Cooperative medium management method and device between medium storage devices and medium

Abstract

The invention discloses a collaborative medium management method and device between medium storage devices and a medium, and relates to the technical field of artificial intelligence and financial science and technology, and the method comprises the steps: responding to the medium access operation of each medium storage device in a target area range, and generating medium state description information matched with the medium access operation in real time; when the collaborative medium management condition is met, determining a target time interval matched with the collaborative medium management condition, and obtaining state description information of each target medium generated in the target time interval; and performing iterative processing by adopting a particle optimization algorithm according to the state description information of each target medium to obtain an ideal medium storage amount of each medium storage device in each time period, and determining a medium management strategy for each target medium storage device in the target area range by combining the current system time so as to update the storage state of each medium. And the management efficiency, the optimization pertinence, the accuracy and the collaborative management precision of the medium storage equipment in the target area are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence and financial technology, and particularly relates to a cooperative medium management method between medium storage devices, a device and a medium. BACKGROUND

[0002] With the deep integration of information technology and service industry, various self-service devices based on medium interaction, such as automatic teller machines (ATMs), vending machines and smart delivery cabinets, have been widely penetrated into daily life. The related medium management, such as the storage, replenishment and allocation of cash, goods or articles, has become a key link to ensure service continuity and efficiency.

[0003] Currently, the medium management of the medium storage device mostly adopts a traditional experience-driven static mode, which usually formulates a medium adding and allocation plan by manual operation based on historical data (such as the medium consumption average in the past months) and a fixed period (such as twice a week).

[0004] This traditional management mode cannot respond to the sudden fluctuations of medium demand in real time due to the dependence on the fixed period and historical experience, which easily leads to medium shortage or medium overstock, reduces the service quality, increases the operating cost, and is difficult to achieve overall collaborative management for the network composed of multiple devices, cannot balance the medium supply and demand relationship in the region, and reduces the collaborative management accuracy. In addition, this mode excessively depends on manual operation, reduces the management efficiency, and easily causes deviation of the cooperative medium management due to human error, which reduces the management accuracy. SUMMARY

[0005] The present application provides a cooperative medium management method between medium storage devices, a device and a medium to solve the problems of low collaborative medium management accuracy, low efficiency, low accuracy, low service quality and high operating cost of the medium storage devices.

[0006] According to an aspect of an embodiment of the present application, a cooperative medium management method between medium storage devices is provided, comprising:

[0007] In response to the medium access operation of each medium storage device in the target region range, real-time generation of medium state description information matched with the medium access operation; wherein the medium state description information includes the set medium access operation performed on the set medium storage device at the set time, and the medium storage amount in the set medium storage device after the access operation is performed;

[0008] When the cooperative medium management condition is met, the target time interval matched with the cooperative medium management condition is determined, and each target medium state description information generated in the target time interval is obtained;

[0009] According to the target medium state description information, a particle optimization algorithm is used for iterative processing to obtain ideal medium storage amounts of the medium storage devices in at least one time period;

[0010] According to the current system time and the ideal medium storage amounts of the medium storage devices in the time periods, a medium management strategy for the target medium storage devices in the target region range is determined to update the medium storage states in the target medium storage devices.

[0011] According to another aspect of the embodiment of the present application, a device for cooperative medium management among medium storage devices is provided, which comprises:

[0012] A description generation module is configured to generate, in real time, medium state description information matched with a medium access operation in response to the medium access operation of the target medium storage devices in the target region range, wherein the medium state description information comprises a set medium access operation performed on a set medium storage device at a set time and a medium storage amount in the set medium storage device after the access operation is performed;

[0013] A description acquisition module is configured to determine a target time interval matched with a cooperative medium management condition when the cooperative medium management condition is met, and acquire each target medium state description information generated in the target time interval;

[0014] A particle optimization module is configured to use a particle optimization algorithm to perform iterative processing according to the target medium state description information to obtain ideal medium storage amounts of the medium storage devices in at least one time period;

[0015] A medium update module is configured to determine a medium management strategy for the target medium storage devices in the target region range according to the current system time and the ideal medium storage amounts of the medium storage devices in the time periods to update the medium storage states in the target medium storage devices.

[0016] According to another aspect of the embodiment of the present application, an electronic device is provided, which comprises:

[0017] At least one processor and a memory connected with the at least one processor in communication; wherein the memory stores a computer program which can be executed by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the method for cooperative medium management among medium storage devices according to any embodiment of the present application.

[0018] According to another aspect of the embodiments of the present application, a computer readable storage medium is provided, which stores computer instructions for causing a processor to implement the method of cooperative media management among media storage devices according to any of the embodiments of the present application.

[0019] According to another aspect of the embodiments of the present application, a computer program product is also provided, which comprises a computer program for implementing the steps of the method according to any of the embodiments of the present application when executed by a processor.

[0020] The technical solution of the embodiments of the present application generates media state description information matching the media access operation in real time in response to the media access operation of each media storage device in the target region, realizes dynamic collection of the running state data of each media storage device, provides reliable data support for cooperative management, reduces the deviation of decision-making caused by data lag or inaccuracy, determines the target time interval matching the cooperative media management condition when the cooperative media management condition is met, and obtains each target media state description information generated in the target time interval, uses the particle optimization algorithm to perform iterative processing according to each target media state description information, obtains the ideal media storage amount of each media storage device in each time period, can focus on key scenes and reduce invalid data processing, outputs the cooperative global optimal ideal storage amount of each device in the region based on multi-dimensional data, determines the media management strategy for each target media storage device in the target region range in combination with the current system time, updates the media storage state, can dynamically adjust the deviation in time, ensures that the storage state is always reasonable in the scene of business fluctuation and the like, improves the service quality, reduces the operation cost, improves the management efficiency, optimizes the pertinence, accuracy and cooperative management precision of the target region media storage device, and enhances the adaptability to complex business scenes.

[0021] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0023] Figure 1 is a flowchart of a method of cooperative media management among media storage devices according to an embodiment of the present application;

[0024] Figure 2 is a flow chart of another method for cooperative media management between media storage devices according to the second embodiment of the present application;

[0025] Figure 3 is a flow chart of yet another method for cooperative media management between media storage devices according to the third embodiment of the present application;

[0026] Figure 4 is a schematic diagram of a method for cooperative media management between media storage devices according to the embodiments of the present application;

[0027] Figure 5 is a structural schematic diagram of a device for cooperative media management between media storage devices according to the fourth embodiment of the present application;

[0028] Figure 6 is a structural schematic diagram of an electronic device implementing a method for cooperative media management between media storage devices according to the embodiments of the present application. DETAILED DESCRIPTION

[0029] In order to make the persons skilled in the art better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all the other embodiments obtained by the persons skilled in the art without creative labor should belong to the protection scope of the present application.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily mean a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0031] Embodiment One

[0032] Figure 1A flowchart of a cooperative medium management method between medium storage devices for the first embodiment of the present application, the embodiment can be applicable to the case of cooperative management of media among multiple medium storage devices, and the method can be executed by a cooperative medium management device between medium storage devices, which can be realized in the form of hardware and / or software and generally configured in an electronic device.

[0033] Correspondingly, as shown in Figure 1 The method comprises:

[0034] S110, in response to the medium access operation of each medium storage device in the target area range, real-time generation of medium state description information matched with the medium access operation.

[0035] Among them, the medium state description information includes the set medium access operation performed on the set medium storage device at the set time, and the medium storage amount in the set medium storage device after the access operation is performed.

