Method and apparatus for charge-discharge control of battery pack

Abstract

The application discloses a battery pack charging and discharging control method and device, and belongs to the technical field of batteries, which comprises the following steps: obtaining the quality parameters of the charging electric energy, setting a preliminary charging method, calculating the charging quality, grading the charging quality through a set quality threshold, classifying the battery packs to be charged according to the grades of the charging quality, realizing the classified storage of the charging electric energy and the reasonable configuration of the battery packs to be charged, and generating a preliminary charging strategy; constructing an identification model, setting a fluctuation identification method, identifying the fluctuation of the charging electric energy under different time scales, setting a fluctuation adjustment method, and dynamically adjusting the preliminary charging strategy according to the fluctuation type; acquiring the discharging demand in real time, setting a discharging optimization method, formulating a discharging plan once the discharging demand is detected, setting the battery discharging priority, and dynamically adjusting the discharging plan, so as to realize the efficient use of energy and guarantee the stability and efficiency of the battery discharging process.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of batteries, in particular to a battery charging and discharging control method and device for the organic combination of photovoltaic power generation, wind power generation, energy storage technology and electric vehicle charging facilities. BACKGROUND

[0002] At present, the global energy transformation is actively promoted, and the drawbacks of traditional energy are highlighted. It is urgent to vigorously develop renewable energy. Although wind power and photovoltaic power generation technologies are constantly improving, their power generation is unstable. The diversified development of energy storage technology and the rise of intelligent scheduling technology provide an opportunity to solve this problem. At the same time, the number of new energy vehicles is increasing rapidly, and the demand for charging is increasing. The traditional charging impacts the power grid. Based on this, building a battery management system that adapts to diversified energy storage technology has become the key to improving energy utilization efficiency.

[0003] The existing Chinese patent with publication number CN106169622A discloses a storage energy battery pack management method suitable for a wind-solar-diesel complementary power generation system. The invention also provides a storage energy battery pack management device suitable for a wind-solar-diesel complementary power generation system. If the technical solution of the invention is used in a wind-solar-diesel complementary power generation system, the following beneficial effects can be obtained: the charging bus and the load bus are separated, the lower voltage batteries are automatically charged first, and the higher voltage batteries are discharged first; ultimately, the voltages of the batteries are consistent; during long-term operation, the voltages of the batteries are consistent, the occurrence of undercharging or overcharging of the batteries is eliminated, and the influence of ring current caused by direct parallel connection is avoided; the working state of each battery can be monitored in real time, and the failed batteries can be detected.

[0004] Although the existing technology separates the charging and load buses and balances the voltages of the batteries, it does not consider the differences in different energy storage technology adaptation scenarios and does not explore the optimization potential of diversified energy storage technology under complex power demand and geographical environment, resulting in rigid energy storage strategies and the inability to establish an advance planning mechanism for energy supply and demand side multi-time scale fluctuations, making it difficult to ensure power supply and demand balance and stability. Therefore, the present application provides a battery charging and discharging control method and device. SUMMARY

[0005] In view of the deficiencies of the prior art, the purpose of the present application is to provide a battery charging and discharging control method and device, which dynamically adjusts the coordination strategy of lithium batteries and lithium titanate batteries by identifying the charging quality and energy supply and demand fluctuations, and solves the efficiency and stability problems when multiple types of batteries are used together.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0007] The battery pack charging and discharging control method comprises:

[0008] Obtaining the quality parameters of the charging power, setting the preliminary charging method, calculating the charging quality, and grading the charging quality through the set quality threshold, and classifying the battery pack to be charged according to the grade of the charging quality, and generating the preliminary charging strategy;

[0009] Building a recognition model, setting a fluctuation recognition method, recognizing the fluctuation of the charging power at different time scales, and setting a fluctuation adjustment method, and dynamically adjusting the preliminary charging strategy according to the fluctuation type;

[0010] Real-time acquisition of discharge demand, setting a discharge optimization method, once the discharge demand is detected, formulating a discharge plan, and setting a battery discharge priority, and dynamically adjusting the discharge plan.

[0011] Further, in generating the preliminary charging strategy, the preliminary charging method comprises:

[0012] Real-time acquisition of the quality parameters of the charging power, including the actual voltage, the actual frequency of the output power, calculation of the real-time voltage deviation rate and the frequency deviation rate;

[0013] Obtaining the effective value U a of each harmonic voltage, calculating the harmonic distortion rate, wherein U1 is the effective value of the fundamental voltage, and a = 1, 2, 3, …;

[0014] Assigning different weights to the voltage deviation rate, the frequency deviation rate and the harmonic distortion rate, and calculating the comprehensive quality score Q RT .

[0015] Further, the preliminary charging method further comprises:

[0016] Setting the sampling interval to t1, collecting the charging quality data in the preset time period, and dividing the charging quality data into a training set and a test set in chronological order;

[0017] Obtaining the tolerance range of the charging quality of the electrical equipment;

[0018] Comprehensively considering the statistical characteristics of the data itself and the equipment tolerance range, the threshold values of the voltage deviation rate, the frequency deviation rate and the harmonic distortion rate are preliminarily set;

[0019] Calculating the preliminary threshold of the comprehensive quality score, including the first quality threshold Q th1.ele , the second quality threshold Q th2.ele ;

[0020] Calculating the comprehensive quality score Q testand adjusting the preliminary threshold value using the test set to generate a quality threshold value, including a first quality threshold value Q th1 and a second quality threshold value Q th2 ;

[0021] According to the quality threshold value, the comprehensive quality score Q RT is divided to determine the category of charging quality; if Q RT <Q th1 , it is determined as high-quality electric energy; if Q th1 ≤Q RT <Q th2 , it is determined as sub-quality electric energy; if Q RT ≥Q th2 , it is determined as low-quality electric energy.

[0022] According to the category of charging quality, the battery pack to be charged is divided into three categories, namely, a first-class battery, a second-class battery, and a third-class battery, the number of each type of battery is calculated, and a preliminary charging strategy is generated.

[0023] Further, the specific steps of configuring the battery pack to be charged include:

[0024] The first-class battery adopts a combination of lithium ion batteries and lithium titanate batteries, and the number N 1.Li of lithium ion batteries and the number N LTO of lithium titanate batteries in the first-class battery are calculated.

[0025] Lithium ion batteries are used as the second-class battery, and the number N 2.Li of second-class batteries is calculated.

[0026] Lithium ion batteries are used as the third-class battery, and the number N 3.Li of third-class batteries is calculated.

[0027] Further, when identifying the fluctuation of charging electric energy, the fluctuation identification method includes:

[0028] Collecting historical fluctuation data and cleaning the collected historical fluctuation data;

[0029] Extracting short-term fluctuation features, medium-term fluctuation features, and long-term fluctuation features from the historical fluctuation data;

[0030] Using a regression logic algorithm to construct a short-term fluctuation identification model, with 0 indicating the existence of short-term fluctuation and 1 indicating the absence of short-term fluctuation; and using a regression logic algorithm to construct a long-term fluctuation identification model, with 1 indicating the existence of long-term fluctuation and 0 indicating the absence of long-term fluctuation;

[0031] Adding an adder after the short-term fluctuation identification model and the long-term fluctuation identification model to generate an identification model.

