CN118536910A - Retired power battery intelligent storage method and device, intelligent storage system and storage medium - Google Patents

Abstract

The present application relates to a method, device, intelligent storage system and storage medium for intelligent storage of retired power batteries, the method comprising: obtaining a plurality of target indicator parameters corresponding to each target AEVB to be stored; after preprocessing the target indicator parameters to generate standard feature parameters, using an evaluation classification model to process the standard feature parameters to obtain classification label data corresponding to each target AEVB; based on the layout parameter information corresponding to the storage area and a plurality of target AEVBs, using the NSGA‑II algorithm to perform storage allocation planning to generate initial storage allocation information; determining the first storage fitness corresponding to all candidate storage areas in the initial storage allocation information, and performing storage planning iterations on the initial storage allocation information according to the difference between the target storage fitness corresponding to the target reorganization heat classification and the target reorganization heat score value in the classification label data and the first storage fitness, until the difference is no greater than a preset threshold, and generating the corresponding target storage allocation information.

Description

Intelligent storage method, device, intelligent storage system and storage medium for retired power batteries

Technical Field

The present application relates to the technical field of cascade utilization of retired power batteries, and in particular to an intelligent storage method, device, intelligent storage system and storage medium for retired power batteries.

Background Art

Power batteries are widely used in electric vehicles and other fields. As time goes by, their performance will gradually decline, and eventually they will be retired and can no longer meet the needs of vehicle use. Retired power batteries still have a certain residual capacity and can be used in energy storage and other scenarios. In related technologies, the ways of recycling and reusing power batteries include disassembly and recycling and cascade utilization. Cascade utilization is becoming more and more popular in power battery recycling because it can extend the life cycle value of the battery and make full use of resources.

In the related art, during the recycling and storage stage of retired power batteries, the efficiency of the reorganization link is not taken into consideration, which will cause uneven quality of the reorganized products when the products are sent to the reorganization workshop for reorganization. At the same time, because some internal characteristic parameters corresponding to different retired power batteries are not available, the retired power batteries are sorted by obtaining limited measurable internal characteristic parameters, and then the sorted power batteries are reorganized, which will cause the retired power batteries to be unable to be effectively classified during the storage process, and thus the quality of the reorganized batteries cannot be effectively controlled, reducing the recycling efficiency of retired power batteries, and will also cause storage chaos of the power batteries to be reorganized during storage, reduce picking efficiency, and due to ineffective classification, it is difficult to store the retired power batteries in a standardized manner, posing a safety risk.

At present, no effective solution has been proposed for the problem in related technologies that retired power batteries cannot be effectively sorted during the storage process, which reduces the recycling rate of power batteries and reduces the effect of intelligent storage.

Summary of the invention

The embodiments of the present application provide a method, device, intelligent storage system and storage medium for intelligent storage of retired power batteries, so as to at least solve the problems in the related art that retired power batteries cannot be effectively sorted during the storage process, reduce the recycling rate of power batteries and reduce the effect of intelligent storage.

In a first aspect, an embodiment of the present application provides an intelligent storage method for retired power batteries, comprising: obtaining target information of multiple target power storage batteries AEVB to be stored, wherein each of the target information includes multiple target indicator parameters, and the target indicator parameters are used to characterize a characteristic parameter of the corresponding target AEVB, and the characteristic parameters include internal characteristic parameters and external characteristic parameters; after preprocessing the multiple target indicator parameters corresponding to each of the target AEVB to generate corresponding multiple standard characteristic parameters, using a trained evaluation classification model to process the multiple standard characteristic parameters corresponding to each of the target AEVB to obtain classification label data corresponding to each of the target AEVB, wherein the classification label data includes a target recombination heat classification and a target recombination heat score value corresponding to the target AEVB, the target recombination heat classification is used to characterize the priority of the target AEVB used during recombination, and the evaluation classification model is based on the extreme gradient boosting decision tree algorithm XGBoos A machine learning model trained by t is trained to generate classification label data corresponding to the corresponding battery according to the input characteristic parameters of the recombinant battery; based on the layout parameter information corresponding to the preset storage area and the multiple target AEVBs, a multi-objective genetic NSGA-II algorithm is used to perform storage allocation planning to generate corresponding initial storage allocation information, wherein the initial storage allocation information includes the candidate storage area allocated to each target AEVB; the first storage fitness corresponding to all the candidate storage areas in the initial storage allocation information is determined, and according to the difference between the target storage fitness corresponding to the target recombination heat classification and the target recombination heat score value and the first storage fitness, the initial storage allocation information is subjected to non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm until the difference is no greater than a preset threshold, and the corresponding target storage allocation information is generated, wherein the target storage allocation information includes the target storage area allocated to each target AEVB.

In a second aspect, an embodiment of the present application provides an intelligent storage device for retired power batteries, comprising:

An acquisition module, used for obtaining target information of a plurality of target power storage batteries AEVB to be stored, wherein each of the target information includes a plurality of target index parameters, and the target index parameters are used for characterizing a characteristic parameter of the corresponding target AEVB, and the characteristic parameters include internal characteristic parameters and external characteristic parameters;

A prediction module, used for preprocessing the multiple target indicator parameters corresponding to each target AEVB to generate the corresponding multiple standard feature parameters, and then using the trained evaluation classification model to process the multiple standard feature parameters corresponding to each target AEVB to obtain classification label data corresponding to each target AEVB, wherein the classification label data includes the target recombination heat classification and the target recombination heat score corresponding to the target AEVB, the target recombination heat classification is used to characterize the priority of the target AEVB to be used during recombination, and the evaluation classification model is a machine learning model trained based on the extreme gradient boosting decision tree algorithm XGBoost, and is trained to generate classification label data corresponding to the corresponding battery according to the input recombination battery characteristic parameters;

A planning module, which is used to perform storage allocation planning based on the layout parameter information corresponding to the preset storage area and the plurality of target AEVBs, using the multi-objective genetic algorithm NSGA-II, and generate corresponding initial storage allocation information, wherein the initial storage allocation information includes the candidate storage area allocated to each target AEVB;

A processing module is used to determine the first storage fitness corresponding to all the alternative storage areas in the initial storage allocation information, and according to the difference between the target storage fitness determined by the target recombination heat classification and the target recombination heat score value and the first storage fitness, perform non-dominated sorting iteration and genetic evolution iteration based on the NSGA-II on the initial storage allocation information until the difference is no more than a preset threshold, and generate corresponding target storage allocation information, wherein the target storage allocation information includes a target storage area assigned to each of the target AEVBs.

In a third aspect, an embodiment of the present application provides an intelligent warehousing system, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the intelligent warehousing method for retired power batteries described in the first aspect.

In a fourth aspect, an embodiment of the present application provides a storage medium on which a computer program is stored, and when the program is executed by a processor, the intelligent storage method for retired power batteries as described in the first aspect above is implemented.