[0036] In the embodiment of the present application, the medium storage device can be specifically understood as a self-service device for storing various media (such as cash, goods or articles, etc.), for example, a cash deposit and withdrawal machine (ATM). The medium access operation can be specifically understood as the medium depositing or withdrawing behavior performed by the user on the device, such as the depositing or withdrawing operation on the ATM.

[0037] The set time can be specifically understood as the specific time point of performing the medium access operation. The set medium storage device can be specifically understood as the specific device being operated, such as the number of ATM machines. The set medium access operation can be specifically understood as the specific action and related parameters performed by the user on the device, such as withdrawing 2000 yuan or depositing 5000 yuan. The medium state description information can be specifically understood as the data information recording the access operation at the set time and the device state after the operation, that is, it can include the set medium access operation performed on the set medium storage device at the set time, and the remaining medium storage amount in the set medium storage device after the operation is completed.

[0038] Among them, the medium storage amount can be weighed by a weight sensor to weigh the total weight of all media in the medium storage device, combined with the standard weight of a single medium, to convert the current remaining medium storage amount; or the light can be emitted and received by the photoelectric sensor, and the medium thickness can be estimated according to the degree of light shielding of the medium stack, and then converted into the medium storage amount. The medium operation information can be recorded by a camera to record the specific time of the operation, and assist to confirm the operation type, such as depositing or withdrawing media; or the user operation data such as medium access quantity and time can be collected by a touch screen. In addition, the medium access frequency can be calculated based on the collected medium access time.

[0039] Specifically, when a user performs a medium access operation on any medium storage device (such as an ATM) in the target area, the system immediately responds to the operation, generates corresponding medium state description information in real time, thereby recording the process information of the medium access operation, i.e., the set time, the set medium storage device, and the set medium access operation, and the device state after the operation, i.e., the medium storage amount in the device.

[0040] S120, when the cooperative medium management condition is met, determining a target time interval matched with the cooperative medium management condition, and acquiring each target medium state description information generated in the target time interval.

[0041] In the embodiment of the present application, the cooperative medium management condition can be specifically understood as a trigger condition for jointly managing multiple medium storage devices in the target area, for example, multiple devices simultaneously experiencing medium shortage or overstock, a specific business peak period, the system detecting unbalanced medium distribution between devices, or reaching a preset joint management period.

[0042] The target time interval can be specifically understood as a specific time range for cooperative medium management corresponding to the cooperative management condition, for example, if the cooperative management is triggered due to a weekend peak, the target time interval can be set to Saturday to Sunday; if it is triggered due to device imbalance, it can be set to a time range of the past 24 hours or between the last device imbalance time point and the current device imbalance time point; if it is triggered due to reaching a preset joint management period, it can be set to a complete management period to ensure that it covers the medium storage state change data of all medium storage devices in the period.

[0043] Specifically, when the system detects that the cooperative medium management condition is met, such as multiple ATMs having cash storage amounts less than a preset safety threshold during the morning peak, determining that there is a risk of shortage, determining a target time interval matched with the cooperative medium management condition, and acquiring target medium state description information generated by all medium storage devices in the target time interval.

[0044] S130, according to each target medium state description information, performing iterative processing using a particle optimization algorithm to obtain an ideal medium storage amount of each medium storage device in at least one time period.

[0045] In the embodiment of the present application, the ideal medium storage amount can be specifically understood as the optimal storage amount of each device in a specific time period obtained after optimization by the particle swarm optimization algorithm.

[0046] Specifically, the particle optimization algorithm takes the medium storage amount of each medium storage device in at least one time period as the position of the particle and the adjustment range of the storage amount as the speed of the particle, wherein the adjustment range can be set according to the deviation degree of the actual storage amount and the expected value.

[0047] The fitness value of each particle is calculated by the fitness function, the individual optimal position and the global optimal position of the particle are determined and reserved based on the fitness value, and after the particle position is updated by the speed adjustment, the better solution is constantly iteratively approximated until the iteration termination condition is met, such as reaching the maximum iteration number, and the ideal medium storage amount of each medium storage device in at least one time period is output.

[0048] S140, according to the current system time and the ideal medium storage amount of each medium storage device in each time period, determine the medium management strategy for each target medium storage device in the target area range, to update the medium storage state in each target medium storage device.

[0049] In the embodiments of the application, the current system time can be specifically understood as the actual time when the medium management strategy is executed. The medium management strategy can be specifically understood as a specific operation scheme (such as supplementing medium or transferring medium across devices) formulated to make the device storage amount reach the ideal value.

[0050] Specifically, the current system time is obtained, and the time period to which the current system time belongs is determined according to the preset time period division rule, the ideal medium storage amount of each medium storage device corresponding to the time period is obtained, and the real-time medium storage amount of each device is compared. If the real-time medium storage amount is greater than the ideal medium storage amount, a transfer-out strategy can be formulated, such as transferring redundant medium to a device short of medium; if the real-time medium storage amount is less than the ideal medium storage amount, a medium supplement strategy can be formulated, which can first be coordinated and scheduled internally, receive redundant medium (i.e. medium transferred out by devices whose storage amount exceeds the ideal value) scheduled from other devices in the target area, and further judge on this basis. If the real-time medium storage amount is greater than the ideal medium storage amount, more medium can be taken out through external interaction to avoid overstocking of the device, and if the real-time medium storage amount is less than the ideal medium storage amount, the medium can be continuously added through external interaction, such as applying for new medium from an external supply center, to update the medium storage state in each target medium storage device, so that the actual medium storage amount reaches the ideal medium storage amount.

[0051] Further, on the basis of the above-mentioned embodiments, the cooperative medium management method between medium storage devices can further include:

[0052] The obtained running parameters of the medium storage device are input into a pre-constructed fault prediction model to obtain a fault prediction result;

[0053] According to the fault prediction result, a corresponding fault warning is triggered.

[0054] In the embodiments of the present application, the operating parameters can be understood as key technical indicators generated during the operation of the medium storage device, and can specifically include current, voltage and temperature signals of the motor collected by the sensing monitoring device, response speed of the device and component vibration frequency, etc.

[0055] It can be understood that the collection period of the operating parameters can be dynamically adjusted according to the degree of parameter abnormality. The higher the collection frequency is, the more accurate the fault precursor can be captured, when the parameter deviates from the normal range more seriously (such as the motor temperature is closer to the warning threshold).

[0056] The fault prediction model can be understood as an algorithm model that can predict the type, probability and time of future possible faults of the device based on the input operating parameters, which is trained in advance by historical fault data and corresponding operating parameters, such as machine learning or deep learning model based on particle optimization algorithm to optimize model parameters.

[0057] Specifically, the system collects the operating parameters of the device, such as current change of the motor during operation and temperature fluctuation of the core components, through the sensing monitoring device such as current sensor and temperature sensor in the ATM according to the collection period. The operating parameters are input into the pre-constructed fault prediction model to analyze the current parameter characteristics and output the fault prediction result. According to the prediction result, the corresponding fault warning is triggered. For example, if the fault probability is greater than or equal to the preset prompt threshold and less than the preset warning threshold, the system background can record the reminder; if the fault probability is greater than the preset warning threshold, an emergency warning is sent to the maintenance personnel, with information such as fault type, predicted occurrence time and recommended treatment scheme.

[0058] By inputting the operating parameters of the medium storage device into the fault prediction model and triggering the corresponding warning according to the prediction result, the system can analyze the current real state of the device, avoid the lag of traditional periodic inspection, the model trained by historical data can identify the fault precursor from the operating parameters, avoid the subjective error of manual judgment, ensure the accuracy of the prediction result, reduce the impact of sudden equipment downtime on service, ensure the continuity of the medium management process, reduce the maintenance cost, and improve the stability of the device operation and the reliability of the service.