[0032] The fluctuation data to be determined is input into the identification model to perform fluctuation determination, where 0 represents short-term fluctuation, 1 represents medium-term fluctuation, and 2 represents long-term fluctuation, and an output of a final prediction fluctuation determination category.

[0033] Further, when dynamically adjusting the preliminary charging strategy, the fluctuation adjustment method comprises:

[0034] When the prediction fluctuation determination category is short-term fluctuation, the energy supply-demand difference ΔE diff.1 in the short term is calculated, and the adjustment amount ΔN LTO of the lithium ion battery is calculated.

[0035] When the prediction fluctuation determination category is medium-term fluctuation, the energy supply-demand difference ΔE diff.2 in the medium term is calculated, and the adjustment amount ΔN 1.Li , ΔN 2.Li , ΔN 3.Li of the lithium ion battery is calculated.

[0036] When the prediction fluctuation determination category is long-term fluctuation, the energy supply-demand difference ΔE diff.3 is calculated, and the adjustment amount ΔN' LTO , ΔN' 1.Li , ΔN' 2.Li , ΔN' 3.Li of each type of battery is calculated.

[0037] Further, when dynamically adjusting the discharge plan, the discharge optimization method comprises:

[0038] When a discharge demand is detected, the discharge parameters of the battery pack to be charged are read, and the predicted discharge power P ex and the predicted discharge time t ex are calculated.

[0039] The proportions r1, r2, and r3 of different charging qualities in the future period are calculated.

[0040] The r1, r2, and r3 are compared, the battery category corresponding to the charging quality with the highest proportion is set as the first discharge priority, the charging quality with the second highest proportion is set as the second discharge priority, and the remaining charging qualities are set as the third discharge priority.

[0041] For the battery category determined as the first discharge priority, the current remaining capacity E S of the battery is calculated, and it is determined whether the battery of the first discharge priority meets the discharge demand.

[0042] If E S ≥ P ex × t ex and P max.1≥P max , then it is determined that the discharge requirement is met, and a discharge operation is performed; wherein, P max.1 is the maximum discharge power;

[0043] Otherwise, the discharge requirement is not met, and the battery of the second discharge priority is judged, and so on.

[0044] Further, the discharge optimization method further comprises:

[0045] In the charging and discharging process, the discharge current I im and the discharge voltage V im are collected in real time, and the standard discharge curve data is obtained;

[0046] When the discharge starts, the constant current discharge phase is entered, so that the discharge current is maintained at I con , and the charging power is P in.1 ;

[0047] When the discharge voltage reaches V con , the constant voltage discharge phase is switched to, the discharge voltage is kept constant at V con , and the charging power is P in.2 ;

[0048] When the discharge current drops to I cut , it is determined that the discharge is completed, the actual discharge time t ac is obtained, the time deviation rate Δt rate , the total discharge cost C total and the energy utilization rate η use are calculated, and the feedback data is generated.

[0049] Further, the battery charging and discharging control method further comprises:

[0050] In the battery charging and discharging process, the battery group to be charged is monitored, the battery management method is set, and the stable work of the battery group is maintained;

[0051] The battery management method comprises:

[0052] Based on the sensors equipped for each battery monomer, the management parameters are obtained in real time;

[0053] The battery electrochemical model is established through offline experiments, and the ampere-hour integral is corrected in combination with the temperature correction coefficient;

[0054] The battery internal resistance change rate and the capacity retention rate are analyzed, the capacity attenuation model is constructed, and the predicted remaining life of the battery is output;

[0055] When the battery monomer voltage deviation exceeds the preset voltage, the active balancing mechanism is triggered;

[0056] The liquid cooling pipeline design was optimized through CFD simulation, and a PID controller was configured to adjust the coolant flow and temperature in real time.

[0057] Establish a three-level early warning system to provide early warning of battery failures.

[0058] The battery pack charging and discharging control device includes: generator set, energy storage battery and charging pile;

[0059] The generator set includes a wind turbine and photovoltaic panels;

[0060] The energy storage battery includes a lithium-ion battery pack, a lithium titanate battery pack, and a bidirectional DC / DC converter.

[0061] The beneficial effects of this invention are:

[0062] By employing a tiered configuration and dynamic adjustment strategy for various types of lithium-ion batteries, the problem of efficient storage and release of energy of different qualities is solved. For high-quality energy, a combination of lithium-ion batteries and lithium titanate batteries is used to ensure stable discharge and compensate for the insufficient power density of a single lithium-ion battery. For medium-quality energy, the charging quality is optimized through the charging and discharging process of the lithium-ion batteries themselves, reducing reliance on external processing. Low-quality energy is stored in low-cost lithium-ion batteries after external processing, balancing cost and performance. At the same time, based on differences in battery characteristics, the number of lithium titanate batteries is dynamically adjusted during short-term fluctuations to quickly respond to power demands, while the lithium-ion battery configuration is optimized during medium- and long-term fluctuations, taking into account both cost and cycle life. Attached Figure Description

[0063] Figure 1 Flowchart of the charging and discharging control method for a battery pack;

[0064] Figure 2 This is a flowchart of the preliminary charging method of the present invention;

[0065] Figure 3 This is a flowchart of the discharge optimization method of the present invention;

[0066] Figure 4 This is a diagram of an integrated wind, solar, energy storage, and charging device. Detailed Implementation

[0067] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0068] Example 1

[0069] refer toFigures 1 to 3 As shown, this embodiment introduces a battery pack charging and discharging control method, including:

[0070] The system acquires electrical energy generated by external power generation equipment to obtain the quality parameters of the energy to be charged. This energy is the power generated by the equipment but has not yet been classified by quality. It is used to charge battery packs of the corresponding category. These battery packs include batteries that have not been charged or have zero charge. A preliminary charging method is set up to calculate the charging quality of the energy to be charged. The energy is then classified into three levels based on a set quality threshold: high-quality, medium-quality, and low-quality. The battery packs are also categorized according to these levels to achieve classified storage of charging energy and rational configuration of battery packs. A preliminary charging strategy is generated, and a battery management method is implemented. This method involves multi-dimensional real-time monitoring and intelligent control of the battery packs to ensure stable battery operation. The battery packs primarily use lithium-ion batteries and are equipped with lithium titanate batteries to meet high power demands in short periods, achieving diversified energy storage.

[0071] A recognition model is built by using multiple binary classifiers, and a fluctuation recognition method is set to identify the fluctuation of charging energy at different time scales. At the same time, a fluctuation adjustment method is set to dynamically adjust the initial charging strategy according to the fluctuation type, rationally plan the number of various types of batteries, and perform fluctuation adjustment to charge the planned battery pack to be charged, and save the charging energy of different qualities into the corresponding batteries.

[0072] It captures charging demands from external devices in real time, such as charging piles and grid demands, and converts them into battery discharge demands. It sets up discharge optimization methods, formulates discharge plans and sets battery discharge priorities once a discharge demand is detected. It monitors and flexibly adjusts the discharge plan in real time during the battery discharge process to achieve efficient energy utilization and ensure the stability and efficiency of the battery discharge process.

