Intelligent Load Identification and Allocation System and Method for Lithium Battery Energy Storage System

Abstract

The present invention belongs to the technical field of energy storage systems, and discloses an intelligent load identification and allocation system and method for a lithium battery energy storage system. The method includes: selecting quantum dots based on a nature-inspired optimization algorithm; labeling the quantum dots at preset monitoring positions of load devices; collecting and correcting to obtain fluorescence signals; extracting fluorescence intensity features, fluorescence lifetime features, and fluorescence spectral features based on the fluorescence signals; using each feature as a dimension to form a device fingerprint; storing the normal device fingerprints with known device information in a fingerprint database; comparing the device fingerprint with all the device fingerprints stored in the fingerprint database to identify the device information of the load device; comparing the device fingerprint with the normal device fingerprint to calculate the device fingerprint deviation, and starting an adaptive adjustment mechanism for power allocation according to the calculation result. The present invention can comprehensively consider various factors and perform optimized power allocation to extend the service life of lithium batteries.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage systems, and more specifically, to an intelligent load identification and allocation system and method for a lithium battery energy storage system. Background Art

[0002] Chinese Patent Application with Publication No. CN119253823A discloses an independent photovoltaic power generation hybrid energy storage system and a power allocation method with a threshold, and the method includes: S1: calculating the total power P that the hybrid energy storage system should bear H ; S2: decomposing the total power P that the hybrid energy storage system needs to bear H into an average power component P av through a moving average algorithm; S3: using the state of charge SOC of the lithium battery LIB and the state of charge SOC of the supercapacitor sc as the input quantities of the fuzzy controller, and the output quantity of the fuzzy controller is an adjustment coefficient K LIB , multiplying P av in step S2 by the adjustment coefficient K LIB to obtain a power value P L'IB ; S4: passing the power value P L'IB in step S3 through a threshold tuning module to obtain the final power value P LIB that the lithium battery needs to suppress, and subtracting P H from the total power P borne by the hybrid energy storage system in step S1 LIB to obtain a remaining power value P sc_1 borne by the supercapacitor; S5: dividing the lithium battery suppression power value P LIB obtained in step S4 by the lithium battery voltage U LIB to obtain the reference current i LIBref of the lithium battery, and taking the difference between the lithium battery reference current i LIBref and the actual current i LIB to obtain the lithium battery current error Δi LIB ; S6: multiplying the lithium battery current error Δi LIB in step S5 by the lithium battery voltage to obtain the uncompensated power demand P sc_2 of the lithium battery, and this uncompensated amount is borne by the supercapacitor; S7: passing P sc_1 in step S4 and P sc_2 in step S6 through a summation module to obtain the power value P sc that the supercapacitor needs to suppress; S8: After the power allocation is completed, a dynamic evolution control strategy is adopted to control the power flow of the energy storage system. The invention can quickly respond to the power imbalance in the system in a timely manner when the external environment and the load change, and maintain the bus voltage and system stability.

[0003] Although the above method can meet most scenarios, through research and practical application of the above method and the existing technology, it is found that the above method and the existing technology have at least the following partial defects:

[0004] In practical applications, the changes in the load are complex and rapid. When the above method copes with the dynamic changes in the load, due to the lack of real-time load recognition ability, it is impossible to adjust the power distribution strategy in a timely manner according to the changes in the load; and the power distribution strategy of adjusting the energy distribution by limiting the threshold of the charging and discharging power of the lithium battery is relatively single, and it is impossible to perform more optimized power distribution for the load device.

[0005] In view of this, the present invention proposes an intelligent load recognition and distribution system and method for a lithium battery energy storage system to solve the above problems. Summary of the Invention

[0006] In order to overcome the above defects of the existing technology and achieve the above object, the present invention provides the following technical solutions: An intelligent load recognition and distribution method for a lithium battery energy storage system, including the following steps:

[0007] Select quantum dots with the best fluorescence emission wavelength and the best stability based on a nature-inspired optimization algorithm;

[0008] Mark the quantum dots at the preset monitoring positions of the load device; aim the probe of the fluorescence spectrometer at the preset monitoring positions marked with the quantum dots, collect the background light signal when the excitation light source is not started and the actual fluorescence signal when the excitation light source is started, and correct the actual fluorescence signal to obtain the fluorescence signal;

[0009] Extract the fluorescence intensity feature, fluorescence lifetime feature and fluorescence spectrum feature based on the corrected fluorescence signal; splice each feature as a dimension to form a device fingerprint;

[0010] Store the device fingerprints and the corresponding device information in the normal operating state with known device information in the fingerprint database, and the device information includes device type code, rated power, rated operating voltage, rated operating current and device priority code;

[0011] Compare the device fingerprint with all the device fingerprints stored in the fingerprint database to identify the device information of the load device;

[0012] Compare the device fingerprint with the device fingerprint in the normal operating state corresponding to the device information, calculate the device fingerprint deviation, compare the device fingerprint deviation with a preset fingerprint deviation threshold, and when the device fingerprint deviation is greater than the preset fingerprint deviation threshold, start an adaptive adjustment mechanism for power distribution.

