity to meet energy storage requirements without exceeding lifespan constraints. Remaining energy storage space refers to the total surplus energy storage capacity that the entire cluster can support when adjusting cluster categories, provided that the difference between the upper limit of the central energy storage device's energy storage capacity and its final energy storage capacity is greater than or equal to 0 (i.e., the device's current final energy storage capacity demand does not exceed the upper limit under the lifespan constraint). This is calculated by multiplying this difference by the number of target energy storage devices within the cluster category. It reflects the additional energy storage capacity that the cluster category can receive without exceeding its lifespan constraint and represents the usable space for subsequent regional-level allocation of total remaining energy storage capacity. Total remaining energy storage capacity refers to the sum of the remaining energy storage capacities of all cluster types with capacity gaps within the target area. This sum represents the total energy storage capacity gap that needs to be supplemented to meet the normal operation needs of all cluster categories within the entire target area. The remaining energy storage capacity allocation ratio refers to the value calculated for each cluster type with remaining energy storage space, by comparing the remaining energy storage space of that cluster to the sum of the remaining energy storage spaces of all clusters with remaining energy storage space. It is used to determine the allocation weight of total remaining energy storage capacity among different clusters with spare space, ensuring the fairness and rationality of capacity allocation. That is, the larger the remaining energy storage space of a cluster, the more total remaining energy storage capacity can be allocated, which can fully utilize the spare space of each cluster while avoiding excessive allocation pressure on any one cluster. The third capacity difference refers to the additional capacity adjustment value set for each target energy storage device for different types of clusters after the total remaining energy storage capacity allocation is completed at the regional level. It is a supplementary adjustment to the previous second capacity difference, ensuring that the capacity allocation of individual devices accurately matches the regional allocation results. For cluster types with remaining energy storage space, the third capacity difference is the ratio of the remaining energy storage capacity allocated to that cluster to the number of target energy storage devices within the cluster, representing the additional energy storage capacity that each device can add to fully utilize the spare space. For cluster types with remaining energy storage capacity, the third capacity difference is fixed at 0 to avoid the device capacity exceeding its lifespan constraint due to additional adjustments.For example, if a cluster with remaining energy storage space is allocated a remaining energy storage capacity of F, and there are G devices in the cluster, then the third capacity difference for each device is F / G, which means that the final energy storage capacity of each device can be increased by F / G units based on the second capacity difference. However, for a cluster with remaining energy storage capacity, the third capacity difference for all devices is 0, and the allocation corresponding to the second capacity difference in the previous period is maintained.

[0047] By analyzing the equipment information and historical usage data of energy storage devices, the lifespan prediction results for each energy storage device are obtained. Furthermore, predictions are made using historical grid load data and historical power generation data from distributed photovoltaic systems to obtain predicted load and power generation data. Combining these data, a first energy storage capacity allocation scheme for the target area is determined, ensuring that the first scheme matches the load and power generation data. Based on the lifespan prediction results, the remaining lifespan capacity corresponding to each energy storage device is obtained. This capacity is then used to adjust the energy storage capacity allocation information for each device within the first scheme, resulting in a second scheme. This second scheme takes into account the lifespan of each energy storage device, ensuring that it is not affected by the lifespan degradation of the devices during implementation. It exhibits a higher degree of matching with the specific conditions of the energy storage system, effectively reducing the probability of system failure and guaranteeing the feasibility of the second scheme.

[0048] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0049] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0050] Figure 3 A schematic diagram of a distributed photovoltaic energy storage distribution device based on lifetime prediction provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 3 As shown, the distributed photovoltaic energy storage distribution device 3 based on lifetime prediction includes: Prediction module 31 is used to input the equipment information and historical usage data of each energy storage device in the energy storage system in the target area into a pre-trained deep learning model to obtain the life prediction result of each energy storage device. Prediction module 31 is used to make predictions based on historical grid load data and historical power generation data of distributed photovoltaic power generation in the target area, so as to obtain the predicted load data and predicted power generation data of the target area in the target time period. The determination module 32 is used to determine the first energy storage capacity allocation scheme of the target area within the target time period based on multiple preset constraints, predicted load data and predicted power generation data. The adjustment module 33 is used to adjust the energy storage capacity allocation information of each energy storage device in the first energy storage capacity allocation scheme according to the life prediction results of all energy storage devices, and obtain the second energy storage capacity allocation scheme for the target area based on all the adjustment results.