[0073] The battery pack's operating parameters are collected in real time by various sensors. The collected real-time data is compared with the preset normal operating thresholds to determine if there are any abnormalities. Once an abnormality or fault is detected, the corresponding redundancy control and protection mechanisms are immediately activated, such as cutting off the power to the faulty part and starting the backup equipment. This is to prevent the fault from escalating and to ensure the safety of personnel and equipment. At the same time, the fault information is recorded to provide a basis for subsequent fault analysis and repair.

[0074] In this embodiment, the integrated wind-solar-storage-charging system mainly includes: collecting solar and wind energy and converting them into electrical energy. Due to the intermittent and unstable nature of solar and wind energy, grid power is introduced as a backup power source to supplement power generation when wind and solar power generation is insufficient, reducing charging interruptions caused by fluctuations in wind and solar power generation. To fully utilize clean energy, a wind and solar priority power generation strategy is implemented, prioritizing the use of wind and solar power to charge the energy storage device. This allows for the activation of grid power generation and the release of electrical energy when wind and solar power generation is insufficient, maximizing the use of clean energy, reducing dependence on grid power, lowering electricity costs, and aligning with environmental protection principles. The solar energy portion is achieved by installing solar panels (photovoltaic panels). Photovoltaic panels utilize the photoelectric effect to convert the photon energy of sunlight into electrical energy, forming direct current output. The wind energy portion relies on wind turbines, powered by wind. The wind turbine blades rotate, driving the generator to produce alternating current (AC). Simultaneously, the system monitors the real-time operating status of the solar photovoltaic panels and wind turbine, collecting power generation data from both via sensors. Utilizing a bidirectional full-bridge topology, it converts AC to DC, storing the AC power generated by the wind in a DC energy storage device. When the DC power from the storage device is fed back to the AC grid or used to power AC loads, inverter technology converts it into AC power that meets grid requirements. Furthermore, the system flexibly adjusts the output active and reactive power based on real-time grid demands, ensuring stable interaction between the system and the grid. It also stores excess energy generated by the power generation module for use during periods of insufficient power generation or peak demand. Lithium batteries are used as the primary energy storage device, supplemented by supercapacitors to handle short-term high power demands, forming a diversified energy storage system.

[0075] Furthermore, when generating the initial charging strategy, the initial charging method includes:

[0076] Real-time acquisition of the quality parameters of the energy to be charged, including the actual voltage U of the energy to be charged. RT and the actual frequency f RT Simultaneously, it acquires benchmark parameters used to measure actual charging quality, including rated voltage U0 and rated frequency f0, and calculates the real-time voltage deviation rate δU. RT and frequency deviation rate δf RT Voltage deviation rate reflects the degree to which the actual voltage deviates from the rated voltage. The smaller the voltage deviation rate, the higher the voltage stability. It is a benchmark parameter for measuring the actual charging quality. Frequency deviation rate measures the frequency stability of the energy being charged. The expression is shown below:

[0077]

[0078] The effective value U of each harmonic voltage is obtained using harmonic detection equipment. aWhere U1 is the effective value of the fundamental voltage, and a = 1, 2, 3, ..., calculate the harmonic distortion rate (THD). U Harmonic distortion rate is used to assess the harmonic content in the electrical energy to be charged; the expression is shown below:

[0079]

[0080] To comprehensively evaluate the quality of the power being charged, a comprehensive quality score is introduced, integrating multiple quality parameters into a single quantitative indicator for a more intuitive and unified assessment of charging quality. Weighting coefficients w1, w2, and w3 are assigned to voltage deviation rate, frequency deviation rate, and harmonic distortion rate, respectively, and the comprehensive quality score Q is calculated. RT This achieves the quantification of indicators; the expression is shown below:

[0081] Q RT =w1×δU RT +w2×δf RT +w3×THD U

[0082] In the formula, w1 + w2 + w3 = 1, and the overall quality score Q is... RT The lower the value, the higher the charging quality; in this embodiment, w1 = 0.4, w2 = 0.3, and w3 = 0.3 are selected.

[0083] The sampling interval is set to t1. Historical data of various quality parameters are collected within a preset time period (e.g., at least one year), including voltage deviation rate, frequency deviation rate, and harmonic distortion rate data. Each data point is accurately timestamped to comprehensively cover the charging energy situation under different seasons, weather conditions, and various operating conditions. This includes multi-scale time information from short-term (e.g., the impact of instantaneous weather changes on charging energy) to long-term (e.g., changes in charging energy due to seasonal changes), ensuring broad data representativeness. The voltage deviation rate data sequence is {δU1,…,δU...}. b The frequency deviation rate data sequence is {δf1,…,δf}. b The harmonic distortion rate data sequence is: b represents the total number of data points, and each data point is also marked with an accurate timestamp for subsequent time series analysis.

[0084] To ensure that stored and used electrical energy meets the normal operating requirements of equipment and to avoid damage caused by charging quality issues, consult the technical manuals of critical electrical equipment, such as electric vehicles, to obtain the equipment's tolerance range for charging quality, including the equipment's voltage deviation rate within ±δU. app Within ±δf, the equipment frequency deviation rate is within ±δf app The harmonic distortion rate of the equipment within THD appWithin this range, according to GB / T 18487.1-2015 and mainstream electric vehicle technical manuals, such as the Tesla V3 Supercharger technical specifications, the tolerance range of key electrical equipment to charging quality is set as follows: voltage deviation rate within ±5%, frequency deviation rate within ±0.4%, and harmonic distortion rate not exceeding 3%. At this time, f0 = 50Hz is set.

[0085] The charging quality data is divided into a training set and a test set in chronological order. The training set is used to initially set the threshold, and the test set is used to independently verify the accuracy of the threshold.

[0086] Taking into account the statistical characteristics of the data itself and the equipment's tolerance range, preliminary thresholds were set for voltage deviation rate, frequency deviation rate, and harmonic distortion rate, including a first voltage threshold. Second voltage threshold Frequency first threshold Second threshold of frequency First harmonic threshold Harmonic second threshold in, σ represents the mean of voltage deviation rate, frequency deviation rate, and harmonic distortion rate in the training set, respectively. δU σ δf σ THD These represent the standard deviations of the voltage deviation rate, frequency deviation rate, and harmonic distortion rate in the training set. For example, if the mean of the voltage deviation rate is 1.0% and the standard deviation is 0.8%, the mean of the frequency deviation rate is 0.2% and the standard deviation is 0.1%, and the mean of the harmonic distortion rate is 1.5% and the standard deviation is 0.5%, then the first voltage threshold δU is... th1 =max{1.0% + 0.8% - 5%} = 1.8%, the second voltage threshold δU th2 =min{1.0% + 2 × 0.8%, 5%} = 2.6%, and δf can be calculated similarly. th1 =0.3%, δf th2 =0.4%, THD th1 =2.0%, THD th2 =2.5%;