Compared with the related art, the retired power battery intelligent storage method, device, intelligent storage system and storage medium provided in the embodiment of the present application obtain target information of multiple target power storage batteries AEVB to be stored, wherein each of the target information includes multiple target indicator parameters, and the target indicator parameters are used to characterize a characteristic parameter of the corresponding target AEVB, and the characteristic parameters include internal characteristic parameters and external characteristic parameters; after preprocessing the multiple target indicator parameters corresponding to each of the target AEVB to generate the corresponding multiple standard characteristic parameters, the multiple target indicator parameters corresponding to each of the target AEVB are processed using the trained evaluation classification model. The target AEVB is characterized by standard characteristic parameters, and the classification label data corresponding to each of the target AEVBs is obtained, wherein the classification label data includes the target recombination heat classification and the target recombination heat score corresponding to the target AEVB, and the target recombination heat classification is used to characterize the priority of the target AEVB to be used in recombination. The evaluation classification model is a machine learning model trained based on the extreme gradient boosting decision tree algorithm XGBoost, and is trained to generate the classification label data corresponding to the corresponding battery according to the input recombinant battery characteristic parameters; based on the layout parameter information corresponding to the preset storage area and the plurality of the target AEVBs, the multi-objective genetic NSGA-II is used. The algorithm is used to perform storage allocation planning and generate corresponding initial storage allocation information, wherein the initial storage allocation information includes the candidate storage warehouse areas allocated to each of the target AEVBs; the first storage fitness corresponding to all the candidate storage warehouse areas in the initial storage allocation information is determined, and according to the difference between the target storage fitness corresponding to the target recombination heat classification and the target recombination heat score value and the first storage fitness, the initial storage allocation information is subjected to non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm until the difference is no greater than a preset threshold, and the corresponding target storage allocation information is generated, wherein the target The storage allocation information includes the target storage area allocated to each of the target AEVBs, which solves the problem in the related art that retired power batteries cannot be effectively sorted during the storage process, reduces the recycling rate of power batteries and reduces the effect of intelligent storage. The obtained external characteristic parameter data is combined with some measurable internal characteristic parameters to manage and sort retired power batteries, thereby realizing intelligent power battery storage and management. In the case of missing some internal characteristic data, the consistency index and storage rules of retired power battery reorganization can be determined according to different scenarios, and the batteries in storage can be effectively classified so that the reorganized batteries can be used in energy storage scenarios in time, thereby improving the utilization rate of retired power batteries.

The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent.

BRIEF DESCRIPTION OF THE DRAWINGS

The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

FIG1 is a hardware structure block diagram of a terminal of an intelligent storage method for retired power batteries according to an embodiment of the present application;

FIG2 is a flow chart of a method for intelligent storage of retired power batteries according to an embodiment of the present application;

FIG3 is a structural block diagram of an intelligent storage device for retired power batteries according to an embodiment of the present application.

DETAILED DESCRIPTION

In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the contents disclosed in the present application, some changes such as design, manufacturing or production based on the technical contents disclosed in the present application are only conventional technical means, and should not be understood as insufficient contents disclosed in the present application.

Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation and may represent the singular or plural. The terms "including", "comprising", "having" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include unlisted steps or units, or may also include other steps or units inherent to these processes, methods, products or devices. The "multiple links" involved in this application refer to links greater than or equal to two. "And/or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and/or B" may represent: A exists alone, A and B exist at the same time, and B exists alone. The terms "first", "second", "third" and the like involved in this application are only used to distinguish similar objects and do not represent a specific ordering of objects.

Before describing the specific embodiments of the present application, the related technologies of the present application are described as follows:

Based on the extreme gradient boosting decision tree algorithm (eXtreme Gradient Boosting, referred to as XGBoost) using the gradient boosting framework, it aims to provide an efficient, flexible and portable gradient boosting library. XGBoost belongs to the Boosting type of integrated learning method.

Non-dominated Sorting Genetic Algorithm II (NSGA-Ⅱ) is a genetic algorithm used to solve multi-objective optimization problems. It is an improved version of NSGA and solves the problems of the original NSGA in computational complexity, convergence and parameter setting.

Non-dominated sorting

Perform non-dominated sorting on the individuals in the population and divide them into different levels. The higher the level, the higher the priority. Sort and select those solutions that are not dominated by other solutions at the same time on multiple objectives. If a solution is not surpassed by another solution on all objectives, it is non-dominated (or not inferior). In multi-objective optimization, the condition for solution A to dominate solution B is that solution A is not worse than solution B on all objectives and solution A is better than solution B on at least one objective. Compare each candidate solution with other solutions to find those solutions that are not dominated by other solutions. These solutions constitute the first layer of non-dominated frontier (Pareto front or layer). After removing the first layer of non-dominated solutions that have been found, repeat the above process for the remaining solutions to find the second layer of non-dominated solutions. Then continue this process until all solutions are sorted into a certain layer. For the first layer of non-dominated solutions, compare the two fitness values of each solution to find those that are not dominated by any other solution. These solutions constitute the first layer. Assuming that solutions A, B, and C are not dominated by any other solution, they belong to the first layer of non-dominated solutions. The second layer of non-dominated solutions, remove the population after the first layer, compare again, find the solution that is not dominated by any other solution, and form the second layer. Assume that solution D and solution E belong to the second layer. Subsequent layers: continue the previous step until all solutions are assigned to a certain layer.

Genetic Evolution of Non-Dominated Sorting Genetic Algorithm II

Once the non-dominated sorting is completed, these sorting layers can be used to perform selection, crossover and mutation operations to generate the next generation of populations. The lower the number of layers, the higher their priority, which means that these solutions are better balanced in multi-objective optimization; selection, crossover and mutation: use the selection, crossover and mutation operations in the genetic algorithm to generate a new population to ensure population diversity and accelerate the optimization process. Iterative solution: through multiple iterations, new populations are continuously generated and selected, gradually approaching the optimal solution. Each generation includes - calculation of fitness function, non-dominated sorting and crowding distance calculation, selection, crossover and mutation operations. The final optimization result is a Pareto frontier containing a set of non-inferior solutions.

The causes of the technical problems solved by the intelligent warehousing in the embodiment of the present application and the corresponding technical means and principles are explained as follows:

The reason for the recycling and reuse of power batteries is that retired power batteries can be used in energy storage facilities in the power system as energy storage devices for renewable energy to improve energy efficiency; retired power batteries can also be used to build power frequency regulation and backup power systems to provide auxiliary services for the power grid. By using intelligent storage systems, retired batteries can be tested and reclassified, and batteries that meet the standards can be reorganized, distributed, and then reused, thereby creating new economic value.

In the prior art, since retired power batteries of different brands have special battery management systems (BMS), the battery status is monitored and managed through their respective BMSs, for example, one or more of the following internal characteristic parameters: voltage, temperature, charge and discharge, and battery balancing. Since the BMS of retired power batteries of each brand is different and the corresponding authorization permissions are different, it is not feasible to obtain all battery information of retired power batteries of different brands. For example, for power batteries of brand A, due to permission issues, after the power batteries are retired, their corresponding battery balancing parameters and charge and discharge parameters cannot be collected or obtained, and some internal characteristic parameters of power batteries of brand A are not available. At this time, it costs a huge cost to obtain all the battery information of the complete retired power batteries, which is not advisable for enterprises that grade-utilize and recycle power batteries. Therefore, the problem of inability to effectively sort retired power batteries arises, which is the technical problem that needs to be solved by the embodiments of the present application.