[0059] The technical scheme of the embodiment of the present application realizes dynamic collection of running state data of each medium storage device by generating medium state description information matched with the medium access operation in real time in response to the medium access operation of each medium storage device in the target area range, provides reliable data support for collaborative management, and reduces decision deviation caused by data lag or inaccuracy; when the collaborative medium management condition is met, a target time interval matched with the collaborative medium management condition is determined, and each target medium state description information generated in the target time interval is obtained; the ideal medium storage amount of each medium storage device in each time period is obtained by iteratively processing each target medium state description information using a particle optimization algorithm, which can focus on key scenes and reduce invalid data processing, and outputs the collaborative global optimal ideal storage amount of each device in the region based on multi-dimensional data; in combination with the current system time, a medium management strategy for each target medium storage device in the target area range is determined to update the medium storage state, which can dynamically adjust deviation in a timely manner, ensures that the storage state is always reasonable in scenes such as business fluctuations, improves service quality, reduces operating costs, improves management efficiency, optimizes pertinence, accuracy, and collaborative management precision of the target area medium storage device, and enhances the adaptability to complex business scenes.

[0060] Embodiment two

[0061] Figure 2 The flowchart of another collaborative medium management method between medium storage devices provided by the second embodiment of the present application, which is a refinement of the above-mentioned embodiment of “obtaining the ideal medium storage amount of each medium storage device in at least one time period by iteratively processing each target medium state description information using a particle optimization algorithm”, can specifically include: initializing the motion information of each particle of the particle swarm based on each target medium state description information, and calculating the particle fitness value of each particle based on the motion information of each particle through an adaptive function; wherein the motion information of the particle includes the position information and the speed information of the particle; the position information is used to represent the medium storage amount of each medium storage device in each time period; the speed information is used to represent the adjustment amplitude of the medium storage amount of each medium storage device in each time period; determining the individual optimal position of each particle and the global optimal position of the particle swarm based on the particle fitness value; updating the motion information of each particle based on the motion information, the individual optimal position and the global optimal position of each particle in the particle swarm; returning to perform the operation of calculating the particle fitness value of each particle based on the motion information of each particle through the adaptive function until a preset iteration termination condition is met; obtaining the ideal medium storage amount of each medium storage device in each time interval according to the particle fitness value of each particle at the iteration termination.

[0062] Correspondingly, as shown in Figure 2 , the method comprises:

[0063] S210, in response to the medium access operation of each medium storage device in the target area range, real-time generation of the medium state description information matched with the medium access operation.

[0064] The medium state description information includes a set of medium access operations performed on a set of medium storage devices at a set time, and a set of medium storage amounts in the set of medium storage devices after the access operation is performed.

[0065] S220, when the cooperative medium management condition is met, determining a target time interval matched with the cooperative medium management condition, and obtaining each target medium state description information generated in the target time interval.

[0066] S230, initializing the motion information of each particle of the particle swarm based on each target medium state description information, and calculating the particle fitness value of each particle based on the motion information of each particle through the fitness function.

[0067] The motion information of the particle includes: position information and speed information of the particle; the position information is used to represent the medium storage amount of each medium storage device in each time period; and the speed information is used to represent the adjustment amplitude of the medium storage amount of each medium storage device in each time period.

[0068] In the embodiment of the application, the particle swarm can be understood as: an algorithm model for simulating group cooperation optimization, composed of a plurality of particles, each particle representing a potential optimization scheme, such as the medium storage scheme of each medium storage device in each time period. The motion information of the particle can be understood as: a parameter used to describe the state of the particle in the particle swarm algorithm, which can specifically include: position information and speed information.

[0069] The position information can be understood as: the coordinates of the particle in the multi-dimensional space, wherein the position information of each dimension corresponds to the medium storage amount of a certain medium storage device in a certain time period, such as dimension 1 for the medium storage amount of a certain device in the morning peak, and dimension 2 for the medium storage amount of a certain device in the flat peak.

[0070] It can be understood that the division of the position information dimension is directly related to the division of the time period: if the overall time is divided into N time periods according to the time period division rule (such as by hour, by peak / flat / valley, or by weekday / weekend, etc.), the position information of the particle needs to be set with N dimensions, each dimension uniquely corresponding to a time period, for representing the medium storage amount of the medium storage device in the time period.

[0071] The speed information can be understood as: the moving amplitude of the particle in each dimension, corresponding to the adjustment amplitude of the storage amount of a certain medium storage device in a certain time period, such as dimension 1 for the adjustment amplitude of the medium storage amount of a certain device in the morning peak, and dimension 2 for the adjustment amplitude of the medium storage amount of a certain device in the flat peak.

[0072] It can be understood that the dimension division of particle velocity information is directly related to the division of time periods. The value of the velocity can be positive or negative, and the positive value represents adding media to the medium storage device in the corresponding time period, and the negative value represents reducing media from the medium storage device in the corresponding time period.

[0073] The fitness function can be specifically understood as a function for evaluating the pros and cons of particles (i.e. medium storage schemes). The weighted sum of the cost, shortage rate or overstock rate of the storage scheme and other indicators can be calculated as the fitness value.

[0074] Specifically, all particles in the particle group are initialized based on the target medium state description information, the number of particles in the particle group is determined, and initial motion information (position and velocity) is assigned to each particle. The number of particles can be set to a predetermined number, such as 50 particles, representing 50 potential storage schemes.

[0075] Wherein, when initializing the position information, the actual storage amount range of each device in different time periods in the target medium state description information data can be referred to, such as the collected target medium state description information shows that the storage amount fluctuation of a certain ATM device in the morning peak is 50,000-120,000 yuan, the initial position of the particle in the corresponding dimension of the morning peak can be set in the range, such as the initial position of a certain ATM device in the morning peak is set to 80,000 yuan, to ensure that the initial scheme conforms to the actual scene.

[0076] When initializing the velocity information, the initial adjustment amplitude can be set for each dimension according to the adjustment rule of the data storage amount in the target medium state description information, such as the average adjustment of a certain ATM device in the morning peak is ±30,000 yuan, the initial velocity of the device in the morning peak can be set to +20,000 yuan, that is, the adjustment plan is to increase 20,000 yuan each time.

[0077] Then, the motion information of each particle is evaluated by the fitness function, the basic cost of the scheme is calculated based on the position information, such as the overstock cost or the shortage cost, the adjustment cost is calculated combined with the velocity information, and the particle fitness value is output by comprehensively considering the two types of costs.

[0078] Further, on the basis of each of the above embodiments, before initializing the motion information of each particle in the particle group based on each target medium state description information, the method can further include:

[0079] Pretreating the target medium state description information.

[0080] Specifically, the preprocessing of the target medium state description information can include data cleaning and normalization operations. Through data cleaning, noise and outliers in the original data are removed, such as error values caused by accidental fluctuations of sensors, and abnormal storage records caused by device failures, and only real and valid data are retained.

[0081] It can be understood that, taking the ATM cash access transaction scenario as an example, the cash storage range of a large ATM can be 100,000-200,000 yuan, and the cash storage range of a small ATM can be 30,000-80,000 yuan; the access amount of the same device in the morning peak period can be concentrated in 1000-5000 yuan, and the access amount in the low valley period can be concentrated in 100-1000 yuan. If these data are directly used for particle initialization or fitness calculation, the algorithm may overemphasize large devices or peak periods and ignore small devices or low valley periods due to the larger absolute value of the ten-thousand-yuan-level data.