[0087] Based on the comprehensive quality scoring formula, the thresholds for voltage deviation rate, frequency deviation rate, and harmonic distortion rate, which were initially set above, are substituted into the formula to calculate the preliminary threshold for the comprehensive quality score, including the first quality threshold Q. th1.ele Second mass threshold Q th2.eleThis system converts specific parameter thresholds into comprehensive quality score thresholds, facilitating direct grading of charging quality based on the comprehensive quality score. Specifically, a first quality threshold is used to distinguish between high-quality and sub-quality charging energy, while a second quality threshold is used to distinguish between sub-quality and low-quality charging energy. Based on the actual calculated values ​​of these thresholds, Q can be calculated. th1.ele =0.4×1.8%+0.3×0.3%+0.3×2.0%=1.41%, Q th2.ele =0.4×2.6% + 0.3×0.4% + 0.3×2.5% = 1.91%;

[0088] The initially set thresholds were inaccurate. Through evaluation of the test set data and threshold adjustments, the thresholds were made more consistent with actual conditions, improving the accuracy of charging quality grading. Using the set initial thresholds, the test set data was classified into quality levels, and the initial thresholds were adjusted based on the feedback from the level classification. For each data point in the test set, a comprehensive quality score Q was calculated based on the voltage deviation rate, frequency deviation rate, and harmonic distortion rate. test Using accuracy as the evaluation metric, adjustments were made in increments of ΔQ1 and ΔQ2 to adjust the initial threshold. The evaluation metric for the test set was then recalculated, and the adjustment effect was observed until the evaluation metric reached a satisfactory level. Finally, the quality threshold, including the first quality threshold Q, was output. th1 Second mass threshold Q th2 ;

[0089] Based on the set quality thresholds, charging quality is categorized, and clear charging quality categorization standards allow for different processing methods to be adopted according to different charging quality levels; if Q RT th1 If the energy is determined to be high-quality electrical energy, it can be directly used to charge the battery, ensuring efficient charging and discharging and a long battery life; if Q... th1 ≤Q RT th2 If Q is determined to be substandard energy, it will be converted into high-quality energy after internal optimization of the battery to be charged (such as adjusting charging and discharging parameters); RT ≥Q th2 If the energy is identified as low-quality energy, it will be processed by external equipment (such as filtering and voltage regulation) before the battery is charged.

[0090] Configure the battery packs to be charged according to the charging quality level, divide the battery packs to be charged into three categories: Category I, Category II and Category III batteries, calculate the number of batteries in each category, and generate a preliminary charging strategy.

[0091] Furthermore, the specific steps for configuring the battery pack to be charged include:

[0092] ​​High-quality electrical energy is discharged directly without processing to power the equipment and is stored in Class I batteries; medium-quality electrical energy is discharged after being optimized by the battery itself to power the equipment and is stored in Class II batteries; low-quality electrical energy is processed by external equipment, charged into the battery, and then discharged after being optimized by the battery itself to power the equipment and is stored in Class III batteries.

[0093] Because lithium-ion batteries have high energy density, they are suitable for storing large amounts of electrical energy for extended periods. Lithium titanate batteries, on the other hand, have rapid charge and discharge characteristics, enabling them to quickly respond to sudden power demands and compensate for the low power density of lithium-ion batteries. One type of battery combines lithium-ion and lithium titanate batteries. In certain special circumstances, such as simultaneous rapid charging of electric vehicles or extreme weather causing a sudden increase in battery discharge demand, lithium-ion batteries alone cannot quickly meet all discharge needs. The presence of lithium titanate batteries provides additional power support in these sudden situations, ensuring smooth battery discharge. The number of batteries in this type is determined based on the expected high-quality electrical energy generation, discharge demand, and individual battery cell capacity, as shown in the following expression:

[0094]

[0095] In the formula, N 1.Li N represents the number of lithium-ion batteries in a class of batteries. LTO E represents the number of lithium titanate batteries. dis To meet the discharge requirements throughout the day, n Li C represents the number of batteries in a lithium-ion battery cluster. Li E1 represents the capacity of a single lithium-ion battery cell, E1 represents the predicted capacity of high-quality electrical energy, and C represents the capacity of a single lithium-ion battery cell. LTO For the capacity of a single lithium titanate battery cell, η LTO For power response efficiency, n LTO K is the number of batteries in the lithium titanate battery cluster, and k1 is the redundancy coefficient, which is used to deal with sudden charging needs or equipment performance fluctuations. The value range is [1.2, 1.5]. In this embodiment, k1 = 1.3 is taken.

[0096] Since the secondary mass of electrical energy can be optimized to some extent through the charging and discharging process of the lithium-ion battery itself, lithium-ion batteries are used as the second type of battery, and the number of second-type batteries is calculated, as shown in the following expression:

[0097]

[0098] In the formula, N 2.LiK represents the number of lithium-ion batteries in the second-class battery, k2 is the loss coefficient, which takes into account the energy loss caused by internal chemical reactions and resistance during the storage and optimization of submass energy of lithium-ion batteries, and the value range is [1.1, 1.3]. In this embodiment, k2 = 1.2 is taken, and E2 is the predicted amount of submass energy.

[0099] Since low-quality electrical energy needs to be processed by external equipment, the requirements for batteries are relatively low. Therefore, lower-cost lithium-ion batteries are selected as the third type of battery, and the quantity of each type of battery is calculated as shown in the following expression:

[0100]

[0101] In the formula, N 3.Li E3 represents the number of lithium-ion batteries among the three types of batteries. k3 is a comprehensive coefficient that takes into account the energy loss caused by the conversion efficiency and transmission loss of low-quality electrical energy when it is processed by external equipment, as well as the redundancy required to ensure storage stability and cope with emergencies. The value range is [1.3, 1.5]. In this embodiment, k3 = 1.4 is taken. E3 is the predicted amount of low-quality electrical energy.

[0102] E1, E2, and E3 are all determined by the energy prediction model. This model is constructed using a neural network, taking collected meteorological data and historical charging energy data as input. The trained model performs calculations, dividing the day into t' time periods, and outputting the predicted energy E for different time periods of the day. pre Based on the charging quality classification results, from the predicted charge E pre The predicted quantities of high-quality electrical energy E1, secondary-quality electrical energy E2, and low-quality electrical energy W3 are separated from the energy.

[0103] In this embodiment, a day is divided into 24 time periods, with one prediction period per hour. Assuming the total daily power generation of a certain wind-solar-storage-charging system is 10000 kWh, the charging quality classification results are: high-quality energy W1 = 6000 kWh, medium-quality energy W2 = 3000 kWh, and low-quality energy W3 = 1000 kWh, and the total daily discharge demand E... dis =4000kWh, with a sudden power demand of 150kWh, and a rapid response within 10 minutes;

[0104] The lithium-ion battery selected is the 300Ah lithium iron phosphate cell with the largest capacity currently in mass production, such as CATL's CTP technology, with a voltage of 3.2V and an energy capacity of 0.96kWh. Due to material limitations, the single-cell capacity of lithium titanate batteries is usually smaller than that of lithium-ion batteries. The largest 100Ah model from Toshiba's SCiB series is selected, which supports charge and discharge rates of 10C and above, with a voltage of 2.4V, an energy capacity of 0.24kWh, and a power response efficiency of 0.9. Because the cells are directly connected, the wire length and number of contacts are huge, resulting in a high failure rate. Usually, battery clusters are formed by connecting them in series / parallel to reduce the number of cells while meeting the system's voltage, current, and energy requirements. Common voltages for energy storage systems include 512V and 1000V. Taking a system requiring 512V as an example, the number of lithium-ion batteries connected in series is 160, and the number of lithium titanate batteries connected in series is 213 (here, the ratio of the system's required voltage to the single-cell voltage is rounded to the nearest integer).