[0013] Further, the method for obtaining quantum dots with the best fluorescence emission wavelength and the best stability includes:

[0014] Obtain a preset quantum dot fluorescence emission wavelength range and corresponding quantization index, a preset stability range and corresponding quantization index;

[0015] Obtain quantum dot influence parameters and a preset value range corresponding to each quantum dot influence parameter;

[0016] The number of ants in the preset ant colony is , the pheromone evaporation coefficient is , the pheromone importance factor is and the heuristic information importance factor is ;

[0017] Discretize the value range of each quantum dot influence parameter to obtain corresponding parameter values, and summarize all parameter values to form a parameter space;

[0018] Initialize the pheromone matrix , whose dimension corresponds to the parameter space;

[0019] Each ant starts searching from the initial position in the parameter space. In each iteration, the ant selects the next parameter value based on the pheromone concentration at the current position and the comprehensive heuristic information according to the transition probability;

[0020] Establish an evaluation function for the comprehensive heuristic information according to the fluorescence emission wavelength and stability of the quantum dot; after each ant constructs a quantum dot parameter combination, calculate the corresponding fitness, where the fitness value is equal to the evaluation function value;

[0021] In the entire parameter space, update the pheromone concentration based on the pheromone concentration update formula;

[0022] For the paths passed by ants with fitness higher than the preset fitness threshold, that is, the selected parameter combinations, enhance the pheromone concentration at the corresponding positions.

[0023] When the number of iterations is the preset maximum number of iterations, terminate the algorithm; or the difference between the maximum values of the population fitness in two adjacent iterations is the preset change threshold of the population fitness value;

[0024] After the update is terminated, select the parameter combination with the highest fitness from the parameter combinations searched by all ants and set the corresponding quantum dot as the quantum dot with the best fluorescence emission wavelength and the best stability.

[0025] Further, the methods for obtaining heuristic information include:

[0026] Calculating respectively the heuristic information of fluorescence emission wavelength corresponding to when the fluorescence emission wavelength is within the preset range of quantum dot fluorescence emission wavelengths, or when the fluorescence emission wavelength exceeds the preset range of quantum dot fluorescence emission wavelengths;

[0027] Calculating respectively the heuristic information of stability corresponding to when the stability is within the preset stability range, or when the stability exceeds the preset stability range;

[0028] Calculating the product of the heuristic information of fluorescence emission wavelength and the heuristic information of stability to obtain the comprehensive heuristic information.

[0029] Further, corresponding fluorescence intensity is obtained based on the fluorescence signal, wherein the value of the fluorescence intensity is equal to the value of the fluorescence signal.

[0030] Further, the methods for obtaining fluorescence intensity features include:

[0031] Calculating the average intensity of the fluorescence signal when the device is in a stable operating state; taking the average intensity of the fluorescence signal as the steady-state fluorescence intensity;

[0032] Calculating the difference between the maximum value and the minimum value of the fluorescence intensity; taking the difference between the maximum value and the minimum value of the fluorescence intensity as the fluorescence intensity change range;

[0033] Calculating the change rate of the fluorescence intensity with time as the fluorescence intensity change rate;

[0034] Concatenating the steady-state fluorescence intensity, the fluorescence intensity change range, and the fluorescence intensity change rate to obtain the fluorescence intensity features.

[0035] Further, the methods for obtaining fluorescence lifetime features include:

[0036] Collecting fluorescence intensities at moments to fit the fluorescence decay law curve to obtain a fluorescence decay fitting curve for the fluorescence lifetime, and calculating the average value of the fluorescence lifetimes of

[0037] fluorescence signals to obtain the average fluorescence lifetime ; Obtaining the fitting maximum value of the fluorescence intensity according to the fluorescence decay fitting curve, calculating the half maximum value, i.e., half of the fitting maximum value of the fluorescence intensity; solving for the two time points and the difference between, at two time points and is used as the fluorescence lifetime distribution width;

[0038] The average fluorescence lifetime and the fluorescence lifetime distribution width are spliced to obtain the fluorescence lifetime feature.

[0039] Furthermore, the method for obtaining the fluorescence spectrum feature includes:

[0040] Based on the curve of the fluorescence intensity varying with the fluorescence emission wavelength obtained by fluorescence spectrum measurement, the fluorescence emission wavelength and intensity corresponding to the peak with the maximum intensity in the curve are obtained. The fluorescence emission wavelength corresponding to the peak with the maximum intensity is used as the main peak position, and the intensity corresponding to the peak with the maximum intensity is used as the main peak intensity;

[0041] Calculate the two fluorescence emission wavelengths corresponding to half of the main peak intensity in the fluorescence spectrum, calculate the difference between the two fluorescence emission wavelengths, and use the difference between the two fluorescence emission wavelengths as the spectral bandwidth;

[0042] Use a Gaussian function to describe the spectral curve, and fit the Gaussian function based on the fluorescence emission wavelength and the corresponding fluorescence intensity to obtain the spectral curve function; perform the first derivative on the spectral curve function to obtain the first derivative, and then perform the second derivative to obtain the second derivative, and find the position as the peak in the spectral curve function; calculate the position offset of the same peak in different states of the device; obtain the fluorescence intensity corresponding to each peak, and the two fluorescence emission wavelengths corresponding to half of the fluorescence intensity corresponding to each peak, calculate the difference between the two fluorescence emission wavelengths, and use the difference between the two fluorescence emission wavelengths as the peak width;

[0043] Based on the fluorescence intensity corresponding to the left and right sides of each peak at a distance of the fluorescence emission wavelength interval from the peak position, calculate the shape parameter of each peak;

[0044] Splice the position offsets, fluorescence intensities, widths and shape parameters of all peaks as the spectral shape;

[0045] Splice the main peak position and main peak intensity, spectral bandwidth and spectral shape to obtain the fluorescence spectrum feature.