[0082] Through normalization, data of different magnitudes and different units are converted to a unified range (such as the 0-1 interval), eliminating the interference caused by the difference in data scales.

[0083] Through preprocessing, the target medium state description information only retains real and valid data, and the medium storage or medium operation of different devices and different time periods has the same comparison weight in the algorithm, ensuring that the motion information (position and speed) of the particles can consider various scenarios and avoid optimization bias to a single dimension.

[0084] Optionally, on the basis of each of the above embodiments, the fitness function calculates the particle fitness value of each particle based on the motion information of each particle, including:

[0085] Based on the preset medium storage device management rules and the motion information of each particle, the corresponding medium addition cost, medium shortage rate, and medium overstock rate are calculated;

[0086] Based on the medium addition cost, medium shortage rate, and medium overstock rate of each particle, the particle fitness value of each particle is calculated.

[0087] In the embodiments of the present application, the medium addition cost can be specifically understood as the total cost of medium addition generated for maintaining the corresponding storage scheme of particles, which can include fixed cost (such as the basic cost of manual or transportation for each medium addition) and variable cost (unit cost related to the amount of medium addition). The medium shortage rate can be specifically understood as the proportion reflecting that the medium storage amount cannot meet the demand. The medium overstock rate can be specifically understood as the proportion reflecting that the medium storage amount exceeds the demand. The medium storage device management rule can be specifically understood as a series of preset specifications, restrictions or operation criteria that need to be followed when managing the medium storage device. For example, the fixed cost and variable cost of single medium addition, the maximum shortage rate threshold and overstock rate threshold allowed, the maximum medium addition amount of single time cannot exceed the remaining capacity of the device, the unit of single medium addition (for example, for an ATM, the banknote adding unit is 100,000 yuan), and the time interval of two medium additions cannot be shorter than the preset time interval and other operation restrictions.

[0088] Specifically, based on the preset medium storage device management rule and the motion information of each particle, the medium addition cost = fixed cost of each medium addition + unit medium addition cost x medium addition amount (the medium addition amount is also the speed of the particle, which represents the addition of medium when the speed is positive, and represents the removal of medium when the speed is negative, and the corresponding medium removal cost). All time periods are traversed, and the medium addition cost of each time period is accumulated to obtain the total medium addition cost (C).

[0089] Based on the historical medium storage data of each medium storage device in each time period, the average value or 95% confidence interval value is determined to determine the actual demand value of each time period, and the adjusted medium storage amount is calculated according to the adjusted medium storage amount = particle position (original storage amount) + speed (adjustment range).

[0090] If the adjusted storage amount < demand value, the unit shortage amount = demand value - adjusted storage amount can be calculated. Correspondingly, the total shortage amount is the sum of the product results of the unit shortage amount of each period and the time length of the period. The total demand reference is the sum of the product results of the demand value of each period and the time length of the period. According to the shortage rate = (total shortage amount / total demand reference) x 100%, the medium shortage rate (S) is calculated. For example, after the adjustment of a certain particle, the storage amount of 40,000 (demand of 50,000) in the morning peak (3 hours), the shortage amount of this period is 10,000 x 3 hours = 30,000 hours. The storage amount of 30,000 (demand of 20,000) in the flat peak (4 hours) has no shortage. The storage amount of 50,000 (demand of 50,000) in the evening peak (3 hours) has no shortage. The total demand reference = 380,000 hours, and S = (3 / 38) x 100% = 7.89%.

[0091] If the adjusted storage quantity is greater than the demand value, the unit backlog quantity can be calculated as adjusted storage quantity-demand value. Correspondingly, the total backlog quantity is the sum of the product of the unit backlog quantity of each time period and the length of the time period. According to the backlog rate=(total backlog quantity / total demand benchmark) x 100%, the medium backlog rate (B) is calculated.

[0092] The fitness function integrates the medium addition cost, the medium shortage rate and the medium backlog rate into a quantitative value by weighted summation, that is, according to F=w1xC+w2XS+w3xB, wherein C is the medium addition cost, S is the medium shortage rate, B is the medium backlog rate, w1, w2, w3 are preset weight coefficients. F is the fitness value corresponding to the particle. The lower the fitness value F is, the smaller the comprehensive index of the cost, the shortage risk and the backlog risk of the scheme is, that is, the particle (storage scheme) is better.

[0093] Based on the preset medium storage device management rules and the motion information of each particle, the calculation of the medium addition cost, the shortage rate and the backlog rate is based on the actual operation specification and the actual medium storage state, which ensures that the fitness value accurately maps the real management scene, lays a foundation for subsequent iterations to obtain executable storage schemes, and avoids one-sidedness caused by single index judgment by comprehensively evaluating the cost, the shortage rate and the backlog rate three indexes. The weighted calculation method improves the adaptability of the optimization process to the scene, and can respond to the needs of different scenes by adjusting the weight coefficients, so that the optimization result matches the actual management target.

[0094] Optionally, based on the motion information of each particle of the particle group, the initial position and the initial velocity of each particle are generated under the compliance condition according to the medium range of the medium storage device in each time period.

[0095] For each time period, the historical statistical value of the historical medium consumption is calculated based on the target medium state description information, and the medium range of the medium storage device in each time period is determined according to the preset medium storage device management rules.

[0096] A preset number of particles are generated, and the initial position and the initial velocity of each particle are generated under the compliance condition according to the medium range of the medium storage device in each time period.

[0097] In the embodiments of the present application, the historical statistical value can be understood as the medium consumption characteristic value calculated based on the target medium state description information, and can specifically include statistical values such as the minimum value (the least consumption in a time period), the maximum value (the most consumption in a time period) and the average value (the average consumption in a time period). The medium range can be understood as the medium storage quantity interval allowed by the medium storage device in each time period, which is jointly constrained by the historical statistical value and the management rules, such as not lower than the safety stock and not more than the maximum capacity of the device.

[0098] The initial motion information of the particles can be understood as that the initial position (medium storage amount of each period) and initial velocity (adjustment range of each period) of each particle are assigned when the particle group is initialized, and the medium range and compliance conditions are met. The compliance conditions can be understood as a series of constraint rules that the initial position (medium storage amount of each period) and initial velocity (adjustment range) of the particles must meet, for example, taking the cash management of an ATM as an example, when the medium range of the morning peak is 30-120 million, if the initial position of a particle is 10 million, the velocity (adjustment range) must be less than or equal to +2 million and greater than or equal to -7 million to meet the medium range of the morning peak.

[0099] Specifically, for each divided time period, the historical statistical value of medium consumption is calculated based on the target medium state description information, such as the minimum value, the maximum value, and the average value. Correspondingly, the demand fluctuation range, i.e., the medium range, can be determined according to the minimum value, the maximum value, the maximum capacity of the medium storage device, and the safety stock specified in the device management rules.

[0100] Correspondingly, the average value can provide implicit constraints for the generation of subsequent initial positions, i.e., when generating the position randomly, a higher probability density is set near the average value, so that the initial position is more likely to be distributed near the historical average demand value, avoiding the particle deviating from the reasonable range at the beginning.