[0105] At this point, calculate the number of individual cells in each type of battery, N. 1.Li =8125 sections, N LTO =36111 sections, N 2.Li =3594 sections, B 3.Li = 1563 cells. After the configuration of the battery clusters, there are 51 lithium-ion battery clusters and 170 lithium titanate battery clusters in the first type of battery, 23 clusters in the second type of battery, and 10 clusters in the third type of battery.

[0106] Lithium titanate batteries can release 7780kWh in 10 minutes, meeting a sudden demand of 1500kWh, with a response time of no more than 2 seconds. In terms of energy utilization, the direct discharge efficiency of high-quality energy is 98%, the efficiency of optimized medium-quality energy is 92%, and the efficiency of low-quality energy after processing is 80%. At the same time, compared with a single lithium-ion battery solution, the hybrid configuration reduces costs by 18%. Lithium titanate accounts for only 35% of the total number of batteries, but undertakes high-frequency charging and discharging tasks, extending the life of the main battery.

[0107] Furthermore, battery management methods include:

[0108] To comprehensively and accurately understand the battery's operating status, high-precision voltage sensors are deployed at both ends of each battery cell, temperature sensors are installed at the tabs, and Hall current sensors are configured in the battery pack circuit to acquire management parameters in real time, such as cell voltage, temperature, and charge / discharge current, with a sampling frequency of f. EMS ;

[0109] Considering that battery capacity can vary significantly under different operating conditions, such as different charge / discharge rates and ambient temperatures, a detailed electrochemical model of the battery was established through offline experiments to accurately estimate the battery's state of charge (SOC). This model describes the battery's performance under different operating conditions. Combined with a temperature correction factor, the ampere-hour integral is corrected, and the SOC estimation error is controlled within ±3%. The temperature correction factor was obtained by fitting a large amount of experimental data to dynamically adjust the ampere-hour integral based on real-time measured temperature values, thereby improving the accuracy of SOC estimation.

[0110] To predict the state of health (SOH) of a battery, we analyze the rate of change of internal resistance and the rate of capacity retention. As the battery is used, its internal resistance gradually increases and its capacity gradually decreases. By monitoring and analyzing the changing trends of these parameters over a long period of time, we construct a capacity decay model and predict the remaining life of the battery based on the current internal resistance and capacity data, providing a scientific basis for battery maintenance and replacement.

[0111] During the use of the battery pack, differences in manufacturing processes and usage environments between individual battery cells can lead to voltage deviations. When the voltage deviation exceeds ±50mV, an active balancing mechanism based on the Buck-Boost circuit is triggered. During the active balancing process, the Buck-Boost circuit transfers energy from the higher-voltage battery cells to the lower-voltage battery cells, thereby achieving voltage balance among the individual cells in the battery pack.

[0112] To ensure that the battery operates within a suitable temperature range, the liquid cooling pipeline design is optimized through CFD simulation, and a PID controller is configured to adjust the coolant flow and temperature in real time. For example, when the battery temperature difference exceeds 5°C, forced air cooling is activated to assist heat dissipation, ensuring that the highest temperature inside the battery pack does not exceed 45°C.

[0113] To promptly detect potential battery faults and ensure the safe and stable operation of the system, a three-level early warning system is established to provide fault warnings for the battery. The first-level warning (SOC>90% or<10%) triggers charging current limiting; the second-level warning (single cell voltage>4.15V or<2.7V) initiates passive equalization; and the third-level warning (temperature difference>10℃ or internal resistance sudden change>15%) automatically disconnects the main contactor and reports a fault code.

[0114] Furthermore, when identifying fluctuations in charging power, the fluctuation identification methods include:

[0115] Collect historical fluctuation data from databases or related record backends. This data should have a sufficiently long time span to ensure that it covers different types of fluctuations. Clean the collected historical fluctuation data to remove outliers, missing values, and erroneous data.

[0116] Short-term, medium-term, and long-term fluctuation characteristics are extracted from historical fluctuation data. Based on the extracted fluctuation characteristics, the historical fluctuation data is divided into short-term, medium-term, and long-term fluctuations. Short-term fluctuations are short in duration, not exceeding 1 hour, such as power fluctuations caused by wind turbine start-up and shutdown. Characteristics include the maximum rate of change of voltage, current, and frequency, fluctuation amplitude (the difference between the maximum and minimum values), and fluctuation frequency (the number of fluctuations per unit time) within the short time window. Medium-term fluctuations are longer than 1 hour but not more than 1 day, such as diurnal changes in sunlight. Characteristics include the average fluctuation amplitude, fluctuation trend (rising, falling, or stable), and periodicity of fluctuations within the medium-term time window. Long-term fluctuations are longer than 1 day, such as seasonal changes in power generation. Characteristics include the average power parameter values, power parameter change trends, and seasonal fluctuation characteristics within the long-term time window.

[0117] A short-term fluctuation identification model is constructed using a regression logic algorithm. The model is trained using known short-term fluctuation data from historical fluctuation data. The model parameters are adjusted, and historical fluctuation data is used as the input feature vector. The output result is whether short-term fluctuations exist, with 0 indicating the presence of short-term fluctuations and 1 indicating the absence of short-term fluctuations.

[0118] A long-term fluctuation identification model is constructed using a regression logic algorithm. The model is trained using known long-term fluctuation data from historical data. The model parameters are adjusted, and historical fluctuation data is used as the input feature vector. The output result is whether long-term fluctuation exists. 1 indicates the existence of long-term fluctuation, and 0 indicates the absence of long-term fluctuation, i.e., medium-term fluctuation exists.

[0119] Add an adder after the short-term volatility identification model and the long-term volatility identification model to generate the identification model;

[0120] The fluctuation data to be determined is input into the identification model, and the output results are obtained for fluctuation determination. 0 indicates the existence of short-term fluctuations, 1 indicates the existence of medium-term fluctuations, and 2 indicates the existence of long-term fluctuations. Since it is difficult to achieve a completely fluctuation-free situation in practice, the case of no fluctuations is not considered here. The fluctuation data to be determined is the predicted output value of the power prediction model.

[0121] Based on the prediction results of the volatility assessment, the final predicted volatility assessment category is output, including short-term volatility, medium-term volatility, and long-term volatility.