For the technical problem to be solved by the embodiment of the present application, the means adopted by the embodiment of the present application is: by proposing to use the available external characteristic parameter data (that is, a large amount of external characteristic data of power battery consistency indicators obtained according to the retired power battery trading platform data, the battery popularity in the network and the measurable structured data, such as: battery name, battery type, battery supply quantity parameters) combined with some measurable internal characteristic parameters (for some measurable data of the disassembled battery, such as: remaining capacity, voltage, current), the retired power batteries are managed, and the retired power batteries are sorted according to the above external characteristic parameters and internal characteristic parameters. That is, when some internal characteristic parameters of the retired power battery are not available or missing, the external characteristic parameters that can characterize the relevant characteristics of the retired power battery obtained based on the existing data are used to replace some of the unavailable internal characteristic parameters. The combination of the two jointly characterizes the consistency of the power battery to be reassembled, and then the retired power batteries are effectively sorted, so that the reassembled batteries can be used in energy storage scenarios in a timely manner, thereby improving the utilization rate of the retired power batteries.

The specific embodiments of the present application are described below:

The method embodiment provided in this embodiment can be executed in a terminal, a computer or a similar computing device. Taking running on a terminal as an example, FIG1 is a hardware structure block diagram of a terminal of the retired power battery intelligent storage method according to an embodiment of the present application. As shown in FIG1 , the terminal may include one or more (only one is shown in FIG1 ) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data. Optionally, the terminal may also include a transmission device 106 and an input/output device 108 for communication functions. It can be understood by a person of ordinary skill in the art that the structure shown in FIG1 is for illustration only and does not limit the structure of the terminal. For example, the terminal may also include more or fewer components than those shown in FIG1 , or have a configuration different from that shown in FIG1 .

The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the intelligent storage method for retired power batteries in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

This embodiment provides a method for intelligent storage of retired power batteries running on the above-mentioned terminal. FIG2 is a flow chart of the method for intelligent storage of retired power batteries according to an embodiment of the present application. As shown in FIG2, the process includes the following steps:

Step S201, obtaining target information of multiple target power storage batteries AEVB to be stored, wherein each target information includes multiple target index parameters, and the target index parameter is used to characterize a characteristic parameter of the corresponding target AEVB, and the characteristic parameter includes an internal characteristic parameter and an external characteristic parameter.

In this embodiment, in order to sort the target AEVB (corresponding to retired power batteries), it is necessary to use the internal characteristic parameters and external characteristic parameters corresponding to the AEVB. After receiving the target AEVB to be stored, the measurable internal characteristic parameters in the target information can be measured and collected after the power battery is retired, or can be measured and collected in real time after the target AEVB is acquired. In this embodiment, the measurable internal characteristic parameters include voltage, temperature, and remaining capacity, and the internal characteristics are necessary parameters; and for the corresponding external characteristic parameters (for example, battery brand, generation date), it means that after the power battery is retired and determined to be used for reorganization, the more important (for example, the indicator parameters are sorted, and the indicator parameters ranked first) indicator parameters are selected from the preset indicator parameters (related to the external characteristic parameters) and assigned to the corresponding target AEVB. That is, after receiving the target AEVB to be stored, the target indicator parameters related to the external characteristic parameters in the target information of each target AEVB already exist.

Step S202, after preprocessing the various target indicator parameters corresponding to each target AEVB and generating the corresponding various standard feature parameters, the various standard feature parameters corresponding to each target AEVB are processed using the trained evaluation classification model to obtain classification label data corresponding to each target AEVB, wherein the classification label data includes a target recombination heat classification and a target recombination heat score value corresponding to the target AEVB, the target recombination heat classification is used to characterize the priority of the target AEVB when being recombined, and the evaluation classification model is a machine learning model trained based on the extreme gradient boosting decision tree algorithm XGBoost, and is trained to generate classification label data corresponding to the corresponding battery according to the input recombination battery characteristic parameters.

In this embodiment, before inputting a plurality of target indicator parameters into a trained evaluation classification model, the plurality of target indicator parameters need to be preprocessed first, that is, the plurality of target indicator parameters are processed into a data format recognizable by the evaluation classification model, and the evaluation classification model is trained using a sample data set of the corresponding data format; in this embodiment, after the preprocessing conversion generates corresponding standard feature parameters, the plurality of standard feature parameters are input into the evaluation classification model, and the plurality of standard feature parameters corresponding to each target AEVB are sorted, and then the important standard feature parameters are selected based on the sorting to perform AEVB classification and AEVB demand forecasting, wherein AEVB classification refers to dividing the AEVB to be stored into battery classifications for whether they are better reorganized, that is, whether the corresponding AEVB will be used for reorganization first; AEVB demand forecasting refers to evaluating the AEVB to be stored, and determining whether the AEVB is a recombinant battery product required in the future according to the evaluation results (corresponding score values), for example: when the score values of the corresponding two AEVBs are 8.555 and 9.122 respectively, at this time, the probability that the AEVB with a high score value will be used as a recombinant battery product in the future is greater than that of the AEVB with a low score value.

In this embodiment, the evaluation and classification model of machine learning is used to sort the consistency indicators, classify the AEVB, and predict the AEVB demand. Then, the corresponding indicators and corresponding weights are determined at different nodes to obtain the target reorganization heat classification and the target reorganization heat score. The corresponding target reorganization heat classification is used to determine the approximate storage area of the AEVB to be stored (that is, it indicates roughly which area), and the target reorganization heat score is used for the specific storage area of the AEVB to be stored. In this embodiment, the actual space of the warehouse is divided into several independent areas. , these areas can be allocated according to the heat level (corresponding to the consistent reorganization heat classification), for example: high-heat (extremely hot, hot) storage areas are set in the core position of the warehouse or in an area that is easy to access. High-heat storage areas have better environmental control systems (for example: temperature and humidity control) and higher safety standards; medium-heat (warm) storage areas are allocated to secondary central areas. The regional environmental control and safety measures of medium-heat storage areas are medium, which are suitable for storing batteries with moderate access frequency; low-heat (cool, cold) storage areas are allocated to the edge of the warehouse or areas that are less frequently visited. The environmental control and safety measures of low-heat storage areas are relatively low. In this way, the reorganization heat classification and target reorganization heat score can be used to reasonably layout the AEVB battery recycling storage to ensure the safety and efficiency of battery storage.