[0101] The particles are generated according to the preset number (such as 50), and each particle is assigned an initial position and an initial velocity. The position of each particle is a vector with a dimension consistent with the number of time periods (such as a 4-dimensional vector corresponding to 4 time periods), and the numerical value of each dimension is randomly generated within the medium range of the period and is implicitly constrained by the average value (more likely to be close to the average value). For example, taking the cash management of an ATM as an example, the initial position of a particle in the morning peak (dimension 1) is randomly taken as 50,000 yuan (close to the average value of 60,000 yuan) within the medium range of [30,000-120,000], 40,000 yuan is taken in the flat peak (dimension 2) within the medium range of [20,000-80,000], and so on. Finally, the position vector [50,000, 40,000, 80,000, 30,000] (corresponding to the morning peak, flat peak, evening peak, and low valley) is formed.

[0102] The speed of each particle is a vector with the same dimension as the position, and the numerical range of each dimension can be set to a certain percentage (such as 10%-30%) of the medium range of the corresponding time period, which represents the maximum adjustment range (which can be positive or negative, positive for adding medium and negative for reducing medium) of the storage amount of the medium in that period. For example, taking the cash management of an ATM as an example, the medium range during the morning peak is [30,000, 120,000], and the speed range is set to [-30,000, 30,000]; the medium range during the flat peak is [20,000, 80,000], and the speed range is set to [-20,000, 20,000]. By randomly selecting a value within the corresponding speed range, the initial speed vector of the particle can be set to [+20,000, -10,000, +10,000, 0] (indicating that the morning peak plan adds 20,000, the flat peak plan reduces 10,000, etc., the evening peak plan adds 10,000, and the low valley does not add or remove operations).

[0103] By using the statistical value of historical medium consumption as the basis, and combining the preset medium storage device management rules to determine the medium range of each period, the rationality of the initial search boundary is ensured, which conforms to the historical demand law, and also considers the device capacity and operation specifications and other practical constraints, avoiding particles from performing invalid searches that are out of the real situation. The initial position and speed are generated under the compliance condition, ensuring that all particle initial solutions are feasible solutions, eliminating states that do not conform to device capacity and operation specifications, avoiding wasting computing power to optimize invalid solutions, improving search efficiency, and the implicit constraint of the average value in the historical statistical value makes the initial position more distributed around the average value, reducing the blindness of the initial search, focusing the particle swarm on the core range that is more likely to contain the optimal solution from the beginning, accelerating the algorithm convergence, generating a preset number of particles, forming a variety of position and speed combinations, avoiding being limited to local optimization due to single initial solution, and ensuring the diversity of initial solutions, improving the probability of finding the global optimal solution.

[0104] Optionally, based on the above embodiments, the initial position and initial speed of each particle are generated under the compliance condition according to the medium range of the medium storage device in each time period, including:

[0105] Generating the initial position and initial speed of each particle according to the medium range of the medium storage device in each time period;

[0106] Judging the compliance of the initial position and initial speed of each particle, and updating the initial speed of the target particle when the position of the target particle after updating according to the initial speed exceeds the medium range.

[0107] Specifically, the initial position and initial velocity of each particle are generated according to the medium range of the medium storage device in each time period, and then the initial position and velocity are judged for compliance, that is, whether the position updated according to the initial velocity exceeds the medium range. For example, taking the cash management of an ATM as an example, if the medium range during the morning peak is 30-120 thousand yuan, the initial position during the morning peak is 110 thousand yuan, and the initial velocity is +20 thousand yuan, then the updated position is 130 thousand yuan, which exceeds the upper limit of the medium range 120 thousand yuan, and at this time the velocity does not meet the compliance condition; for example, the medium range during the flat peak is 20-80 thousand yuan, the initial position during the flat peak is 20 thousand yuan, and the initial velocity is -10 thousand yuan, and the updated position is 1 thousand yuan, which is lower than the lower limit 2 thousand yuan, and does not meet the compliance.

[0108] When the position updated according to the initial velocity of the target particle exceeds the medium range, it is determined that it does not meet the compliance, and the initial velocity of the target particle can be adjusted so that the updated position is at the boundary of the medium range, that is, it meets the medium range. For example, the initial position during the morning peak is 110 thousand yuan, and the original velocity +20 thousand yuan leads to exceeding the upper limit, and the velocity can be adjusted to +10 thousand yuan (the updated position is 120 thousand yuan, which meets the upper limit); the initial position during the flat peak is 2 thousand yuan, and the original velocity -10 thousand yuan leads to being lower than the lower limit, and the velocity can be adjusted to 0 (the updated position is still 2 thousand yuan, which meets the lower limit).

[0109] By judging the compliance of the updated position and correcting the non-compliant velocity, the dynamic feasibility of the particle state is strengthened, the situation that the initial state of the particle is compliant but the adjusted state is non-compliant is avoided, the state of each iteration is ensured to have practical operability, the waste of algorithm computing power is reduced, the velocity correction mechanism balances the exploration and constraint by adjusting the velocity to the critical value that does not violate the rules, satisfies the compliance requirement, and retains the driving force of the particle to adjust to a better solution, avoids the search range being too narrow or out of control, so that the algorithm can concentrate computing power in the effective range to optimize, accelerate convergence to an optimal medium storage scheme that meets the actual constraints and balances the cost and efficiency, and improve the algorithm iteration efficiency and the actual applicability of the medium management scheme.

[0110] S240, determining the individual optimal position of each particle and the global optimal position of the particle group based on the particle fitness value.

[0111] In the embodiments of the present application, the individual optimal position can be understood as the optimal position (i.e., the storage scheme with the lowest fitness value) reached by each particle in the iteration process, which is the best solution in the historical search range of the particle. The global optimal position can be understood as the optimal position reached by all particles in the iteration process (i.e., the scheme with the lowest fitness value among the individual optimal positions of all particles), which is the global best solution in the current group search range.

[0112] Specifically, in the particle swarm iteration process, each particle will constantly adjust its own position and calculate the fitness value of the new position. Each particle compares the fitness value of the current position with the historical optimal fitness value of the particle (i.e. the lowest fitness value of all past positions of the particle), and if the current fitness value is less than the historical optimal fitness value, the current position is updated as the individual optimal position of the particle; otherwise, the individual optimal position remains unchanged.

[0113] Among all the individual optimal positions of the particles, the position with the lowest fitness value is selected as the global optimal position of the entire particle swarm. As the iteration proceeds, if the individual optimal position fitness value of a particle is lower than the fitness value of the current global optimal position, the global optimal position is updated to the individual optimal position of the particle.

[0114] S250, based on the motion information, individual optimal position and global optimal position of each particle in the particle swarm, updating the motion information of each particle.

[0115] Specifically, based on the motion information, individual optimal position and global optimal position of each particle in the particle swarm, the position and speed of the particle are updated according to the update formula of the particle optimization algorithm V id k+1 = w x v k id + c1 x r1 x (p id k - x id k ) + c2 x r2 x (p gd k - x id k ), X id k+1 = x id k + V id k+1 , wherein v k id is the d-dimensional speed of the i-th particle at the k-th iteration, x id k is the d-dimensional position of the i-th particle at the k-th iteration, w is the inertia weight, used to adjust the degree of retaining the speed of the particle in the last iteration, c1 and c2 are the individual learning factor and the group learning factor respectively, used to control the influence strength of individual experience and the influence strength of group experience respectively, r1 and r2 are random numbers between 0 and 1, used to increase the randomness and diversity of the search, p id k is the individual optimal position of the i-th particle at the k-th iteration, and p gd k is the global optimal position at the k-th iteration.

[0116] S260, return to execute the operation of calculating the particle fitness value of each particle based on the motion information of each particle through the fitness function until the preset iteration termination condition is met.