[0051] In one possible implementation, the prediction module 31 is specifically used for: collecting equipment information and historical usage data of each energy storage device in the energy storage system of the target area, wherein the equipment information includes the equipment model and the equipment installation time; the historical usage data includes historical charging and discharging behavior data and historical fault data; for each energy storage device, extracting all charging start times, the SOC of all charging start times, all charging end times, and the SOC of all charging end times from the historical charging and discharging behavior data of the energy storage device, and calculating the number of cycles of the energy storage device based on the SOC of all charging start times, all charging start times, all charging end times, and the SOC of all charging end times; inputting the equipment model, equipment installation time, historical fault data, and number of cycles of each energy storage device into a pre-trained deep learning model to obtain the lifespan prediction result of each energy storage device; wherein the pre-trained deep learning model is trained using the equipment model, equipment installation time, fault data, number of cycles, and lifespan data of multiple energy storage devices; the lifespan data includes the lifespan intervals corresponding to all remaining lifespans of the energy storage devices and the capacity intervals corresponding to the remaining lifespan of each lifespan interval.

[0052] In one possible implementation, the prediction module 31 is further configured to: input historical grid load data within the target area into the load prediction model of the target area to obtain the predicted load data of the target area within the target time period; and input historical power generation data and predicted environmental data of distributed photovoltaic power into the photovoltaic power generation prediction model to obtain the predicted power generation data of the target area within the target time period.

[0053] In one possible implementation, module 32 is specifically used to: construct an objective function with the goals of maximizing the marginal benefit of energy storage operation, minimizing line loss, and minimizing voltage deviation, and construct an energy storage capacity allocation model with branch power flow constraints, node voltage constraints, and energy storage charging and discharging power constraints as preset constraints; input the predicted load data and predicted power generation data into the energy storage capacity allocation model to obtain the first energy storage capacity allocation scheme for the target area.

[0054] In one possible implementation, the adjustment module 33 is specifically used for: extracting the energy storage capacity allocation information of each energy storage device from the first energy storage capacity allocation scheme; wherein, the energy storage capacity allocation information of the energy storage device includes the initial energy storage capacity, the final energy storage capacity, and the energy storage rate; calculating the first capacity difference between the final energy storage capacity and the initial energy storage capacity of each energy storage device; determining the capacity corresponding to the remaining lifespan of each energy storage device based on the device information and lifespan prediction results of each energy storage device; using the K-means clustering algorithm to cluster the first capacity difference, energy storage rate, energy storage information, and capacity corresponding to the remaining lifespan of all energy storage devices to obtain multiple cluster categories; for each cluster category, adjusting the first capacity difference of all energy storage devices within that cluster category based on the capacity corresponding to the remaining lifespan of all energy storage devices within that cluster category to obtain the adjustment result; and determining the second energy storage capacity allocation scheme for the target area based on the adjustment results of all cluster categories.

[0055] In one possible implementation, the adjustment module 33 is further configured to: determine the energy storage device with the shortest distance to the cluster center of the cluster as the central energy storage device of the cluster; determine the upper limit of the energy storage capacity of the central energy storage device as the product of the remaining lifetime capacity and the storage margin of the central energy storage device; calculate the difference between the upper limit of the energy storage capacity of the central energy storage device and the final energy storage capacity; if the difference is greater than or equal to 0, then it is not necessary to adjust the energy storage capacity allocation information of all energy storage devices in the cluster, and the difference is compared with the number of energy storage devices in the cluster. The product of the quantities is used to determine the remaining energy storage space for the cluster category; if the difference is less than 0, the upper limit of the energy storage capacity of the central energy storage device is determined as the final energy storage capacity of all energy storage devices in the cluster category, and the product of the difference and the number of energy storage devices in the cluster category is determined as the remaining energy storage capacity of the cluster category; the difference between the corresponding final energy storage capacity of the central energy storage device and the initial energy storage capacity is determined as the second capacity difference of all energy storage devices in the cluster category; the second capacity difference of all energy storage devices is determined as the adjustment result of the cluster category.

[0056] In one possible implementation, the adjustment module 33 is further configured to: extract the device usage time of each energy storage device from the device information of each energy storage device; extract the capacity percentage range to which the remaining lifespan of each energy storage device belongs from the lifespan prediction results based on the device usage time of each energy storage device; and determine the lower bound of the capacity range to which the remaining lifespan of each energy storage device belongs as the capacity corresponding to the remaining lifespan of each energy storage device.

[0057] In one possible implementation, the adjustment module 33 is further configured to: extract the remaining energy storage capacity or remaining energy storage space of each cluster type within the target area from the adjustment results; determine the sum of the remaining energy storage capacity of all cluster types within the target area as the total remaining energy storage capacity of the target area; calculate the ratio of the remaining energy storage space of each cluster type to the sum of the remaining energy storage spaces of all cluster types, and determine the ratio as the remaining energy storage capacity allocation ratio for each cluster type; allocate the total remaining energy storage capacity to all energy storage devices of each cluster type according to all the remaining energy storage capacity allocation ratios; for each cluster type with remaining energy storage space, determine the ratio of the remaining energy storage capacity allocated to the cluster type to the number of energy storage devices in the cluster type as the third capacity difference for each energy storage device in the cluster type; for each cluster type with remaining energy storage capacity, determine 0 as the third capacity difference for all energy storage devices in the cluster type; and the second capacity difference and the third capacity difference of all energy storage devices constitute the second energy storage capacity allocation scheme for the target area.