[0046] Furthermore, start the adaptive adjustment mechanism for power distribution;

[0047] Construct a linear priority evaluation function regarding the fluorescence intensity change range, average fluorescence lifetime, rated power and device priority encoding; according to the number of load devices and the total power of the lithium battery energy storage system, perform power distribution on the th device according to the priority of the device; the power allocated to the th device Under the conditions of meeting the power limit, the power storage system's power quantity constraint, and the device's voltage constraint, optimize the power allocated to the nth device based on the nature-inspired optimization algorithm.

[0048] Furthermore, during the optimization process of the power allocated to the nth device based on the nature-inspired optimization algorithm, construct a corresponding parameter space for optimization based on the power limit, the power storage system's power quantity constraint, and the device's voltage constraint.

[0049] Furthermore, the method for correcting the actual fluorescence signal to obtain the fluorescence signal includes: subtracting the background fluorescence signal from the actual fluorescence signal to obtain the corrected fluorescence signal.

[0050] Furthermore, the method for obtaining the device information of the load device includes:

[0051] Use a distance metric-based method to compare the device fingerprint with all the device fingerprints in the fingerprint database, and calculate the similarity between the device fingerprint and all the device fingerprints stored in the fingerprint database; compare the similarity with a preset similarity threshold, and when the similarity exceeds the preset similarity threshold, obtain the corresponding device information.

[0052] The intelligent load identification and allocation system of the lithium battery energy storage system, implementing the intelligent load identification and allocation method of the lithium battery energy storage system, includes:

[0053] Quantum dot optimization module: Select quantum dots with the best fluorescence emission wavelength and stability based on the nature-inspired optimization algorithm;

[0054] Signal acquisition module: Label the quantum dots at the preset monitoring positions of the load device; align the probe of the fluorescence spectrometer with the preset monitoring positions marked with quantum dots, collect the background light signal when the excitation light source is not started and the actual fluorescence signal when the excitation light source is started, and correct the actual fluorescence signal to obtain the fluorescence signal;

[0055] Feature extraction module: Extract fluorescence intensity features, fluorescence lifetime features, and fluorescence spectrum features based on the corrected fluorescence signal; splice each of the fluorescence intensity features, fluorescence lifetime features, and fluorescence spectrum features as a dimension to form a device fingerprint;

[0056] Fingerprint pre-storage module: Store the device fingerprints and the corresponding device information in the normal operating state of known device information in the fingerprint database, where the device information includes device type code, rated power, rated operating voltage, rated operating current, and device priority code;

[0057] ​​Load identification module: Compare the device fingerprint with all the device fingerprints stored in the fingerprint database to identify the device information of the load device;

[0058] Power distribution module: Compare the device fingerprint with the device fingerprint under the normal operating state corresponding to the device information, calculate the device fingerprint deviation, compare the device fingerprint deviation with a preset fingerprint deviation threshold, and when the device fingerprint deviation is greater than the preset fingerprint deviation threshold, start the adaptive adjustment mechanism for power distribution.

[0059] Technical effects and advantages of the intelligent load identification and distribution system and method for the lithium battery energy storage system of the present invention:

[0060] The present invention collects the fluorescence signal of the load device based on quantum dots with the best fluorescence emission wavelength and the best stability, performs further feature extraction to obtain the device fingerprint, and can accurately identify loads of different types and different power requirements based on the device fingerprint, so as to achieve more refined distribution; it also performs power distribution through the adaptive adjustment mechanism, which can comprehensively consider various factors and perform more optimized power distribution to extend the service life of the lithium battery and improve the overall performance of the system. Brief description of the drawings

[0061] Figure 1 Schematic diagram of the intelligent load identification and distribution method for the lithium battery energy storage system of the present invention;

[0062] Figure 2 Schematic diagram of the quantum dot optimization result of the present invention;

[0063] Figure 3 Block diagram of the intelligent load identification and distribution system for the lithium battery energy storage system of the present invention;

[0064] Figure 4 Schematic diagram of the fingerprint pre-storage module interface of the present invention. Detailed implementation manners

[0065] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0066] Embodiment 1

[0067] Please refer to Figure 1 As shown, the intelligent load identification and distribution method for the lithium battery energy storage system in this embodiment includes:

[0068] Quantum dots with optimal fluorescence emission wavelength and best stability are selected based on a nature-inspired optimization algorithm.