[0117] Specifically, return to execute the operation of calculating the particle fitness value of each particle based on the motion information of each particle through the fitness function until the preset iteration termination condition is met, for example, the preset number of iterations is reached or the change amount of the fitness value of the global optimal position is less than the preset change threshold.

[0118] S270, according to the particle fitness value of each particle at the iteration termination, the ideal medium storage amount of each medium storage device in each time interval is obtained.

[0119] Specifically, at the iteration termination, the particle with the lowest fitness value is selected from all particles, and the position of the particle is a multi-dimensional vector, each dimension corresponding to the storage amount of a medium storage device in a time interval. If multiple medium storage devices are involved, the position vector of the particle will be further expanded. For example, taking the cash management of ATM as an example, for the ideal medium storage amount of 2 ATMs in 4 time periods, the position vector of the optimal particle can be [70,000, 50,000, 90,000, 30,000, 60,000, 40,000, 80,000, 20,000], the first 4 dimensions correspond to the storage amount of the first device in each time period, and the last 4 dimensions correspond to the storage amount of the second device in each time period.

[0120] S280, according to the current system time and the ideal medium storage amount of each medium storage device in each time period, determine the medium management strategy for each target medium storage device in the target area range, to update the medium storage state in each target medium storage device.

[0121] The technical scheme of the embodiment of the present application generates medium state description information in real time in response to a medium access operation; determines a target time interval and obtains each generated target medium state description information when a cooperative medium management condition is met; initializes the motion information of each particle of a particle swarm based on each target medium state description information, calculates the particle fitness value through a fitness function, determines the individual optimal position of each particle and the global optimal position of the particle swarm; updates the motion information of each particle based on the motion information, the individual optimal position and the global optimal position of each particle in the particle swarm; returns to perform the operation of calculating the particle fitness value through the fitness function until a preset iteration termination condition is met; obtains the ideal medium storage amount of each medium storage device under each time interval according to the particle fitness value of each particle, realizes the conversion of medium management optimization into a multi-dimensional space optimization problem, ensures that the optimization direction is consistent with the actual target through the quantitative evaluation of the fitness function, reduces the subjective deviation of experience judgment, avoids falling into a local optimum through the dynamic updating mechanism of the individual optimal position and the global optimal position, expands the search range through group iteration, finally finds the global ideal medium storage amount, realizes the cooperation of multiple devices and resource balance; in combination with the current system time, determines the medium management strategy of each target medium storage device in the target area range to update the medium storage state, can dynamically adjust the deviation in time, ensures that the storage state is always reasonable under the scene of business fluctuation and the like, improves the service quality, reduces the operation cost, improves the management efficiency, optimization pertinence, accuracy and the cooperative management precision of the target area medium storage device, and enhances the adaptability to complex business scenarios.

[0122] Embodiment three

[0123] Figure 3A flowchart of another method for cooperative medium management between medium storage devices is provided for the third embodiment of the present application. This embodiment is a refinement of the above-mentioned embodiment of "determining a medium management strategy for each target medium storage device in the target area range according to the current system time and the ideal medium storage amount of each medium storage device in each time period". Specifically, it can include: identifying each medium storage device as a first type of medium storage device that needs to be supplemented with medium and a second type of medium storage device that can transfer medium according to the current system time and the ideal medium storage amount of each medium storage device in each time period; obtaining a current processing device in each first type of medium storage device and obtaining a target ideal medium storage amount of the current processing device at the current system time; if the real-time medium storage amount of the current processing device at the current system time is less than the target ideal medium storage amount, calculating the medium addition cost value when directly adding medium to the current processing device; calculating the minimum medium scheduling cost value when scheduling medium to the current processing device using at least one second type of medium storage device; if the minimum medium scheduling cost value is less than the medium addition cost value, generating a medium scheduling strategy matching the minimum medium scheduling cost value for the current processing device; and if the minimum medium scheduling cost value is greater than or equal to the medium addition cost value, generating a medium addition strategy for the current processing device.

[0124] Accordingly, as shown in Figure 3 the method includes:

[0125] S310, in response to the medium access operation of each medium storage device in the target area range, real-time generating medium state description information matching the medium access operation.

[0126] The medium state description information includes a set of medium access operations performed on a set of medium storage devices at a set time, and the medium storage amount in the set of medium storage devices after the access operation is performed.

[0127] S320, when the cooperative medium management condition is met, determining a target time interval matching the cooperative medium management condition, and obtaining each target medium state description information generated in the target time interval.

[0128] S330, according to each target medium state description information, using a particle optimization algorithm for iterative processing to obtain the ideal medium storage amount of each medium storage device in at least one time period.

[0129] S340, according to the current system time and the ideal medium storage amount of each medium storage device in each time period, identifying each medium storage device as a first type of medium storage device that needs to be supplemented with medium and a second type of medium storage device that can transfer medium.

[0130] Specifically, according to the current system time, a time period to which the time belongs is matched, and ideal medium storage amounts of each medium storage device in the corresponding time period are determined. Real-time medium storage amounts of each medium storage device at the current time point are obtained (such as being read by a sensor or a system record), which are compared with the ideal medium storage amounts of the device in the current time period, and a device state is judged.

[0131] When the real-time medium storage amount of the device is less than the ideal medium storage amount in the current time period, it is indicated that the device is short of medium, and the device is determined as a first-type medium storage device that needs to be supplemented with medium. When the real-time medium storage amount of the device is greater than the ideal medium storage amount in the current time period, it is indicated that the device has a medium surplus, which exceeds the optimal demand and may cause resource idling, and the device is determined as a second-type medium storage device that can transfer medium to other devices.

[0132] S350, a current processing device is obtained in each first-type medium storage device, and a target ideal medium storage amount of the current processing device at the current system time is obtained.

[0133] S360, if a real-time medium storage amount of the current processing device at the current system time is less than the target ideal medium storage amount, a medium addition cost value when medium is directly added to the current processing device is calculated.

[0134] Specifically, in the identified first-type medium storage device, the current processing device is obtained, for example, which can be selected according to a preset rule such as a device priority or a medium shortage amount. The target ideal medium storage amount of the current processing device at the current system time is obtained.

[0135] If the real-time medium storage amount is less than the target ideal medium storage amount, it is indicated that the device has a medium gap and needs to be supplemented with medium. According to a medium addition cost value = an addition fixed cost + (a unit medium addition cost × a medium gap amount), the medium addition cost value is calculated. The medium gap amount = the target ideal medium storage amount - the real-time medium storage amount.

[0136] S370, a minimum medium scheduling cost value when medium is scheduled to the current processing device by using at least one second-type medium storage device is calculated.

[0137] Specifically, the transferable amount of each second-type medium storage device is determined according to the formula: transferable amount = real-time medium storage amount of the second-type device - target ideal medium storage amount of the second-type device. When the total transferable amount of all second-type devices is greater than or equal to the gap amount of the current processing device, it means that the medium gap of the current processing device can be satisfied by scheduling, and a scheduling scheme based on a single device or multiple devices is generated. According to the scheduling scheme and the formula: device medium scheduling cost = scheduling fixed cost corresponding to the device + (unit medium scheduling cost corresponding to the device x medium scheduling amount), the device medium scheduling cost of each second-type device is calculated, and then the device medium scheduling costs involved in the scheduling scheme are summed to obtain the medium scheduling cost corresponding to each scheduling scheme. The minimum medium scheduling cost is selected from the medium scheduling costs corresponding to the scheduling schemes.

[0138] S380, if the minimum medium scheduling cost is less than the medium addition cost, a medium scheduling strategy matching the minimum medium scheduling cost is generated for the current processing device.