In this embodiment, the target reorganization heat classification and the target reorganization heat score are the expected targets of the corresponding target AEVB when allocating them in the storage area, that is, the most ideal storage location of the target AEVB to be stored at present. For example, the storage areas with high reorganization heat classification (corresponding heat score range of 8.000-10.000) are storage area 1, storage area 2, storage area 3, storage area 4 and storage area 5, and the corresponding reorganization heat score values are 8.25, 8.5, 8.75, 9.0, 9.5 respectively; and one target AEVB corresponds to If the target recombinant heat classification in the classification label data is high heat and the target recombinant heat score value is 8.5, then the target storage area corresponding to the target AEVB is storage area 2; in the storage allocation planning process, the target recombinant heat classification and the target recombinant heat score value in the classification label data are used to guide the storage area allocated to the target AEVB from the initial randomly allocated storage area to the corresponding target storage area gradually, until it is allocated to the target storage area whose corresponding recombinant heat classification and recombinant heat score value correspond to the target recombinant heat classification and target recombinant heat score value respectively.

Step S203, based on the layout parameter information corresponding to the preset storage area and multiple target AEVBs, a multi-objective genetic NSGA-II algorithm is used to perform storage allocation planning to generate corresponding initial storage allocation information, wherein the initial storage allocation information includes the alternative storage area allocated to each target AEVB.

In this embodiment, after determining the classification label data corresponding to the target AEVB, storage allocation planning is performed according to the layout parameter information of the current warehouse area, that is, the corresponding optimal storage allocation is solved using the NSGA-II algorithm; in the solution process, the population is first randomly initialized, that is, a storage allocation plan is randomly generated, and the individuals in the randomly initialized population (corresponding to the initial storage allocation information) represent a consistent possible AEVB storage area configuration plan, that is, an alternative storage area allocated to each target AEVB, and the alternative storage area is a currently idle one in the preset storage area; it can be understood that the initial storage allocation information is a randomly initialized generated plan, and the corresponding storage fitness is likely to be greatly deviated from the target storage fitness. Therefore, it is necessary to perform non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm on the initial storage allocation information.

Step S204, determine the first storage fitness corresponding to all the alternative storage areas in the initial storage allocation information, and according to the difference between the target storage fitness corresponding to the target recombination heat classification and the target recombination heat score value and the first storage fitness, perform non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm on the initial storage allocation information until the difference is no more than a preset threshold, and generate the corresponding target storage allocation information, wherein the target storage allocation information includes the target storage area assigned to each target AEVB.

In this embodiment, a fitness function is used to measure the quality of the individuals (corresponding to a storage area assigned to a target AEVB) corresponding to the storage allocation information generated during the storage allocation planning process, that is, by calculating the first storage fitness of the corresponding storage area in the storage allocation information solved based on the NSGA-II algorithm, the quality of the storage area allocation scheme solved at the time is measured, and by comparing the first storage fitness corresponding to the storage area assigned to all the target AEVBs at the time with the target storage fitness corresponding to the corresponding target AEVB, the storage allocation information generated at the time is guided to perform non-dominated sorting iterations and genetic evolution operation iterations based on the NSGA-II algorithm until the target storage allocation information is generated; it can be understood that in the process of performing non-dominated sorting iterations and genetic evolution operation iterations, the target recombination heat classification and target recombination heat score value in the classification label data are used to guide the storage area assigned to the corresponding target AEVB to gradually move closer to the corresponding target storage area until the corresponding recombination heat classification and recombination heat score value are respectively assigned to the target storage area corresponding to the target recombination heat classification and target recombination heat score value.

It should be noted that in this embodiment, the steps of non-dominated sorting and corresponding genetic evolution operations are corresponding operations in the relevant technology. Those skilled in the art can understand the steps of non-dominated sorting and corresponding genetic evolution operations, and they do not constitute unclear limitations on the present application.

Through the above steps S201 to S204, target information of multiple target power storage batteries AEVB to be stored is obtained, each target information includes multiple target index parameters, and the target index parameters are used to characterize a characteristic parameter of the corresponding target AEVB, and the characteristic parameters include internal characteristic parameters and external characteristic parameters; after preprocessing the multiple target index parameters corresponding to each target AEVB to generate the corresponding multiple standard characteristic parameters, the multiple standard characteristic parameters corresponding to each target AEVB are processed by using the trained evaluation classification model to obtain classification label data corresponding to each target AEVB, and the classification label data includes the target reorganization heat classification and the target reorganization heat score value corresponding to the target AEVB, and the target reorganization heat classification is used to characterize the priority of the target AEVB to be used during reorganization; based on the layout parameter information corresponding to the preset storage area and the multiple target AEVB, the multi-objective genetic NSGA-II algorithm is used to perform storage allocation planning to generate the corresponding initial storage allocation information, and the initial storage allocation information includes the alternative storage area allocated to each target AEVB; determine the location of all the alternative storage areas in the initial storage allocation information The corresponding first storage fitness is obtained, and according to the difference between the target storage fitness corresponding to the target reorganization heat classification and the target reorganization heat score and the first storage fitness, the initial storage allocation information is iterated by non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm until the difference is no more than the preset threshold, and the corresponding target storage allocation information is generated, and the target storage allocation information includes the target storage area allocated to each target AEVB; the problem that retired power batteries cannot be effectively sorted during the storage process, the recycling rate of power batteries is reduced, and the intelligent storage effect is reduced in the related technology is solved, and the obtained external characteristic parameter data is combined with some measurable internal characteristic parameters to manage and sort retired power batteries, so as to realize intelligent power battery storage and management, and can determine the reorganization consistency index and storage rules of retired power batteries according to different scenarios when some internal characteristic data are missing, effectively classify the incoming batteries, so that the reorganized batteries can be used in the energy storage scenario in time, thereby improving the utilization rate of retired power batteries, and reasonably plan and arrange the storage locations of the batteries in stock according to the storage location optimization rules, so that they meet certain safety and storage efficiency.

In some embodiments, preprocessing is performed on a plurality of target indicator parameters corresponding to each target AEVB to generate a plurality of corresponding standard feature parameters, which is achieved by the following steps:

Step 21 : extracting a plurality of target indicator parameters corresponding to each target AEVB from the target information of each target AEVB, wherein the target indicator parameter includes one of the following: AEVB type, AEVB brand, AEVB type, and AEVB electrical parameter.

In this embodiment, the target indicator parameters corresponding to the external characteristic parameters (for example: AEVB type, AEVB brand, AEVB type) refer to that after the power battery is retired and determined to be used for reorganization, more important indicator parameters are screened out from the preset indicator parameters and assigned to the corresponding target AEVB. When the target AEVB to be stored is received, the target information of each target AEVB already has the target indicator parameters corresponding to the external characteristic parameters; and the target indicator parameters characterizing the internal characteristic parameters can be measured when the power battery is retired as a target AEVB, or can be measured when it is stored in the warehouse.

Step 22, filtering, denoising and cleaning the target indicator parameters corresponding to all target AEVBs to generate standard indicator parameters.