[0069] Methods for obtaining quantum dots with optimal fluorescence emission wavelength and optimal stability include:

[0070] Obtain a preset quantum dot fluorescence emission wavelength range and a corresponding quantitative index, a preset stability range and a corresponding quantitative index.

[0071] Get Quantum dots affect parameters And the preset value range corresponding to each quantum dot influencing parameter; such as working parameters, fluorescence emission performance and stability indicators.

[0072] The number of ants in the default ant colony is , the pheromone volatility coefficient is , the pheromone importance factor is And the heuristic information importance factor is .

[0073] The value range of each quantum dot influencing parameter is discretized to obtain the corresponding parameter value, and all parameter values are summarized to form a parameter space.

[0074] Initialize the pheromone matrix , The dimension of corresponds to the parameter space.

[0075] Each ant starts searching from an initial position in the parameter space, and in each iteration, the ant selects the next parameter value based on the transition probability according to the pheromone concentration at the current position and the comprehensive heuristic information.

[0076] Methods for obtaining heuristic information include:

[0077] Calculate fluorescence emission wavelength heuristic information: When the fluorescence emission wavelength When the wavelength of the quantum dot fluorescence emission is within the preset range, the heuristic information ;in, is a mathematical constant; is the optimal fluorescence emission wavelength; is the fluorescence emission wavelength scale parameter; when the fluorescence emission wavelength Fluorescence emission wavelength heuristic information when the preset quantum dot fluorescence emission wavelength range is exceeded ;in, is the maximum value of the preset quantum dot fluorescence emission wavelength range; It is the minimum value of the preset quantum dot fluorescence emission wavelength range.

[0078] Compute stability heuristic information; when stability When within the preset stability range, the stability heuristic information ; where is the optimal stability; is the stability scale parameter; when the stability exceeds the preset stability range, the stability heuristic information ; where is the maximum value of the preset stability range; is the minimum value of the preset stability range.

[0079] Calculate the comprehensive heuristic information ; where is the fluorescence emission wavelength heuristic information; is the stability heuristic information; in the comprehensive heuristic information , only when the fluorescence emission wavelength and the stability are both close to the optimal values, the comprehensive heuristic information will approach 1, and any deviation of a characteristic from the ideal value will cause to decrease rapidly.

[0080] Obtain the transition probability for the th ant to select the position from the current position as:

[0081]

[0082] where is the pheromone concentration from position to position at the th iteration; is the comprehensive heuristic information from position to position ; is the comprehensive heuristic information from position to position ; is the set of parameter values that the current ant can select.

[0083] Establish an evaluation function based on the fluorescence emission wavelength and stability of the quantum dots; after each ant constructs a quantum dot parameter combination , calculate the corresponding fitness ; where is the quantum dot parameter combination corresponding to the th ant; is the The th quantum dot parameter in the quantum dot parameter combination corresponding to one ant; is the quantum dot parameter combination corresponding to the th iteration and the th ant; is the th iteration and the th quantum dot parameter in the quantum dot parameter combination corresponding to the th ant; The higher the fitness, the closer the corresponding parameter combination is to the desired quantum dot characteristics.

[0084] In the entire parameter space, the pheromone concentration is updated based on the pheromone concentration update formula ; where is the pheromone concentration from position to position at the th iteration; This means that over time, the pheromone concentration will gradually decrease to prevent ants from always relying on early search paths.

[0085] For the paths (i.e., the selected parameter combinations) passed by ants with fitness higher than the preset fitness threshold, enhance the pheromone concentration at the corresponding positions. Let the fitness of the th ant in the th iteration be , and the pheromone enhancement amount be , then , where is a constant; After pheromone enhancement, the enhanced pheromone concentration is .

[0086] When the number of iterations reaches which is the preset maximum number of iterations, terminate the algorithm; or calculate the difference between the maximum values of the population fitness in two adjacent iterations. When , it indicates that the algorithm has converged and terminate the algorithm, where is the preset change threshold of the population fitness value; is the maximum value of the population fitness at the th iteration; is the maximum value of the population fitness at the th iteration.

[0087] Refer to Figure 2After the algorithm terminates, select the parameter combination with the highest fitness from all the parameter combinations searched by the ants, that is, the optimization result corresponding to the figure. Take the quantum dot fluorescence emission wavelength and stability corresponding to the parameter combination with the highest fitness as the best fluorescence emission wavelength and the best stability, and set the parameters of the quantum dot to the parameter combination with the highest fitness to obtain a quantum dot with the best fluorescence emission wavelength and the best stability.

[0088] Label the quantum dots at the preset monitoring positions of the load device by physical adsorption method; align the probe of the fluorescence spectrometer with the preset monitoring positions where the quantum dots are labeled, and use the fluorescence spectrometer to collect the fluorescence signals of the quantum dots during the operation of the device, which can ensure accurate collection of the fluorescence signals. Collect the background light signals when the excitation light source is not started to obtain the background fluorescence intensity sequence ; where is the acquisition time; collect the actual fluorescence signals when the excitation light source is started , subtract the background fluorescence signals from the actual fluorescence signals to obtain the corrected fluorescence signals ; where the fluorescence intensity .