[0139] Specifically, when the minimum medium scheduling cost is less than the medium addition cost, it means that it is more economical to transfer medium from other devices than to directly add, and a medium scheduling strategy matching the minimum medium scheduling cost is generated for the current processing device. The medium scheduling strategy can include: scheduling source device (involving second-type medium storage device), medium amount to be transferred out by each source device, scheduling target device (current processing device), and scheduling time, etc. information, to update the medium storage state in each target medium storage device.

[0140] S390, if the minimum medium scheduling cost is greater than or equal to the medium addition cost, a medium addition strategy is generated for the current processing device to update the medium storage state in each target medium storage device.

[0141] Specifically, when the minimum medium scheduling cost is greater than or equal to the medium addition cost, it means that it is more economical to directly add medium than to transfer from other devices or the costs are comparable, and a medium addition strategy is generated for the current processing device. The medium addition strategy can include: addition amount, addition method, and addition time, etc. information, to update the medium storage state in each target medium storage device.

[0142] In one specific example, the application scenario of cash management of a cash depositing machine is taken as an example, Figure 4 is a schematic diagram of a medium storage device cooperative medium management method suitable for embodiments of the present application, as Figure 4 shown, describes the core module and logical relationship for realizing cash storage, monitoring, communication and intelligent management in an ATM machine.

[0143] The cash storage unit is a cash container of the ATM, which internally includes a plurality of cash boxes (such as 100 yuan banknote boxes and 50 yuan banknote boxes) for storing different denominations of cash, and is an object of system management.

[0144] The sensing and monitoring unit is composed of a weight sensor or a photoelectric sensor and a collection module, and is used for real-time monitoring of the state of the cash storage unit. The weight sensor converts the cash amount by weighing the weight of the banknote box (such as calculating the total number of banknotes according to the weight of each 100 yuan banknote); the photoelectric sensor identifies the stacking height of the cash in the banknote box by the light shielding principle to determine the remaining amount; the collection module aggregates the sensor data and converts it into the cash storage amount to provide raw data for subsequent analysis.

[0145] The communication unit is a data transmission medium of the ATM and external systems, which can communicate with the bank back-end management system and other ATMs or intelligent devices. For example, the cash storage amount and device status (such as fault alarm) collected by the sensing and monitoring unit are sent to the bank back-end through the communication unit. The instructions (such as scheduling or adding banknotes of the target cash storage amount) issued by the bank back-end are received through the communication unit to synchronize the device with the bank operation strategy.

[0146] The control processing unit includes a particle swarm optimization algorithm function module, which calculates the target ideal cash storage amount of the current ATM through the particle swarm optimization algorithm based on the cash storage amount obtained by the communication unit, and judges to add or schedule banknotes based on this, and generates the corresponding adding or scheduling strategy. The strategy generated by the control processing unit is fed back to the bank operation team through the communication unit (or automatically triggers the adding or scheduling operation of the ATM), so that the state of each ATM cash storage unit is updated to the target ideal cash storage amount.

[0147] The technical scheme of the embodiment of the application generates medium state description information in real time in response to medium access operations; when the cooperative medium management condition is met, a target time interval is determined, and each target medium state description information generated is obtained; according to each target medium state description information, a particle optimization algorithm is used for iterative processing to obtain ideal medium storage amounts of each medium storage device in each time period; according to the current system time and the ideal medium storage amounts of each medium storage device in each time period, each medium storage device is identified as a first type and a second type of medium storage device; the current processing device and the target ideal medium storage amount are obtained; if the real-time medium storage amount is less than the target ideal medium storage amount, the medium addition generation value when the medium is directly added to the current processing device is calculated; the minimum medium scheduling generation value when the medium is scheduled from the second type of medium storage device to the current processing device is calculated; if the minimum medium scheduling generation value is less than the medium addition generation value, a medium scheduling strategy is generated; otherwise, a medium addition strategy is generated to update the medium storage state, a cost-optimal decision mechanism is constructed by comparing the cost of directly adding the medium and scheduling the medium from other devices, the resource allocation is more targeted, the operating cost is reduced while the efficiency is taken into account, the strategy can dynamically respond to the medium state change based on the current system time and the real-time storage amount, sudden situations can be quickly responded to, service interruption is avoided, service stability is improved, the global optimization of medium resources in the region is realized through the cooperation of the second type of device and the first type of device, the waste and shortage risks are reduced, single-device imbalance is avoided, the multi-device cooperative management efficiency is improved, the adaptability to complex business scenarios is enhanced, and the automation level of medium management is improved.

[0148] Embodiment four

[0149] Figure 5 A structural schematic diagram of a cooperative medium management device between medium storage devices provided by the fourth embodiment of the application is shown in FIG. 5. Figure 5 As shown in the figure, the device comprises a description generation module 510, a description acquisition module 520, a particle optimization module 530 and a medium update module 540, wherein:

[0150] The description generation module 510 is configured to generate medium state description information matching the medium access operation in real time in response to the medium access operation of each medium storage device in the target region range; wherein the medium state description information comprises a set of medium access operations performed on a set of medium storage devices at a set time, and the medium storage amount in the set of medium storage devices after the access operation is performed;

[0151] The description acquisition module 520 is configured to determine a target time interval matching the cooperative medium management condition when the cooperative medium management condition is met, and to obtain each target medium state description information generated in the target time interval;

[0152] a particle optimization module 530, configured to perform iterative processing on the target medium state description information according to a particle optimization algorithm to obtain ideal medium storage amounts of the medium storage devices in at least one time period;

[0153] a medium updating module 540, configured to determine a medium management strategy for the target medium storage devices in the target region range according to the current system time and the ideal medium storage amounts of the medium storage devices in the time periods, to update the medium storage states of the target medium storage devices.

[0154] The technical scheme of the embodiment of the present application can, in response to the medium access operation of the medium storage devices in the target region range, generate medium state description information matched with the medium access operation in real time, realize dynamic collection of the running state data of the medium storage devices, provide reliable data support for collaborative management, and reduce decision deviation caused by data lag or inaccuracy. When the collaborative medium management condition is met, a target time interval matched with the collaborative medium management condition is determined, and each target medium state description information generated in the target time interval is obtained. According to the target medium state description information, iterative processing is performed on the target medium state description information according to a particle optimization algorithm to obtain ideal medium storage amounts of the medium storage devices in the time periods, which can focus on key scenes and reduce invalid data processing, output the collaborative global optimal ideal storage amounts of the devices in the region based on multi-dimensional data, and determine a medium management strategy for the target medium storage devices in the target region range in combination with the current system time to update the medium storage states, which can dynamically adjust deviation in a timely manner, ensure that the storage states are always reasonable in scenes such as business fluctuation, improve service quality, reduce operating costs, improve management efficiency, optimization pertinence, accuracy, and collaborative management precision of the target region medium storage devices, and enhance the adaptability to complex business scenes.

[0155] On the basis of the above embodiments, the particle optimization module 530 is specifically configured to:

[0156] initialize motion information of each particle of the particle group based on the target medium state description information, and calculate a particle fitness value of each particle based on the motion information of each particle through a fitness function;

[0157] The motion information of the particle includes position information and speed information of the particle. The position information is used to represent the medium storage amounts of the medium storage devices in the time periods. The speed information is used to represent adjustment amplitudes of the medium storage amounts of the medium storage devices in the time periods.