In this embodiment, the target indicator parameters are filtered, that is, deduplication processing is performed to remove repeated items in the target indicator parameters to ensure that each data is unique, and then the format is converted to standardize different units of measurement, for example: the working voltage is uniformly converted into volts (V), and the battery capacity is uniformly converted into kilowatt-hours (kWh); the target indicator parameters are denoised, which refers to detecting abnormal values, detecting and eliminating obviously abnormal data points, for example: the power supply of the target AEVB is obviously too high or too low, and for example: the missing values in the target indicator parameters are uniformly processed by filling the missing values or eliminating data rows with seriously incomplete information; the target indicator parameters are cleaned, including: the target indicator parameters are formatted in a unified manner to ensure that all text fields are normalized, and redundant information is removed to retain core data.

Step 23, normalize the standard index parameters corresponding to all target AEVBs to generate standard feature parameters.

In this embodiment, the standard indicator parameters are normalized so that all standard feature parameters input into the evaluation classification model are at the same level.

Through the above steps, a variety of target indicator parameters corresponding to each target AEVB are extracted from the target information of each target AEVB, wherein the target indicator parameters include one of the following: AEVB type, AEVB brand, AEVB type, AEVB electrical parameters; the target indicator parameters corresponding to all target AEVBs are filtered, denoised and cleaned to generate standard indicator parameters; the standard indicator parameters corresponding to all target AEVBs are normalized to generate standard feature parameters, thereby preprocessing the target indicator parameters in the target information of all target AEVBs, so that the formed standard feature parameters have high quality, high consistency and analyzability, so as to evaluate the classification model for AEVB classification and demand forecasting.

In some of the embodiments, before obtaining target information of a plurality of target power storage batteries AEVB to be stored, the following steps are further implemented:

Step 31 uses a preset data collector to obtain corresponding hypertext markup language HTML page data from the target network platform.

In this embodiment, before collecting data, the target network platform is also identified to identify the network platform for trading retired power batteries; then, different data collectors (for example: Python's BeautifulSoup and Scrapy) are configured for different network platforms, and the data collectors regularly and automatically access the target network platform to obtain dynamic data and extract HTML page data.

Step 32, extracting the transaction data corresponding to the AEVB from the HTML page data, and parsing the first indicator parameter corresponding to the external characteristic parameter from the transaction data.

In this embodiment, the transaction data corresponding to the AEVB is extracted by parsing the HTML page data. The corresponding transaction data includes the AEVB name, brand, AEVB type, supply quantity, application scenario, and power supply. Among them, the AEVB names include automotive batteries, industrial batteries, and energy storage batteries. The brand refers to the corresponding battery supplier; the AEVB types include battery packs (Pack), modules (Module), and cells (Cell); the supply quantity provides batch quantity, total quantity, units such as kWh or number of groups and other data; the application scenario refers to the corresponding usage scenario description, for example: electric vehicles, power tools, energy storage systems; the power supply includes voltage (V) and capacity (Ah or kWh).

Step 33: Use the second indicator parameter and the first indicator parameter corresponding to the collected intrinsic characteristic parameters as target indicator parameters.

In this embodiment, after the first indicator parameter is obtained, the first indicator parameter is used as an indicator parameter library. When an AEVB is retired and used for reorganization, the corresponding first indicator parameter is selected from the parameter library and assigned to the corresponding AEVB, so that the target information of the AEVB has target indicator parameters representing external characteristic parameters. At the same time, some internal characteristic parameters can be obtained through measurement, and then the second indicator parameters corresponding to the measured internal characteristic parameters are associated with the AEVB, so that the target information of the AEVB has complete target indicator parameters. It should be understood that using the second indicator parameter and the first indicator parameter as target indicator parameters refers to a parameter library corresponding to the corresponding target indicator parameters, and the target indicator parameters associated with a specific AEVB need to be generated or obtained by assigning the corresponding first indicator parameter and measuring the second indicator parameter.

By utilizing the preset data collector in the above steps, the corresponding hypertext markup language HTML page data is obtained from the target network platform; the transaction data corresponding to the AEVB is extracted from the HTML page data, and the first indicator parameters corresponding to the external characteristic parameters are parsed from the transaction data; the second indicator parameters and the first indicator parameters corresponding to the collected internal characteristic parameters are used as target indicator parameters, thereby realizing the construction and preset of the target indicator parameters, and then the target indicator parameters of the retired AEVB as the recombinant battery can be quickly determined, and then the target AEVB can be quickly sorted and stored.

In some of the embodiments, based on the layout parameter information corresponding to the preset storage area and multiple target AEVBs, a multi-objective genetic NSGA-II algorithm is used to perform storage allocation planning and generate corresponding initial storage allocation information, which is achieved through the following steps:

Step 41, obtain the coding information corresponding to multiple target AEVBs and the layout parameter information of each storage area, wherein the layout parameter information includes the reorganization picking efficiency parameters, storage safety parameters and storage status parameters corresponding to each storage area, the reorganization picking efficiency parameters are determined based on the location information of the storage area, and the storage safety parameters are determined based on the storage environment and location information corresponding to the storage area.

Step 42, after determining the currently idle storage area according to the storage status parameters, randomly assign codes to the currently idle storage area and each coding information to obtain multiple coding bodies corresponding to multiple target AEVBs, and use the storage area corresponding to each coding body as the corresponding candidate storage area.

Step 43, determine the first storage fitness corresponding to each alternative storage area, and generate initial storage allocation information based on the alternative storage areas and the first storage fitness corresponding to all target AEVBs, wherein the first storage fitness is generated by weighting the reorganization picking efficiency parameters and storage safety parameters corresponding to the corresponding storage areas.

In this embodiment, the degree of superiority of each storage allocation scheme is determined based on the two indicators of reorganization picking efficiency and storage safety parameters, that is, the first storage fitness corresponding to the alternative storage area allocated to each target AEVB is calculated to measure the excellence of the allocation scheme, wherein the reorganization picking efficiency is calculated based on the storage record data, the picking path length and time corresponding to each storage area, and then the corresponding reorganization picking efficiency is calculated; the reorganization safety parameter is based on the storage environment and location of the corresponding storage area. The heat classification and corresponding safety score, for example: the storage area equipped with an environmental control system and safety protection measures has a high corresponding safety score.

Through the above steps 41 to 43, a random generation of a storage allocation plan is achieved, providing data for determining target storage allocation information.

In some of the embodiments, according to the difference between the target storage fitness and the first storage fitness corresponding to the target recombination heat classification and the target recombination heat score value, the initial storage allocation information is subjected to non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm, including the following steps:

Step 51, determine the fitness interval corresponding to the target reorganization heat classification corresponding to each target AEVB, and determine whether the first storage fitness corresponding to each candidate storage area is in the corresponding fitness interval.

Step 52, when it is determined that the first storage fitness corresponding to at least one alternative storage area is not in the fitness interval, the initial storage allocation information is subjected to multiple non-dominated sorting iterations and genetic evolution operation iterations based on the NSGA-II algorithm, until the first storage fitness corresponding to the current storage area allocated to each target AEVB in the generated current storage allocation information is in the corresponding fitness interval, wherein the genetic evolution operation includes a coding body selection operation, a coding body crossover operation and a coding body mutation operation.