[0089] Extract the fluorescence intensity features, fluorescence lifetime features and fluorescence spectrum features based on the corrected fluorescence signals; splice each feature as a dimension to form a device fingerprint; the resulting device fingerprint makes each load device correspond to a unique fingerprint vector, which can accurately distinguish the corresponding device from other devices; the power distribution corresponding to the load device can be performed according to the device fingerprint, thereby improving the power utilization efficiency of the entire system.

[0090] When the device load increases, it will cause situations such as the increase in the temperature inside the device, the change in the electric or magnetic field, and the change in the chemical reaction rate. These factors will affect the surrounding environment of the quantum dots, and then change their fluorescence intensity features. Under high load conditions, more heat, higher electric field intensity or stronger chemical substance interactions may be generated inside the device. These factors will change the microscopic environment where the quantum dots are located and affect the fluorescence lifetime features of their excited states. When the device load changes, factors such as the electric field, magnetic field, temperature and chemical environment around the quantum dots will change, thus affecting the energy level structure and transition probability of the quantum dots, and further affecting the fluorescence spectrum features. Therefore, extracting the fluorescence intensity features, fluorescence lifetime features and fluorescence spectrum features can comprehensively reflect the dynamic changes in the environment where the quantum dots are located during the operation of the device, providing richer information for device fingerprint recognition.

[0091] The methods for obtaining the fluorescence intensity features include:

[0092] Calculate the average intensity of the fluorescence signals of the device in a stable operating state ; where and are the start time and end time for the stable operation of the fluorescence signal, respectively; is the number of time points corresponding to the fluorescence intensity; the average intensity of the fluorescence signal is used as the steady-state fluorescence intensity.

[0093] Calculate the difference between the maximum value and the minimum value of the fluorescence intensity ; where, and are the maximum value and the minimum value of the fluorescence intensity, respectively; the difference between the maximum value and the minimum value of the fluorescence intensity is used as the fluorescence intensity change range.

[0094] Calculate the change rate of the fluorescence intensity over time ; where, is the fluorescence intensity at time is the time interval; the change rate of the fluorescence intensity over time is used as the fluorescence intensity change rate.

[0095] Concatenate the steady-state fluorescence intensity, the fluorescence intensity change range, and the fluorescence intensity change rate to obtain the fluorescence intensity feature.

[0096] The fluorescence lifetime feature includes the average fluorescence lifetime and the fluorescence lifetime distribution width; the methods for obtaining the fluorescence lifetime feature include:

[0097] Collect fluorescence intensities at time points to fit the fluorescence decay curve to obtain a fluorescence decay fitting curve for the fluorescence lifetime, such as where is the fluorescence lifetime; is the fluorescence signal at the start time of stable operation; calculate the average value of the fluorescence lifetimes of

[0098] fluorescence signals to obtain the average fluorescence lifetime According to the fluorescence decay fitting curve, obtain the fitting maximum value of the fluorescence intensity, calculate the half maximum value , that is, half of the fitting maximum value and of the fluorescence intensity; solve for the two time points and corresponding to when the value of the fluorescence decay fitting curve is equal to the half maximum value; based on the fluorescence decay fitting curve, solve for the difference between the two time points such as and to obtain the difference between the two time points and The difference is used as the fluorescence lifetime distribution width.

[0099] The average fluorescence lifetime and the fluorescence lifetime distribution width are spliced to obtain the fluorescence lifetime characteristics.

[0100] The fluorescence spectrum characteristics include the main peak position and intensity, spectral bandwidth, and spectral shape.

[0101] The methods for obtaining the fluorescence spectrum characteristics include:

[0102] Based on the curve of the fluorescence intensity varying with the fluorescence emission wavelength obtained from the fluorescence spectrum measurement, the fluorescence emission wavelength and intensity corresponding to the peak with the maximum intensity in the curve are obtained. The fluorescence emission wavelength corresponding to the peak with the maximum intensity is used as the main peak position, and the intensity corresponding to the peak with the maximum intensity is used as the main peak intensity.

[0103] Calculate the two fluorescence emission wavelengths corresponding to half of the main peak intensity in the fluorescence spectrum, calculate the difference between the two fluorescence emission wavelengths, and use the difference between the two fluorescence emission wavelengths as the spectral bandwidth.

[0104] Use a Gaussian function to describe the spectral curve, and fit the Gaussian function based on the fluorescence emission wavelength and the corresponding fluorescence intensity to obtain the spectral curve function; take the first derivative of the spectral curve function to obtain the first derivative, then take the second derivative to obtain the second derivative, and find the position as the peak in the spectral curve function; calculate the position offset of the same peak of the device in different states ; where and are the positions of the same peak of the device in different states respectively; obtain the fluorescence intensity corresponding to each peak, and the two fluorescence emission wavelengths corresponding to half of the fluorescence intensity corresponding to each peak, calculate the difference between the two fluorescence emission wavelengths, and use the difference between the two fluorescence emission wavelengths as the width of the peak.

[0105] Calculate the shape parameter of each peak ; where and are the fluorescence intensities corresponding to the right side and the left side of the peak at a fluorescence emission wavelength interval from the position of the peak respectively.