[0058] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device 4 in this embodiment includes a processor 40 and a memory 41. The memory 41 stores a computer program 42. When the processor 40 executes the computer program 42, it implements the steps in the various method embodiments described above. Alternatively, when the processor 40 executes the computer program 42, it implements the functions of each module / unit in the various device embodiments described above.

[0059] For example, computer program 42 may be divided into one or more modules / units, which are stored in memory 41 and executed by processor 40 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 42 in electronic device 4.

[0060] Electronic device 4 may include, but is not limited to, processor 40 and memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 4 may also include input / output devices, network access devices, buses, etc.

[0061] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0062] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0063] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A distributed photovoltaic energy storage allocation method based on lifetime prediction, characterized in that, include: The equipment information and historical usage data of each energy storage device in the energy storage system in the target area are input into a pre-trained deep learning model to obtain the lifespan prediction results of each energy storage device. Based on the historical power grid load data and historical power generation data of distributed photovoltaic power in the target area, predictions are made to obtain the predicted load data and predicted power generation data of the target area within the target time period. Based on multiple preset constraints, the predicted load data, and the predicted power generation data, a first energy storage capacity allocation scheme for the target area within the target time period is determined. Based on the lifespan prediction results of all energy storage devices, the energy storage capacity allocation information of each energy storage device in the first energy storage capacity allocation scheme is adjusted, and based on all the adjustment results, a second energy storage capacity allocation scheme for the target area is obtained.

2. The distributed photovoltaic energy storage allocation method based on lifetime prediction according to claim 1, characterized in that, The step involves adjusting the energy storage capacity allocation results for each energy storage device within the first energy storage capacity allocation scheme based on the predicted lifetime data of all energy storage devices, and obtaining a second energy storage capacity allocation scheme for the target area based on all adjustment results, including: From the first energy storage capacity allocation scheme, extract the energy storage capacity allocation information of each energy storage device; wherein, the energy storage capacity allocation information of the energy storage device includes the initial energy storage capacity, the final energy storage capacity, and the energy storage rate; Calculate the first capacity difference between the final energy storage capacity and the initial energy storage capacity of each energy storage device; Based on the equipment information and lifespan prediction results of each energy storage device, determine the corresponding capacity for the remaining lifespan of each energy storage device. The K-means clustering algorithm was used to cluster all energy storage devices based on their first capacity difference, energy storage rate, energy storage information, and remaining lifetime capacity, resulting in multiple cluster categories. For each cluster category, based on the remaining lifetime capacity of all energy storage devices within that cluster category, the first capacity difference of all energy storage devices within that cluster category is adjusted to obtain the adjustment result; Based on the adjustment results of all cluster categories, a second energy storage capacity allocation scheme for the target area is determined.

3. The distributed photovoltaic energy storage allocation method based on lifetime prediction according to claim 2, characterized in that, For each cluster category, based on the remaining lifetime capacity of all energy storage devices within that cluster category, the first capacity difference of all energy storage devices within that cluster category is adjusted to obtain the adjustment result, including: The energy storage device that is closest to the cluster center of the cluster category is identified as the central energy storage device for that cluster category. The product of the remaining lifespan capacity of the central energy storage device and its storage margin is determined as the upper limit of the energy storage capacity of the central energy storage device. Calculate the difference between the upper limit of the energy storage capacity of the central energy storage device and the final energy storage capacity; If the difference is greater than or equal to 0, then there is no need to adjust the energy storage capacity allocation information of all energy storage devices in the cluster category, and the product of the difference and the number of energy storage devices in the cluster category is determined as the remaining energy storage space of the cluster category. If the difference is less than 0, the upper limit of the energy storage capacity of the central energy storage device is determined as the final energy storage capacity of all energy storage devices in the cluster category, and the product of the difference and the number of energy storage devices in the cluster category is determined as the remaining energy storage capacity of the cluster category. The difference between the final energy storage capacity and the initial energy storage capacity of the central energy storage device is determined as the second capacity difference of all energy storage devices in this cluster category. The second capacity difference of all energy storage devices is determined as the adjustment result for this cluster category.

4. The distributed photovoltaic energy storage allocation method based on lifetime prediction according to claim 2, characterized in that, The lifespan prediction results include the capacity range corresponding to the remaining lifespan range of the energy storage device. The process of determining the remaining capacity of each energy storage device based on its equipment information and lifespan prediction results includes: Extract the usage time of each energy storage device from its device information; Based on the usage time of each energy storage device, the remaining lifespan of the energy storage device is extracted from the lifespan prediction results to determine the capacity percentage range. The lower bound of the capacity range to which the remaining lifespan of each energy storage device belongs is determined as the corresponding capacity for the remaining lifespan of each energy storage device.