[0122] Furthermore, when dynamically adjusting the initial charging strategy, the fluctuation adjustment method includes:

[0123] When the predicted fluctuation is classified as short-term fluctuation, lithium titanate batteries, due to their high-quality stored energy, can more efficiently and stably meet discharge demands or balance energy surplus in the face of short-term fluctuations. In contrast, lithium-ion batteries store lower-quality energy, resulting in insufficient response speed and stability in dealing with short-term fluctuations, making it difficult to quickly and accurately match changes in short-term discharge demand. Therefore, the energy supply-demand difference ΔE under short-term fluctuations needs to be calculated. diff.1 And the adjustment amount ΔN of lithium titanate batteries LTO The expression is as follows:

[0124] ΔE diff.1 =E charge.1 -E power.1

[0125]

[0126] In the formula, E charge.1 This refers to the energy required for discharge within short-term fluctuations, and T' represents the number of time periods that a day is divided into, and E power.1 The energy that the battery pack can supply during short-term fluctuations is determined by the energy prediction model. It is the floor function; when ΔN LTO ≥0, discharge demand exceeds available energy, the additional number of lithium titanate batteries is ΔN. LTO To meet the discharge demand during future short-term fluctuations, when ΔN LTO <0, indicating excess energy supply, resulting in a reduction of -ΔN in the number of lithium titanate batteries. LTO To avoid over-discharge leading to battery life degradation, increasing the number of batteries results in more energy being stored. However, if this energy is already excessive, further storage would be wasteful and could pose safety risks, such as ΔE. diff.1 =50kWh, at this point, demand is excessive and the number of lithium titanate batteries needs to be reduced, then ΔN LTO = -2 clusters, so 2 clusters need to be reduced;

[0127] When the predicted fluctuation category is classified as medium-term fluctuation, the storage quantity of lithium-ion batteries is adjusted based on the adaptability of battery characteristics and application scenarios. While lithium titanate batteries have high energy density and charge / discharge efficiency, their high cost and relatively short cycle life make them more suitable for short-term, high-frequency fluctuation scenarios with extremely high energy response speed requirements, such as dealing with instantaneous peak electricity demand. In comparison, lithium-ion batteries have advantages in cost, cycle life, and energy storage characteristics, and can better adapt to the needs of medium-term fluctuation adjustments. The energy supply and demand difference ΔE under medium-term fluctuations is calculated. diff.2 And the adjustment amount ΔN of the lithium-ion battery 1.Li ΔN 2.Li ΔN3.Li The expression is as follows:

[0128] ΔE diff.2 =E charge.2 -E power.2

[0129]

[0130] In the formula, ΔN 1.Li ΔN represents the adjustment amount for lithium-ion batteries in a certain type of battery. 2.Li For the adjustment amount of the second type of battery, ΔN 3.Li For the adjustment amount of the three types of batteries, E power.2 The energy that the battery pack can supply during medium-term fluctuations is determined by the energy prediction model, E. charge.2 This refers to the energy required for discharge during medium-term fluctuations, and T'1 represents the number of time periods encompassed by the medium-term fluctuation. If it is predicted that the available energy supply will gradually decrease while the discharge demand will gradually increase during the medium-term fluctuation, the storage capacity of lithium-ion batteries will be increased to reserve energy in advance; conversely, the storage capacity will be reduced, for example, ΔE diff.2 = -200kWh, at this point the demand is insufficient and more lithium-ion batteries need to be added, then ΔN 1.Li =ΔN 2.Li =ΔN 3.Li = -4 clusters, each type needs to reduce 4 clusters;

[0131] When the predicted fluctuation is classified as a long-term fluctuation, the total storage quantity of various types of batteries is adjusted on a large scale based on the forecast of long-term energy supply and discharge demand, combined with seasonal factors; the energy supply and demand difference ΔE is calculated. diff.3 And the adjustment amounts for various types of batteries, expressed as follows:

[0132] ΔE diff.3 =E charge.3 -E power.3

[0133]

[0134] In the formula, ΔN' LTO , ΔN' 1.Li ΔN' represents the adjustment amount for lithium titanate batteries and lithium-ion batteries in a certain type of battery. 2.Li The adjustment amount for Class II batteries is ΔN'. 3.Li For the adjustment amount of the three types of batteries, E power.3 The energy that the battery pack can supply under long-term fluctuations is determined by the energy prediction model, E. charge.3 This refers to the energy required for discharge under long-term fluctuations, and T'2 represents the number of time periods encompassed by the long-term fluctuation; for example, a decrease in power generation during winter, ΔEdiff.3 = -500kWh, then ΔN' LTO = -20 clusters, ΔN' 1.Li =ΔN' 2.Li =ΔN' 3.Li = -10 clusters.

[0135] Furthermore, the discharge optimization methods include:

[0136] Once the battery pack has finished charging, the system monitors external charging demands. When a charging request is detected from an external charging device, key parameters are read from the device's built-in communication equipment and converted into discharge parameters for the battery pack, including the rated discharge capacity C. rated State of charge (SOC) of external devices now Maximum permissible discharge current I max and rated discharge voltage V rated Combined with the maximum output power P of the external device max Calculate the expected discharge power P ex and the expected discharge time t ex The expression is as follows:

[0137] P ex =min(P max ,I max ×V rated )

[0138]

[0139] In the formula, η is the discharge efficiency, and also the charging efficiency of the external device, such as the charging efficiency of the charging pile itself.

[0140] Call the power prediction model to obtain the predicted amount of high-quality power E in the future. ' 1. Predicted quantity E of secondary mass electrical energy ' 2 and the predicted amount of low-quality electrical energy E ' 3. Calculate the percentage of different charging qualities in future time periods, as shown in the following expression:

[0141]

[0142] In the formula, r1 is the proportion of high-quality electrical energy, r2 is the proportion of medium-quality electrical energy, and r3 is the proportion of low-quality electrical energy.

[0143] Compare these three ratios r1, r2, and r3. Set the battery category corresponding to the highest charging quality ratio as the first discharge priority, the second highest ratio as the second discharge priority, and the rest as the third discharge priority. For example, if the charging quality ratio in the future period is r1 = 50%, r2 = 30%, and r3 = 20%, then the first priority is a type of battery with an expected discharge power of 100kW and an expected time of 2 hours.

[0144] For the battery category determined to have the highest discharge priority, obtain the number of available batteries M, the capacity of each individual battery C1, the current state of charge (SOC1), and the maximum discharge power P. max.1 Calculate the current remaining battery capacity E. S The expression is as follows:

[0145] E S =M×C1×SOC1

[0146] In the formula, if the battery category with the first discharge priority is a Class I battery, the remaining capacity of the lithium titanate battery and the lithium-ion battery are calculated separately. At this time, the remaining capacity of the Class I battery is the sum of the remaining capacity of the lithium titanate battery and the lithium-ion battery. If the battery category with the first discharge priority is a Class II or Class III battery, the battery that has not completely converted the sub-quality energy or low-quality energy into high-quality energy is an unusable battery and is not included in the calculation of the remaining capacity.

[0147] Determine whether the battery with the highest discharge priority meets the discharge requirements; if E S ≥P ex ×t ex And P max.1 ≥P max If the discharge requirement is met, the battery pack is controlled to discharge to the external device; otherwise, if the discharge requirement is not met, the same judgment method is used to judge the second priority battery, and so on. For example, if the remaining capacity of a battery is 624.896 kWh, and the current state of charge is 80% with a maximum discharge power of 779.9 kW, and the maximum output power of the external device is 100 kW, then E S P max.1 All of the above judgment conditions are met, therefore, the first type of battery should be used for discharge.