[0106] Splice the position offsets, fluorescence intensities, widths, and shape parameters of all peaks as the spectral shape.

[0107] Splice the main peak position and intensity, spectral bandwidth, and spectral shape to obtain the fluorescence spectrum characteristics.

[0108] Store the device fingerprint and the corresponding device information in the normal operating state of the known device information in the fingerprint database. The device information includes device type code, rated power, rated operating voltage, rated operating current, and device priority code, which facilitates subsequent identification and energy allocation.

[0109] After the load device is accurately identified, add the corresponding device fingerprint and device information to the database. At the same time, audit and update the fingerprint data in the database according to the preset update interval to ensure the accuracy and timeliness of the data. If it is found that some device fingerprints have changed due to device aging, maintenance, etc., update the corresponding records in the database in a timely manner.

[0110] Compare the device fingerprint with all the device fingerprints stored in the fingerprint database to identify the device information of the load device. The methods for obtaining the device information of the load device include:

[0111] Use a distance metric-based method to compare the device fingerprint with all the device fingerprints in the fingerprint database, and calculate the similarity between the device fingerprint and all the device fingerprints stored in the fingerprint database; compare the similarity with the preset similarity threshold, and when the similarity exceeds the preset similarity threshold, obtain the corresponding device information.

[0112] Compare the device fingerprint with the device fingerprint in the normal operating state corresponding to the device information, calculate the device fingerprint deviation, and compare the device fingerprint deviation with the preset fingerprint deviation threshold. When the device fingerprint deviation is greater than the preset fingerprint deviation threshold, start the adaptive adjustment mechanism for power distribution.

[0113] Construct a priority evaluation function: ; ; ; ; where 、 and are the results after transformation of the fluorescence intensity change range, average fluorescence lifetime, and rated power respectively; is the preset reference fluorescence lifetime value; is the adjustment index; is the rated power of the device; is the value encoded according to the device type; is the activation parameter.

[0114] Let the number of load devices be , the total power of the energy storage system be , and the priority of the th device be , then the power allocated to the th device; where is the priority of the th device; when the power allocated to the Under the conditions of meeting the power limit, the power storage system's power constraint, and the device's voltage constraint, optimization is performed based on a nature-inspired optimization algorithm.

[0115] Embodiment 2

[0116] Please refer to Figure 3 as shown. The intelligent load identification and allocation system of the lithium battery energy storage system described in this embodiment includes:

[0117] Quantum dot optimization module: Select quantum dots with the best fluorescence emission wavelength and stability based on a nature-inspired optimization algorithm.

[0118] Signal acquisition module: Label the quantum dots at the preset monitoring positions of the load devices; Align the probe of the fluorescence spectrometer with the preset monitoring positions labeled with quantum dots, collect the background light signal when the excitation light source is not started and the actual fluorescence signal when the excitation light source is started, and correct the actual fluorescence signal to obtain the fluorescence signal.

[0119] Feature extraction module: Extract fluorescence intensity features, fluorescence lifetime features, and fluorescence spectral features based on the corrected fluorescence signal; Concatenate each of the fluorescence intensity features, fluorescence lifetime features, and fluorescence spectral features as a dimension to form a device fingerprint.

[0120] Fingerprint pre-storage module: Store the device fingerprints and the corresponding device information in the normal operating state of the known device information in the fingerprint database. The device information includes device type code, rated power, rated operating voltage, rated operating current, and device priority code; Among them, the device name and device description can also be correspondingly input, and reference can be made to Figure 4 for the convenience of later device differentiation.

[0121] Load identification module: Compare the device fingerprint with all the device fingerprints stored in the fingerprint database to identify the device information of the load device.

[0122] Power allocation module: Compare the device fingerprint with the device fingerprint in the normal operating state corresponding to the device information, calculate the device fingerprint deviation, compare the device fingerprint deviation with the preset fingerprint deviation threshold, and when the device fingerprint deviation is greater than the preset fingerprint deviation threshold, start the adaptive adjustment mechanism for power allocation.

[0123] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all of them should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the said claims.

[0124] Finally, the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent load identification and allocation method for a lithium battery energy storage system, characterized in that, The steps include: Selecting quantum dots with the best fluorescence emission wavelength and the best stability based on a nature-inspired optimization algorithm; Marking quantum dots at a preset monitoring position of a load device; aligning a probe of a fluorescence spectrometer with the preset monitoring position marked with quantum dots, collecting a background light signal when the excitation light source is not started and an actual fluorescence signal when the excitation light source is started, and correcting the actual fluorescence signal to obtain a fluorescence signal; Extract fluorescence intensity features, fluorescence lifetime features, and fluorescence spectrum features based on the corrected fluorescence signal; splice each feature as a dimension to form a device fingerprint; The device fingerprint and corresponding device information of the known device information in the normal operating state are stored in the fingerprint database, and the device information includes the device type code, rated power, rated working voltage, rated working current and device priority code; Compare the device fingerprint with all device fingerprints stored in the fingerprint database to identify and obtain the device information of the payload device: The device fingerprint is compared with all the device fingerprints in the fingerprint database using a distance measurement-based method, and the similarity between the device fingerprint and all the device fingerprints stored in the fingerprint database is calculated; the similarity is compared with a preset similarity threshold, and when the similarity exceeds the preset similarity threshold, the corresponding device information is obtained; Compare the device fingerprint with the device fingerprint in normal operation corresponding to the device information, calculate the device fingerprint deviation, and compare the device fingerprint deviation with the preset fingerprint deviation threshold. When the device fingerprint deviation is greater than the preset fingerprint deviation threshold, start the adaptive adjustment mechanism to allocate power: Construct a linear priority evaluation function regarding the fluorescence intensity change range, average fluorescence lifetime, rated power, and device priority encoding; according to the number of load devices and the total power of the lithium battery energy storage system, allocate power to the th device according to the priority of the device; the power allocated to the th device Under the conditions of meeting the power limit, the power storage system power constraint, and the voltage constraint of the device, perform optimization of the power allocated to the th device based on the nature-inspired optimization algorithm. Among them, the corresponding parameter space is constructed for optimization based on power limitation, energy storage system power constraint and equipment voltage constraint.