Step 53, determine whether the difference between the first storage fitness corresponding to the current storage area and the target storage fitness corresponding to the target reorganization heat score is greater than a preset threshold. When it is determined that the difference between the first storage fitness corresponding to the current storage area and the target storage fitness is not greater than the preset threshold, the current storage area is used as the target storage area corresponding to the target AEVB, and the target storage allocation information is generated.

Step 54, when it is determined that the difference between the first storage fitness corresponding to the current storage area and the target storage fitness corresponding to the target reorganization heat score value is greater than a preset threshold, the current storage allocation information is iterated by non-dominated sorting and genetic evolution operations based on the NSGA-II algorithm.

In some of the embodiments, before the first storage fitness corresponding to at least one current storage area is in the corresponding fitness interval, the following steps are also implemented: repeatedly executing non-dominated sorting iterations and genetic evolution operation iterations based on the NSGA-II algorithm on the storage allocation information generated this time.

In some embodiments, evaluating the training of the classification model includes:

Step 61, obtaining a preset data set, wherein the data set includes target indicator parameters corresponding to a plurality of AEVBs.

Step 62, after preprocessing the target indicator parameters corresponding to each AEVB in the data set to generate a feature parameter set, all standard feature parameters corresponding to each AEVB in the feature parameter set are divided into recombinant battery feature groups, and the recombinant battery feature groups corresponding to each AEVB are annotated with corresponding classification label data to generate a recombinant battery feature parameter set.

Step 63, cutting the recombinant battery characteristic parameter set into a recombinant battery characteristic parameter training set and a recombinant battery characteristic parameter test set according to a preset ratio, and using the recombinant battery characteristic parameter training set to train the initially constructed XGBoost machine learning model until regression fitting is achieved to obtain a trained XGBoost machine learning model.

In this embodiment, the XGBoost machine learning model is initially constructed, the relevant parameters and objective function of the XGBoost model are configured, and all standard feature parameters in the recombinant battery feature parameter training set are used as input, and the corresponding classification label data is used as supervision for training; the optimal parameters are selected by setting parameters such as the number of trees, depth, and learning rate, and performing cross-validation.

Step 64, using the recombinant battery characteristic parameter test set, the trained XGBoost machine learning model is tested and evaluated to train and generate an evaluation classification model.

In this embodiment, when the recombinant battery characteristic parameter test set is used to evaluate the model, all standard characteristic parameters in the recombinant battery characteristic parameter test set are used as input, and the deviation between the classification label data output by the model and the classification label data annotated for the recombinant battery characteristic parameter test set is compared, thereby evaluating the effect of the evaluation classification model; in this embodiment, the evaluation classification model effect is evaluated by using indicators such as accuracy, precision, recall rate, and F1-score.

This embodiment also provides an intelligent storage device for retired power batteries, which is used to implement the above-mentioned embodiments and preferred embodiments, and will not be repeated here. As used below, the terms "module", "unit", "subunit", etc. can be a combination of software and/or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

FIG3 is a structural block diagram of a retired power battery intelligent storage device according to an embodiment of the present application. As shown in FIG3 , the device includes an acquisition module 31, a prediction module 32, a planning module 33 and a processing module 34, wherein:

An acquisition module 31 is used to obtain target information of multiple target power storage batteries AEVB to be stored, wherein each target information includes multiple target index parameters, and the target index parameter is used to characterize a characteristic parameter of the corresponding target AEVB, and the characteristic parameter includes an internal characteristic parameter and an external characteristic parameter;

The prediction module 32 is coupled to the acquisition module 31, and is used to pre-process the multiple target indicator parameters corresponding to each target AEVB to generate the corresponding multiple standard feature parameters, and then use the trained evaluation classification model to process the multiple standard feature parameters corresponding to each target AEVB to obtain the classification label data corresponding to each target AEVB, wherein the classification label data includes the target recombination heat classification and the target recombination heat score value corresponding to the target AEVB, the target recombination heat classification is used to characterize the priority of the target AEVB to be used during recombination, and the evaluation classification model is a machine learning model trained based on the extreme gradient boosting decision tree algorithm XGBoost, and is trained to generate the classification label data corresponding to the corresponding battery according to the input recombination battery characteristic parameters;

The planning module 33 is coupled to the prediction module 32 and is used to perform storage allocation planning based on the layout parameter information corresponding to the preset storage area and multiple target AEVBs using the multi-objective genetic algorithm NSGA-II to generate corresponding initial storage allocation information, wherein the initial storage allocation information includes the candidate storage area allocated to each target AEVB;

The processing module 34 is coupled to the planning module 33, and is used to determine the first storage fitness corresponding to all the alternative storage areas in the initial storage allocation information, and according to the difference between the target storage fitness determined by the target recombination heat classification and the target recombination heat score value and the first storage fitness, perform non-dominated sorting iteration and genetic evolution iteration based on NSGA-II on the initial storage allocation information until the difference is no more than a preset threshold, and generate the corresponding target storage allocation information, wherein the target storage allocation information includes the target storage area assigned to each target AEVB.

The intelligent storage device for retired power batteries according to the embodiment of the present application is adopted.

In some embodiments, the prediction module 32 further includes:

An extraction unit, used to extract a plurality of target indicator parameters corresponding to each target AEVB from the target information of each target AEVB, wherein the target indicator parameter includes one of the following: AEVB type, AEVB brand name, AEVB type, and AEVB electrical parameter;

A processing unit, coupled to the extraction unit, is used to filter, reduce noise and clean the target indicator parameters corresponding to all target AEVBs to generate standard indicator parameters;

The generating unit is coupled to the processing unit and is used for normalizing the standard index parameters corresponding to all target AEVBs to generate standard characteristic parameters.

In some of the embodiments, the intelligent storage device for retired power batteries is also used to obtain corresponding hypertext markup language HTML page data from the target network platform using a preset data collector before obtaining target information of multiple target power batteries AEVB to be stored; extract transaction data corresponding to AEVB from the HTML page data, and parse out first indicator parameters corresponding to external characteristic parameters from the transaction data; and use the second indicator parameters and the first indicator parameters corresponding to the collected internal characteristic parameters as target indicator parameters.

In some embodiments, the planning module 33 further includes:

An acquisition unit is used to acquire the coding information corresponding to the plurality of target AEVBs and the layout parameter information of each storage area, wherein the layout parameter information includes the reorganization picking efficiency parameter, storage safety parameter and storage status parameter corresponding to each storage area, the reorganization picking efficiency parameter is determined according to the location information of the storage area, and the storage safety parameter is determined according to the storage environment and location information corresponding to the storage area;

The encoding unit is coupled to the acquisition unit and is used to randomly assign the encoding to the currently idle storage area and each encoding information after determining the currently idle storage area according to the storage state parameter, so as to obtain multiple encoding bodies corresponding to multiple target AEVBs, and use the storage area corresponding to each encoding body as the corresponding candidate storage area;

The determination unit is coupled to the encoding unit, and is used to determine the first storage fitness corresponding to each alternative storage area, and generate initial storage allocation information based on the alternative storage areas and the first storage fitness corresponding to all target AEVBs, wherein the first storage fitness is generated by weighting the reorganization picking efficiency parameters and storage safety parameters corresponding to the corresponding storage areas.