[0148] During the battery pack's discharge process, the discharge current I is collected in real time by the battery's sensors. im and discharge voltage V im It also acquires the charging curve of the external device and converts it into standard battery discharge curve data, including the target current value I during the constant current discharge stage. con The target voltage value V during the constant voltage discharge stage con and the cutoff discharge current value Icut ;

[0149] When discharge begins, the system enters a constant current discharge phase, controlling the battery pack's output power to maintain the discharge current at I. con At this time, the discharge power is P in.1 The expression is as follows:

[0150] P in.1 =I con ×V im

[0151] As the discharge proceeds, when the discharge voltage reaches V con At this time, switch to the constant voltage discharge stage, during which the discharge voltage is kept constant at V. con The discharge current naturally decreases as the external device's power gradually reaches saturation, at which point the discharge power is P. in.2 The expression is as follows:

[0152] P in.2 =I im ×V con

[0153] When the discharge current drops to I cut At that time, the discharge is determined to be complete, and the actual discharge time t is obtained. ac Calculate the time deviation rate Δt rate The expression is as follows:

[0154]

[0155] Calculate the total discharge cost C by analyzing the amount of electricity consumed for each type of charge quality during this discharge and the corresponding cost of that charge quality. total ;

[0156] Obtain the effective charge E actually used for discharge ef And the total power E provided by the scheduling battery total Calculate the energy utilization rate η use The expression is as follows:

[0157]

[0158] For example, in the constant current discharge stage, the current is 300A, the voltage is 3.2V, and the power is 960W; in the constant voltage discharge stage, the voltage is 3.0V, the cutoff current is 10A, and the actual discharge time is 1.9 hours. The time deviation rate at this time is... When the cost of high-quality electricity is 0.5 yuan / kWh, the cost of medium-quality electricity is 0.3 yuan / kWh, and the cost of low-quality electricity is 0.1 yuan / kWh, the total cost of discharging is 400 yuan, and the energy utilization rate is...

[0159] These evaluation data are compiled into feedback data, including deviation rate, discharge cost, energy utilization rate, and records of abnormal situations during the discharge process. This feedback is then fed back into the fluctuation identification method so that its parameters can be adjusted based on this feedback data. For example, the feature weights used to identify fluctuations at different time scales can be optimized, and the parameters of the power prediction model can be improved to enhance the accuracy and effectiveness of future discharge optimization methods.

[0160] Furthermore, the specific steps to activate the protection mechanism include:

[0161] In the off-grid state, various devices cannot rely on the external power grid to start up, so they need to have their own independent starting capability. When the device is in the off-grid state and needs to start up, a specific starting battery pack in the energy storage module is activated. These starting battery packs have pre-stored enough energy to power the most critical control equipment and some small auxiliary equipment in the system.

[0162] The wind turbine's starter motor and the solar photovoltaic panel's drive circuit are started sequentially according to a predetermined order. Driven by the wind turbine's starter motor, the wind turbine blades begin to rotate slowly, gradually reaching a speed that can sustain power generation. When the sunlight conditions are met, the solar photovoltaic panel also begins to output DC power. When the electrical energy output by the wind turbine and the solar photovoltaic panel reaches a certain level and is stabilized after being detected by the energy management module, this electrical energy is gradually introduced into the energy storage module to charge other energy storage batteries.

[0163] During startup, the remaining power of the startup battery pack is calculated in real time to ensure it can continuously power critical equipment; let the initial power of the startup battery pack be E. 0.off The total power of the key equipment is P total.off The startup time is t. off Calculate the remaining power E re.off , when E re.off Below the set minimum power threshold E th.off When necessary, appropriate measures should be taken, such as switching to the backup starter battery pack or adjusting the starter strategy; the expression is as follows:

[0164] E re.off =E 0.off -P total.off

[0165] During the startup process of a wind turbine, the time t required for the turbine to reach its self-sustaining power generation speed is calculated based on the turbine's characteristic curve. wind and the power P required to start the motor wind For solar photovoltaic panels, the output power P is calculated based on the light intensity and the conversion efficiency of the photovoltaic panel. pv Determine whether the power requirements for system startup are met;

[0166] The system continuously monitors parameter changes between various power generation devices and the load, employing a combination of active and passive islanding detection methods to determine if islanding occurs. Once islanding is detected, the system adjusts the output power of the power generation modules and controls the charging and discharging status of the energy storage modules to maintain stable system operation. It also promptly issues alarms to notify relevant personnel, improving the safety and stability of the system during off-grid operation, preventing equipment failures and safety accidents caused by islanding, and ensuring that the system can continuously and reliably supply power to the load.

[0167] Example 2

[0168] Please see Figure 4 Another embodiment of the present invention provides a battery pack charging and discharging control device, comprising: a generator set, an energy storage battery, and a charging pile;

[0169] The generator set includes a wind turbine and a photovoltaic panel. The wind turbine has a starting wind speed of <3m / s, and the photovoltaic panel uses monocrystalline silicon PERC modules with a conversion efficiency of ≥22%. The tilt angle is installed according to the local latitude +5°.

[0170] The energy storage battery includes a lithium-ion battery pack, a lithium titanate battery pack, and a bidirectional DC / DC converter. The lithium-ion battery pack is used to balance 24-hour energy fluctuations, while the lithium titanate battery pack is used to buffer 10-second power fluctuations. The two are connected in parallel through the bidirectional DC / DC converter. The lithium-ion battery pack has a capacity of 500kWh and uses CATL lithium iron phosphate cells, while the lithium titanate battery pack has a capacity of 5kWh and uses Maxwell 3000F cells. The bidirectional DC / DC converter has an efficiency of ≥98%.

[0171] The charging pile enables 150kW fast charging of electric vehicles in off-grid environments, with an energy storage utilization rate of ≥95%.

[0172] In summary, this invention collects charging energy quality parameters, calculates voltage deviation rate, frequency deviation rate, and harmonic distortion rate, introduces a comprehensive quality score, sets a quality threshold, and classifies charging quality levels. Based on these levels, three types of batteries are configured: Type I batteries use a combination of lithium-ion and lithium titanate batteries; Types II and III batteries use only lithium-ion batteries; and the quantity of each type of battery is calculated. A fluctuation type is identified using an identification model; for short-term fluctuations, the number of lithium titanate batteries is adjusted; for medium- and long-term fluctuations, the number of lithium-ion batteries is adjusted. Simultaneously, the battery operating status is monitored in real time, SOC and SOH are estimated, an active balancing mechanism is triggered, temperature is controlled, a three-level early warning system is established to ensure stable battery operation, and during discharge, priorities are set and the discharge plan is adjusted in real time.