2. The intelligent load identification and allocation method for the lithium battery energy storage system according to claim 1, wherein Methods for obtaining quantum dots with optimal fluorescence emission wavelength and optimal stability include: Obtain a preset quantum dot fluorescence emission wavelength range and corresponding quantization index, a preset stability range and corresponding quantization index; obtain quantum dot influence parameters and a preset value range corresponding to each quantum dot influence parameter; The number of ants in the preset ant colony is , the pheromone evaporation coefficient is , the importance factor of pheromone is and the importance factor of heuristic information is ; Discretize the value range of each quantum dot influencing parameter to obtain the corresponding parameter value, and summarize all parameter values to form a parameter space; Initialize the pheromone matrix , whose dimension corresponds to the parameter space; Each ant starts searching from an initial position in the parameter space. In each iteration, the ant selects the next parameter value based on the transition probability according to the pheromone concentration at the current position and the comprehensive heuristic information; According to the fluorescence emission wavelength and stability of quantum dots, an evaluation function for comprehensive heuristic information is established; after each ant constructs a quantum dot parameter combination, the corresponding fitness is calculated; wherein the fitness value is equal to the evaluation function value; In the entire parameter space, the pheromone concentration is updated based on the pheromone concentration update formula; For the paths taken by ants whose fitness is higher than the preset fitness threshold, i.e. the selected parameter combination, the pheromone concentration at the corresponding position is enhanced; When the number of iterations reaches the preset maximum number of iterations, stop the update; or when the difference between the maximum values of the population fitness in two adjacent iterations is less than the preset change threshold of the population fitness value, stop the update, where After the update is terminated, the parameter combination with the highest fitness is selected from all the parameter combinations searched by the ants and the corresponding quantum dots are set as the quantum dots with the best fluorescence emission wavelength and the best stability.

3. The intelligent load identification and allocation method for the lithium battery energy storage system according to claim 2, wherein, Methods for obtaining heuristic information include: Calculate respectively when the fluorescence emission wavelength is within the preset quantum dot fluorescence emission wavelength range, or when the fluorescence emission wavelength exceeds the preset quantum dot fluorescence emission wavelength range, the corresponding fluorescence emission wavelength heuristic information; Calculate respectively when the stability is within the preset stability range, or when the stability exceeds the preset stability range, the corresponding stability heuristic information; The comprehensive heuristic information is obtained by calculating the product of the fluorescence emission wavelength heuristic information and the stability heuristic information.

4. The intelligent load identification and allocation method for the lithium battery energy storage system according to claim 1, wherein A corresponding fluorescence intensity is obtained based on the fluorescence signal, wherein a value of the fluorescence intensity is equal to a value of the fluorescence signal.

5. The intelligent load identification and allocation method for the lithium battery energy storage system according to claim 4, wherein The methods for obtaining fluorescence intensity characteristics include: Calculating the average intensity of the fluorescence signal when the computing device is in a stable operating state; using the average intensity of the fluorescence signal as the steady-state fluorescence intensity; Calculate the difference between the maximum and minimum values of the fluorescence intensity; use the difference between the maximum and minimum values of the fluorescence intensity as the fluorescence intensity change range; Calculating the change rate of the fluorescence intensity over time as the fluorescence intensity change rate; Concatenating the steady-state fluorescence intensity, the fluorescence intensity change range, and the fluorescence intensity change rate to obtain the fluorescence intensity characteristics.

6. The intelligent load identification and allocation method for the lithium battery energy storage system according to claim 4, characterized in that, The methods for obtaining fluorescence lifetime characteristics include: Collection The fluorescence intensity at each moment is used to fit the fluorescence decay curve to obtain a fluorescence decay fitting curve for the fluorescence lifetime, and calculate the average value of the fluorescence lifetimes of the fluorescence signals to obtain the average fluorescence lifetime ; Obtain the fitting maximum value of the fluorescence intensity according to the fluorescence decay fitting curve, calculate the half maximum value, that is, half of the fitting maximum value of the fluorescence intensity; solve the two time points corresponding to the value of the fluorescence decay fitting curve equal to the half maximum value and ; solve the two time points based on the fluorescence decay fitting curve and ; calculate the difference between the two time points and , and take the difference between the two time points as the fluorescence lifetime distribution width Concatenating the average fluorescence lifetime and the fluorescence lifetime distribution width to obtain the fluorescence lifetime characteristics.