5. The distributed photovoltaic energy storage allocation method based on lifetime prediction according to claim 3, characterized in that, The step of determining a second energy storage capacity allocation scheme for the target area based on the adjustment results of the energy storage capacity allocation information of all energy storage devices includes: From the adjustment results, extract the remaining energy storage capacity or remaining energy storage space for each cluster type within the target area; The sum of the remaining energy storage capacity of all cluster types within the target area is determined as the total remaining energy storage capacity of the target area; Calculate the ratio of the remaining energy storage space of each cluster type to the sum of the remaining energy storage spaces of all cluster types, and determine the ratio as the remaining energy storage capacity allocation ratio for each cluster type; The total remaining energy storage capacity is allocated to all energy storage devices in each cluster type according to the allocation ratio of all remaining energy storage capacity; For each cluster type with remaining energy storage space, the ratio of the remaining energy storage capacity allocated to the cluster type to the number of energy storage devices in the cluster type is determined as the third capacity difference for each energy storage device in the cluster type. For each cluster type with remaining energy storage capacity, 0 is defined as the third capacity difference among all energy storage devices within that cluster type; The second and third capacity differences of all energy storage devices constitute the second energy storage capacity allocation scheme for the target area.

6. The distributed photovoltaic energy storage allocation method based on lifetime prediction according to claim 1, characterized in that, The equipment information and historical usage data of each energy storage device within the energy storage system in the target area are input into a pre-trained deep learning model to obtain the lifetime prediction results for each energy storage device, including: Collect equipment information and historical usage data for each energy storage device within the energy storage system in the target area. The equipment information includes the device model and installation time; the historical usage data includes historical charging and discharging behavior data and historical fault data. For each energy storage device, extract all charging start times, SOC at all charging start times, all charging end times, and SOC at all charging end times from the charging and discharging history data of the energy storage device. Then, calculate the number of cycles for the energy storage device based on the SOC at all charging start times, SOC at all charging start times, and SOC at all charging end times. The device model, installation time, historical fault data, and cycle count of each energy storage device are input into a pre-trained deep learning model to obtain the lifespan prediction result of each energy storage device. The pre-trained deep learning model is trained using the device model, installation time, fault data, cycle count, and lifespan data of multiple energy storage devices. The lifespan data includes all lifespan intervals corresponding to the remaining lifespan of the energy storage device and the capacity interval corresponding to the remaining lifespan of each lifespan interval.

7. The distributed photovoltaic energy storage allocation method based on lifetime prediction according to claim 1, characterized in that, The step of forecasting based on historical grid load data and historical distributed photovoltaic power generation data within the target area to obtain predicted load data and predicted power generation data for the target area within the target time period includes: Historical power grid load data within the target area are input into the load forecasting model of the target area to obtain the predicted load data of the target area within the target time period; The historical power generation data and predicted environmental data of the distributed photovoltaic system are input into the photovoltaic power generation prediction model to obtain the predicted power generation data of the target area within the target time period.

8. The distributed photovoltaic energy storage allocation method based on lifetime prediction according to claim 1, characterized in that, The process of determining a first energy storage capacity allocation scheme for the target area based on multiple preset constraints, the predicted load data, and the predicted power generation data includes: With the objectives of maximizing the marginal benefits of energy storage operation, minimizing line loss, and minimizing voltage deviation, an objective function is constructed, and an energy storage capacity allocation model is constructed with branch power flow constraints, node voltage constraints, and energy storage charging and discharging power constraints as preset constraints. The predicted load data and the predicted power generation data are input into the energy storage capacity allocation model to obtain the first energy storage capacity allocation scheme for the target area.

9. A distributed photovoltaic energy storage distribution device based on lifetime prediction, characterized in that, include: The prediction module is used to input the equipment information and historical usage data of each energy storage device in the energy storage system in the target area into a pre-trained deep learning model to obtain the life prediction result of each energy storage device. The prediction module is used to make predictions based on historical grid load data and historical power generation data of distributed photovoltaic power in the target area, so as to obtain the predicted load data and predicted power generation data of the target area in the target time period. The determination module is used to determine the first energy storage capacity allocation scheme of the target area within the target time period based on multiple preset constraints, the predicted load data and the predicted power generation data. The adjustment module is used to adjust the energy storage capacity allocation information of each energy storage device in the first energy storage capacity allocation scheme according to the life prediction results of all energy storage devices, and obtain the second energy storage capacity allocation scheme for the target area based on all the adjustment results.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images