In some embodiments, the processing module 34 further includes:

A processing unit, used to determine the fitness interval corresponding to the target recombination heat classification corresponding to each target AEVB, and determine whether the first storage fitness corresponding to each candidate storage area is in the corresponding fitness interval;

An iteration unit is coupled to the processing unit and is used to perform multiple non-dominated sorting iterations and genetic evolution operation iterations based on the NSGA-II algorithm on the initial storage allocation information when it is determined that the first storage fitness corresponding to at least one candidate storage area is not within the fitness interval, until the first storage fitness corresponding to the current storage area allocated to each target AEVB in the generated current storage allocation information is within the corresponding fitness interval, wherein the genetic evolution operation includes a coding body selection operation, a coding body crossover operation and a coding body mutation operation;

The judgment unit is coupled to the iteration unit and is used to judge whether the difference between the first storage fitness corresponding to the current storage area and the target storage fitness corresponding to the target reorganization heat score value is greater than a preset threshold. When it is judged that the difference between the first storage fitness corresponding to the current storage area and the target storage fitness is not greater than the preset threshold, the current storage area is used as the target storage area corresponding to the target AEVB, and the target storage allocation information is generated.

In some of the embodiments, before the iteration unit determines that the first storage fitness corresponding to at least one current storage area is not in the corresponding fitness interval, the processing module 34 is also used to repeatedly execute the non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm for the storage allocation information generated this time; when the judgment unit determines that the difference between the first storage fitness corresponding to the current storage area and the target storage fitness corresponding to the target recombination heat score value is greater than a preset threshold, the processing module 34 is also used to perform non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm for the current storage allocation information.

In some of the embodiments, the intelligent storage device for retired power batteries is also used to obtain a preset data set, wherein the data set includes target indicator parameters corresponding to multiple AEVBs; after preprocessing the target indicator parameters corresponding to each AEVB in the data set to generate a feature parameter set, all standard feature parameters corresponding to each AEVB in the feature parameter set are divided into recombinant battery feature groups, and the corresponding classification label data is annotated for the recombinant battery feature group corresponding to each AEVB to generate a recombinant battery feature parameter set; the recombinant battery feature parameter set is cut into a recombinant battery feature parameter training set and a recombinant battery feature parameter test set according to a preset ratio, and the recombinant battery feature parameter training set is used to train the initially constructed XGBoost machine learning model until regression fitting is achieved to obtain a trained XGBoost machine learning model; the trained XGBoost machine learning model is tested and evaluated using the recombinant battery feature parameter test set to train and generate an evaluation classification model.

This embodiment also provides an intelligent warehousing system, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

Optionally, the intelligent warehousing system may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.

Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

S1, obtaining target information of multiple target power storage batteries AEVB to be stored, wherein each target information includes multiple target index parameters, and the target index parameter is used to characterize a characteristic parameter of the corresponding target AEVB, and the characteristic parameter includes an internal characteristic parameter and an external characteristic parameter.

S2, after preprocessing the various target indicator parameters corresponding to each target AEVB and generating the corresponding various standard feature parameters, the various standard feature parameters corresponding to each target AEVB are processed using the trained evaluation classification model to obtain classification label data corresponding to each target AEVB, wherein the classification label data includes a target recombination heat classification and a target recombination heat score value corresponding to the target AEVB, wherein the target recombination heat classification is used to characterize the priority of the target AEVB to be used during recombination, and the evaluation classification model is a machine learning model trained based on the extreme gradient boosting decision tree algorithm XGBoost, and is trained to generate classification label data corresponding to the corresponding battery according to the input recombination battery feature parameters.

S3, based on the layout parameter information corresponding to the preset storage area and multiple target AEVBs, a multi-objective genetic NSGA-II algorithm is used to perform storage allocation planning and generate corresponding initial storage allocation information, wherein the initial storage allocation information includes the alternative storage area allocated to each target AEVB.

S4, determine the first storage fitness corresponding to all the alternative storage areas in the initial storage allocation information, and according to the difference between the target storage fitness corresponding to the target recombination heat classification and the target recombination heat score value and the first storage fitness, perform non-dominated sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm on the initial storage allocation information until the difference is no more than a preset threshold, and generate the corresponding target storage allocation information, wherein the target storage allocation information includes the target storage area assigned to each target AEVB.

It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

In addition, in combination with the retired power battery intelligent storage method in the above embodiment, the embodiment of the present application can provide a storage medium for implementation. The storage medium stores a computer program; when the computer program is executed by a processor, any one of the retired power battery intelligent storage methods in the above embodiment is implemented.

Those skilled in the art should understand that the technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims (10)

1. An intelligent storage method for retired power batteries is characterized by comprising the following steps:

Obtaining target information of a plurality of target power storage batteries AEVB to be stored, wherein each target information comprises a plurality of target index parameters, the target index parameters are used for representing one characteristic parameter of the corresponding target AEVB, and the characteristic parameters comprise an inner characteristic parameter and an outer characteristic parameter;

after preprocessing the multiple target index parameters corresponding to each target AEVB to generate multiple corresponding standard characteristic parameters, processing the multiple standard characteristic parameters corresponding to each target AEVB by using a trained evaluation classification model to obtain classification label data corresponding to each target AEVB, wherein the classification label data comprises target recombination heat classification corresponding to the target AEVB and target recombination heat score value, the target recombination heat classification is used for representing the priority of the target AEVB used during recombination, and the evaluation classification model is a machine learning model trained by an extreme gradient lifting decision tree algorithm XGBoost and is trained to generate classification label data corresponding to a corresponding battery according to the input recombination battery characteristic parameters;

Based on the preset layout parameter information corresponding to the warehouse area and a plurality of target AEVB, carrying out warehouse allocation planning by utilizing a multi-target genetic NSGA-II algorithm to generate corresponding initial warehouse allocation information, wherein the initial warehouse allocation information comprises an alternative warehouse area allocated to each target AEVB;

Determining first storage fitness corresponding to all the alternative storage warehouse areas in the initial storage allocation information, and carrying out non-dominant sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm on the initial storage allocation information according to the target storage fitness corresponding to the target reorganization heat classification and the target reorganization heat scoring value and the difference value of the first storage fitness until the difference value is not larger than a preset threshold value, so as to generate corresponding target storage allocation information, wherein the target storage allocation information comprises target storage warehouse areas allocated for each target AEVB.