7. The intelligent load identification and allocation method for the lithium battery energy storage system according to claim 4, wherein The methods for obtaining fluorescence spectrum characteristics include: Based on the curve of the fluorescence intensity varying with the fluorescence emission wavelength obtained from the fluorescence spectrum measurement, obtaining the fluorescence emission wavelength and intensity corresponding to the peak with the maximum intensity in the curve, using the fluorescence emission wavelength corresponding to the peak with the maximum intensity as the main peak position, and using the intensity corresponding to the peak with the maximum intensity as the main peak intensity; Calculating the two fluorescence emission wavelengths corresponding to half of the main peak intensity in the fluorescence spectrum, calculating the difference between the two fluorescence emission wavelengths, and using the difference between the two fluorescence emission wavelengths as the spectral bandwidth; Describing the spectral curve using a Gaussian function, and fitting the Gaussian function based on the fluorescence emission wavelength and the corresponding fluorescence intensity to obtain the spectral curve function; taking the first derivative of the spectral curve function to obtain the first-order derivative, then taking the second derivative to obtain the second-order derivative, and finding the positions of the peaks in the spectral curve function; calculating the position offset of the same peak in different states of the computing device; obtaining the fluorescence intensity corresponding to each peak, as well as the two fluorescence emission wavelengths corresponding to half of the fluorescence intensity of each peak, calculating the difference between the two fluorescence emission wavelengths, and using the difference between the two fluorescence emission wavelengths as the peak width; Calculate the shape parameter of each peak based on the fluorescence intensity corresponding to the positions on both sides of each peak that are separated from the peak position by the fluorescence emission wavelength interval. When calculating the shape parameter of each peak, use the fluorescence intensity corresponding to the positions on both sides of each peak that are separated from the peak position by the fluorescence emission wavelength interval. Concatenating the position offsets, fluorescence intensities, widths, and shape parameters of all peaks as the spectral shape; Concatenating the main peak position, the main peak intensity, the spectral bandwidth, and the spectral shape to obtain the fluorescence spectrum characteristics.

8. The intelligent load identification and allocation method for the lithium battery energy storage system according to claim 1, characterized in that, The method for correcting the actual fluorescence signal to obtain a fluorescence signal includes: subtracting the background fluorescence signal from the actual fluorescence signal to obtain a corrected fluorescence signal.

9. An intelligent load identification and allocation system for a lithium battery energy storage system, which implements the intelligent load identification and allocation method of the lithium battery energy storage system according to any one of claims 1-8, characterized in that, Including: Quantum dot optimization module: Selecting quantum dots with the best fluorescence emission wavelength and stability based on a nature-inspired optimization algorithm; Signal acquisition module: Labeling the quantum dots at the preset monitoring positions of the load device; aligning the probe of the fluorescence spectrometer with the preset monitoring positions labeled with quantum dots, collecting the background light signal when the excitation light source is not started and the actual fluorescence signal when the excitation light source is started, and correcting the actual fluorescence signal to obtain a fluorescence signal; Feature extraction module: Extracting fluorescence intensity characteristics, fluorescence lifetime characteristics, and fluorescence spectrum characteristics based on the corrected fluorescence signal; concatenating each of the fluorescence intensity characteristics, fluorescence lifetime characteristics, and fluorescence spectrum characteristics as a dimension to form a device fingerprint; Fingerprint pre-storage module: Storing the device fingerprints and the corresponding device information in the normal operating state with known device information into the fingerprint database, where the device information includes device type code, rated power, rated working voltage, rated working current, and device priority code; Load identification module: Compare the device fingerprint with all the device fingerprints stored in the fingerprint database, and identify the device information of the load device: Adopt a distance metric-based method to compare the device fingerprint with all the device fingerprints in the fingerprint database, calculate the similarity between the device fingerprint and all the device fingerprints stored in the fingerprint database; compare the similarity with a preset similarity threshold, and when the similarity exceeds the preset similarity threshold, obtain the corresponding device information; Power distribution module: Compare the device fingerprint with the device fingerprint under the normal operating state corresponding to the device information, calculate the device fingerprint deviation, compare the device fingerprint deviation with a preset fingerprint deviation threshold, and when the device fingerprint deviation is greater than the preset fingerprint deviation threshold, start an adaptive adjustment mechanism for power distribution: Construct a linear priority evaluation function regarding the fluorescence intensity change range, average fluorescence lifetime, rated power, and device priority encoding; according to the number of load devices and the total power of the lithium battery energy storage system, perform power allocation for the th device according to the priority of the device; the power allocated to the th device Under the conditions of meeting the power limit, the power storage system power constraint, and the voltage constraint of the device, perform optimization of the power allocated to the th device based on the nature-inspired optimization algorithm; Among them, a corresponding parameter space is constructed based on power limit, energy storage system power constraint, and device voltage constraint for optimization.

Images