[0173] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A charge-discharge control method of a battery pack, characterized by, Comprise: Obtain the quality parameter of the electric energy to be charged, set the preliminary charging method, calculate the charging quality, and grade the charging quality through the set quality threshold, and classify the battery pack to be charged according to the grade of the charging quality, and generate the preliminary charging strategy; Among them, according to the category of charging quality, the battery pack to be charged is divided into three categories, namely, one type of battery, two types of battery and three types of battery, the number of each type of battery is calculated, and the specific steps of configuring the battery pack to be charged include: One type of battery employs a combination of lithium-ion batteries and lithium-titanate batteries, and the number of lithium-ion batteries in the one type of battery is calculated and the number of lithium-titanate batteries ; the expression is as follows: ； ； wherein, is the total daily discharge requirement, is the number of cells in the lithium-ion battery cluster, is the lithium-ion battery monobloc capacity, is the predicted energy of high-quality electrical energy, is the lithium titanate-ion battery monobloc capacity, is the power response efficiency, is the number of cells in the lithium titanate-ion battery cluster, is the redundancy factor; adopting lithium ion batteries as the secondary batteries, and calculating the number of the secondary batteries ; the expression is as follows: ； wherein is the predicted amount of secondary quality electric energy, is the loss coefficient; Lithium-ion batteries are used as the third type of battery, and the number of third type of batteries is calculated The expression is as follows: ； wherein is the predicted energy of low quality electrical energy, is the integrated coefficient; Construct an identification model, set a fluctuation identification method, identify the fluctuation of the charging electric energy under different time scales, and set a fluctuation adjustment method, and dynamically adjust the preliminary charging strategy according to the fluctuation type; Real-time acquisition of discharge demand, setting discharge optimization method, once the discharge demand is detected, making discharge plan, and setting battery discharge priority, dynamically adjusting the discharge plan; When identifying the fluctuation of the charging electric energy, the fluctuation identification method comprises: Collect historical fluctuation data, clean the collected historical fluctuation data; Extract short-term fluctuation features, medium-term fluctuation features and long-term fluctuation features from the historical fluctuation data; A short-term fluctuation identification model is constructed by using a regression logic algorithm, and 0 indicates that there is a short-term fluctuation, and 1 indicates that there is no short-term fluctuation; And a long-term fluctuation identification model is constructed by using a regression logic algorithm, and 1 indicates that there is a long-term fluctuation, and 0 indicates that there is no long-term fluctuation; Add an adder after the short-term fluctuation identification model and the long-term fluctuation identification model to generate an identification model; The fluctuation data to be judged is input into the identification model for fluctuation judgment, 0 indicates that there is a short-term fluctuation, 1 indicates that there is a medium-term fluctuation, and 2 indicates that there is a long-term fluctuation; And output the final prediction fluctuation judgment category.

2. The battery pack charging and discharging control method according to claim 1, wherein: When generating the preliminary charging strategy, the preliminary charging method comprises: Real-time acquisition of quality parameters of the electric energy to be charged, including actual voltage, actual frequency of output electric energy, calculation of real-time voltage deviation rate and frequency deviation rate; obtaining the effective value of each harmonic voltage , calculating the harmonic distortion rate, wherein, is the effective value of the fundamental voltage, and ; assigning different weights to the voltage deviation rate, the frequency deviation rate, and the harmonic distortion rate, and calculating a comprehensive quality score .

3. The charge-discharge control method of the battery pack according to claim 2, characterized by, The preliminary charging method further comprises: The sampling interval is set as , collect the charging quality data in a preset time period, and divide the charging quality data into a training set and a test set in chronological order; Obtain the tolerance range of the charging quality of the electric equipment; Comprehensively consider the statistical characteristics of the data itself and the equipment tolerance range, and preliminarily set the threshold values of the voltage deviation rate, the frequency deviation rate and the harmonic distortion rate; calculating a preliminary threshold for the overall quality score, including a first quality threshold , a second quality threshold ; computing a composite quality score for each data point in the test set and adjusting the preliminary threshold using the test set to generate a quality threshold, including a first quality threshold and a second quality threshold ; dividing the comprehensive quality score according to the quality threshold value to determine the category of charging quality; if , the high-quality electric energy is determined; if , the sub-quality electric energy is determined; if , the low-quality electric energy is determined According to the category of charging quality, the battery pack to be charged is divided into three categories, namely, one type of battery, two types of battery and three types of battery, the number of each type of battery is calculated, and the preliminary charging strategy is generated.

4. The battery pack charging and discharging control method according to claim 3, wherein: When dynamically adjusting the preliminary charging strategy, the fluctuation adjustment method comprises: When the prediction fluctuation determination category is a short-term fluctuation, an energy supply and demand difference in a short term is calculated , and an adjustment amount of the lithium titanate ion battery ; When the prediction fluctuation determination category is a medium-term fluctuation, an energy supply and demand difference in a medium term is calculated , and an adjustment amount of the lithium ion battery , , ; When the predicted volatility is classified as long-term volatility, the energy supply and demand difference is calculated. And the adjustment amount of various batteries , , , .

5. The battery pack charging and discharging control method according to claim 4, wherein: When dynamically adjusting the discharge plan, the discharge optimization method comprises: When a discharge demand is detected, the discharge parameters of the battery pack to be charged are read and the expected discharge power is calculated and the expected discharge time ; Calculating the proportion of different charging qualities in future time periods , , ; Comparing , , the highest proportion of the charging quality corresponding to the battery category is set as the first discharge priority, the second highest proportion is set as the second discharge priority, and the rest is the third discharge priority. For the battery category determined as the first discharge priority, calculate the current remaining power of the battery and determine whether the battery of the first discharge priority meets the discharge requirement; If and then it is determined that the discharge requirement is met and a discharge operation is performed, wherein is the maximum discharge power; Otherwise, the discharge demand is not met, the second discharge priority battery is judged, and so on.

6. The charge-discharge control method of the battery pack according to claim 5, characterized by, The discharge optimization method further comprises: In the charging and discharging process, the discharge current and the discharge voltage are collected in real time, and standard discharge curve data is obtained; When the discharging is started, a constant current discharging phase is entered, so as to maintain the discharging current at , and the charging power is ; When the discharge voltage reaches , the discharge voltage is switched to a constant voltage discharge phase, and the discharge voltage is kept constant at , and the charging power is ; When the discharge current drops to , it is determined that the discharge is completed, and the actual discharge time is obtained, the time deviation rate , the total discharge cost and the energy utilization rate are calculated, and the feedback data is generated.

7. The charge-discharge control method of the battery pack according to claim 6, characterized by, Further comprising: During the battery charging and discharging process, the battery pack to be charged is monitored, a battery management method is set, and the stable operation of the battery pack is maintained; The battery management method comprises: Real-time acquisition of management parameters based on sensors equipped for each battery monomer; Establishing a battery electrochemical model through offline experiments and correcting ampere-hour integration by combining temperature correction coefficients; Analyzing the battery internal resistance change rate and capacity retention rate, constructing a capacity attenuation model, and outputting the predicted battery remaining life; Triggering the active balancing mechanism when the battery monomer voltage deviation exceeds the preset voltage; Optimizing the liquid cooling pipe design through CFD simulation and configuring a PID controller to adjust the cooling liquid flow and temperature in real time; Establishing a three-level early warning system to provide fault early warning for the battery.

8. A charge-discharge control device of a battery pack for implementing the charge-discharge control method of the battery pack according to any one of claims 1 to 7, characterized by It includes: A generator set, energy storage batteries, and charging piles; The generator set includes a wind turbine and a photovoltaic panel; The energy storage batteries include lithium ion battery packs, lithium titanate ion battery packs, and bidirectional DC / DC converters.

Images