2. The method of claim 1, wherein preprocessing the plurality of target indicator parameters corresponding to each target AEVB to generate a corresponding plurality of standard feature parameters, comprising:

Extracting a plurality of target index parameters corresponding to each target AEVB from the target information of each target AEVB, wherein the target index parameters comprise one of the following: AEVB type, AEVB product name, AEVB type, AEVB electrical parameters;

filtering, denoising and cleaning the target index parameters corresponding to all the target AEVB to generate standard index parameters;

and carrying out normalization processing on the standard index parameters corresponding to all the target AEVB to generate the standard characteristic parameters.

3. The method of claim 2, wherein prior to obtaining target information for a plurality of target power storage batteries AEVB to be stocked, the method further comprises:

acquiring corresponding hypertext markup language (HTML) page data from a target network platform by using a preset data acquisition device;

Extracting transaction data corresponding to AEVB from the HTML page data, and analyzing a first index parameter corresponding to the external characteristic parameter from the transaction data;

And taking the second index parameter and the first index parameter corresponding to the acquired internal characteristic parameter as the target index parameters.

4. The method of claim 1, wherein the generating the corresponding initial warehouse allocation information based on the preset warehouse lot area corresponding layout parameter information and the plurality of target AEVB and using a multi-target genetic NSGA-II algorithm for warehouse allocation planning comprises:

Acquiring coding information corresponding to a plurality of target AEVB and layout parameter information of each warehouse area, wherein the layout parameter information comprises a reorganization picking efficiency parameter, a warehouse safety parameter and a warehouse status parameter corresponding to each warehouse area, the reorganization picking efficiency parameter is determined according to the position information of the warehouse area, and the warehouse safety parameter is determined according to a storage environment corresponding to the warehouse area and the position information;

After determining the currently idle warehouse area according to the warehouse status parameters, carrying out random allocation coding on the currently idle warehouse area and each piece of coding information to obtain a plurality of coding bodies corresponding to a plurality of target AEVB, and taking the warehouse area corresponding to each coding body as a corresponding alternative warehouse area;

Determining the first storage fitness corresponding to each alternative storage bin, and generating the initial storage allocation information based on the alternative storage bins and the first storage fitness corresponding to all the target AEVB, wherein the first storage fitness is generated by weighting according to the recombination picking efficiency parameter and the storage safety parameter corresponding to the corresponding storage bin.

5. The method of claim 4, wherein performing non-dominant ranking iterations and genetic evolution operation iterations based on the NSGA-II algorithm on the initial bin allocation information based on a difference between a target bin fitness corresponding to the target reorganization heat classification and the target reorganization heat score value and the first bin fitness comprises:

determining an fitness interval corresponding to the target recombination heat classification corresponding to each target AEVB, and determining whether the first storage fitness corresponding to each alternative storage bin is in the corresponding fitness interval;

Under the condition that at least one first warehouse fitness corresponding to the alternative warehouse area is not in the fitness interval, carrying out non-dominant sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm on the initial warehouse allocation information for a plurality of times until the first warehouse fitness corresponding to the current warehouse area allocated to each target AEVB in the generated current warehouse allocation information is in the corresponding fitness interval, wherein the genetic evolution operation comprises a coding body selection operation, a coding body crossing operation and a coding body variation operation;

Judging whether the difference value between the first storage fitness corresponding to the current storage bin and the target storage fitness corresponding to the target recombination heat scoring value is larger than a preset threshold value, and taking the current storage bin as the target storage bin corresponding to the target AEVB and generating the target storage allocation information under the condition that the difference value between the first storage fitness corresponding to the current storage bin and the target storage fitness is not larger than the preset threshold value.

6. The method of claim 5, wherein before the first bin fitness corresponding to at least one of the current bin regions is not within the corresponding fitness interval, the method further comprises: repeatedly executing non-dominant sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm on the storage allocation information generated at present;

In the case that the difference between the first storage fitness corresponding to the current storage area and the target storage fitness corresponding to the target recombination heat scoring value is greater than a preset threshold, the method further includes: and carrying out non-dominant sorting iteration and genetic evolution operation iteration based on the NSGA-II algorithm on the current warehouse allocation information.

7. The method of claim 1, wherein the training of the assessment classification model comprises:

Acquiring a preset data set, wherein the data set comprises a plurality of target index parameters corresponding to AEVB;

Preprocessing the target index parameters corresponding to each AEVB in the data set to generate a characteristic parameter set, dividing all the standard characteristic parameters corresponding to each AEVB in the characteristic parameter set into a recombined battery characteristic set, and labeling the recombined battery characteristic set with the corresponding classification label data to generate a recombined battery characteristic parameter set;

Cutting the recombined battery characteristic parameter set into a recombined battery characteristic parameter training set and a recombined battery characteristic parameter test set according to a preset proportion, and training an initially constructed XGBoost machine learning model by utilizing the recombined battery characteristic parameter training set until regression fitting is carried out to obtain a trained XGBoost machine learning model;

And testing and evaluating the trained XGBoost machine learning model by using the recombinant battery characteristic parameter testing set so as to train and generate the evaluating and classifying model.

8. An intelligent storage device for retired power batteries, which is characterized by comprising:

the storage system comprises an acquisition module, a storage module and a storage module, wherein the acquisition module is used for acquiring target information of a plurality of target power storage batteries AEVB to be stored, each piece of target information comprises a plurality of target index parameters, the target index parameters are used for representing one characteristic parameter of the corresponding target AEVB, and the characteristic parameters comprise an inner characteristic parameter and an outer characteristic parameter;

The prediction module is configured to pre-process the multiple target index parameters corresponding to each target AEVB to generate multiple corresponding standard feature parameters, and then process the multiple standard feature parameters corresponding to each target AEVB by using a trained evaluation classification model to obtain classification tag data corresponding to each target AEVB, where the classification tag data includes a target recombination heat classification corresponding to the target AEVB and a target recombination heat score value, the target recombination heat classification is used to characterize a priority of the target AEVB used during recombination, and the evaluation classification model is a machine learning model trained based on an extreme gradient lifting decision tree algorithm XGBoost and is trained to generate classification tag data corresponding to a corresponding battery according to the input recombination battery feature parameters;

The planning module is used for carrying out warehouse allocation planning by utilizing a multi-target genetic algorithm NSGA-II based on the layout parameter information corresponding to the preset warehouse areas and a plurality of target AEVB, and generating corresponding initial warehouse allocation information, wherein the initial warehouse allocation information comprises an alternative warehouse area allocated to each target AEVB;

the processing module is configured to determine a first storage fitness corresponding to all the candidate storage warehouse areas in the initial storage allocation information, and perform non-dominant sorting iteration and genetic evolution iteration based on the NSGA-II on the initial storage allocation information according to a difference value between the target storage fitness determined by the target recombination heat classification and the target recombination heat scoring value and the first storage fitness until the difference value is not greater than a preset threshold, so as to generate corresponding target storage allocation information, where the target storage allocation information includes a target storage warehouse area allocated for each target AEVB.

9. A smart warehousing system comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the steps of the retired power battery smart warehousing method of any one of claims 1 to 7.

10. A storage medium having stored thereon a computer program, which when executed by a processor implements the intelligent warehousing method of retired power battery according to any one of claims 1-7.