[0158] determine an individual optimal position of each particle and a global optimal position of the particle group based on the particle fitness value;

[0159] update the motion information of each particle based on the motion information of each particle in the particle swarm, the individual optimal position and the global optimal position;

[0160] The operation of calculating the particle fitness value of each particle based on the motion information of each particle through the fitness function is returned to be performed until a preset iteration termination condition is met.

[0161] According to the particle fitness value of each particle at the iteration termination, the ideal medium storage amount of each medium storage device under each time interval is obtained.

[0162] On the basis of the above embodiments, the particle optimization module 530 is further configured to:

[0163] For each time period, calculate the historical statistical value of historical medium consumption based on the target medium state description information, and determine the medium range of the medium storage device in each time period according to the preset medium storage device management rule.

[0164] Generate a preset number of particles, and generate the initial position and initial velocity of each particle under the compliance condition according to the medium range of the medium storage device in each time period.

[0165] On the basis of the above embodiments, the particle optimization module 530 is further configured to:

[0166] Generate the initial position and initial velocity of each particle according to the medium range of the medium storage device in each time period;

[0167] Determine the compliance of the initial position and the initial velocity of each particle, and update the initial velocity of the target particle when the position of the target particle after updating the initial velocity at the initial position exceeds the medium range.

[0168] On the basis of the above embodiments, the particle optimization module 530 is further configured to:

[0169] Calculate the corresponding medium addition cost, medium shortage rate and medium overstock rate based on the preset medium storage device management rule and the motion information of each particle;

[0170] Calculate the particle fitness value of each particle based on the medium addition cost, medium shortage rate and medium overstock rate of each particle.

[0171] On the basis of the above embodiments, the medium updating module 540 is specifically configured to:

[0172] According to the current system time and the ideal medium storage amount of each medium storage device under each time period, identify each medium storage device as a first type of medium storage device that needs to be supplemented with medium and a second type of medium storage device that can transfer out medium.

[0173] acquire a current processing device in each first-type medium storage device, and acquire a target ideal medium storage amount of the current processing device at a current system time;

[0174] if the real-time medium storage amount of the current processing device at the current system time is less than the target ideal medium storage amount, calculate a medium adding cost value when directly adding medium into the current processing device;

[0175] calculate a minimum medium scheduling cost value when scheduling medium to the current processing device using at least one second-type medium storage device;

[0176] if the minimum medium scheduling cost value is less than the medium adding cost value, generate a medium scheduling strategy matching the minimum medium scheduling cost value for the current processing device;

[0177] if the minimum medium scheduling cost value is greater than or equal to the medium adding cost value, generate a medium adding strategy for the current processing device.

[0178] Further, on the basis of each of the above embodiments, the coordinated medium management device among medium storage devices can further include a fault prediction module and a fault early warning module, wherein:

[0179] the fault prediction module is configured to input the acquired operating parameters of the medium storage device into a pre-constructed fault prediction model to obtain a fault prediction result;

[0180] the fault early warning module is configured to trigger a corresponding fault early warning according to the fault prediction result.

[0181] The coordinated medium management device among medium storage devices provided by the embodiments of the present application can execute the coordinated medium management method among medium storage devices provided by any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0182] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information comply with relevant laws and regulations and do not violate public order and good customs.

[0183] In the technical solution of the present disclosure, the collected information is information and data authorized by the user or authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws and regulations and standards of relevant countries and regions, necessary security measures are taken, public order and good customs are not violated, and appropriate operation portals are provided for users to choose authorization or refusal.

[0184] In the technical solutions of the present disclosure, if an automatic decision is involved, a corresponding operation portal is provided for a user to select to agree or reject the automatic decision result; if the user selects to reject, an expert decision process is entered.

[0185] Embodiment Five

[0186] Figure 6 A structural schematic diagram of an electronic device 10 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the present application described and / or claimed in this document.

[0187] As Figure 6 shown, the electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11, wherein the memory stores computer programs that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer programs stored in the read-only memory (ROM) 12 or loaded into the random access memory (RAM) 13 from the storage unit 18. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0188] Various components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, a speaker, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0189] The processor 11 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the collaborative media management method among media storage devices, i.e.:

[0190] In response to the media access operation of each media storage device in the target area range, real-time generation of media state description information matched with the media access operation; wherein the media state description information includes a set of media access operations performed on a set of media storage devices at a set time, and the media storage amount in the set of media storage devices after the access operation is executed;

[0191] When the collaborative media management condition is met, determine the target time interval matched with the collaborative media management condition, and obtain each target media state description information generated in the target time interval;

[0192] According to each target media state description information, the particle optimization algorithm is used for iterative processing to obtain the ideal media storage amount of each media storage device in at least one time period;

[0193] According to the current system time and the ideal media storage amount of each media storage device in each time period, determine the media management strategy for each target media storage device in the target area range, to update the media storage state in each target media storage device.

[0194] In some embodiments, the collaborative media management method among media storage devices can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the collaborative media management method among media storage devices described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the collaborative media management method among media storage devices by any other appropriate means (e.g., by means of firmware).

[0195] The various embodiments of the systems and techniques described above can be implemented in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a load programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0196] Computer programs used to implement the processes of the application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer program

[0197] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store computer programs for use by or in connection with an instruction execution system, apparatus, or device. Computer-readable storage media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium will include one or more lines of a program of instructions in a transitory signal, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0198] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0199] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0200] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. A server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0201] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the scope of the present disclosure. For example, the steps recited in the present disclosure can be executed in parallel, executed in sequence, or executed in a different order, as long as the desired results of the present disclosure are achieved, and the present disclosure is not limited herein.

[0202] The specific embodiments described above are not intended to be limiting, and persons skilled in the art will appreciate that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the spirit and principles of the disclosure. Accordingly, the disclosure is not limited to the specific embodiments described above, but only by the scope of the appended claims.

Claims

1. A method for collaborative media management among media storage devices, characterized in that, include: In response to media access operations on various media storage devices within the target area, media status description information matching the media access operations is generated in real time; wherein, the media status description information includes the specified media access operation performed on the specified media storage device at a specified time, and the amount of media stored in the specified media storage device after the access operation is performed; When the conditions for collaborative media management are met, a target time interval matching the conditions for collaborative media management is determined, and the status description information of each target medium generated within the target time interval is obtained. Based on the state description information of each target medium, the particle optimization algorithm is used for iterative processing to obtain the ideal medium storage capacity of each medium storage device in at least one time period. Based on the current system time and the ideal media storage capacity of each media storage device in each time period, determine the media management strategy for each target media storage device within the target area, so as to update the media storage status in each target media storage device.

2. The method according to claim 1, characterized in that, Based on the state description information of each target medium, a particle optimization algorithm is used for iterative processing to obtain the ideal medium storage capacity of each medium storage device in at least one time period, including: The motion information of each particle in the particle swarm is initialized based on the state description information of each target medium, and the particle fitness value of each particle is calculated based on the motion information of each particle through the fitness function. The particle motion information includes: particle position information and velocity information; the position information is used to represent the amount of media storage in each media storage device in each time period; the velocity information is used to represent the adjustment range of the amount of media storage in each media storage device in each time period. The optimal position of each particle and the global optimal position of the particle swarm are determined based on the particle fitness value. The motion information of each particle is updated based on its motion information, individual optimal position, and global optimal position. Return to the operation of calculating the particle fitness value of each particle based on the motion information of each particle using the fitness function, until the preset iteration termination condition is met; Based on the particle fitness value of each particle at the end of the iteration, the ideal media storage capacity of each media storage device in each time interval is obtained.

3. The method according to claim 2, characterized in that, The motion information of each particle in the particle swarm is initialized based on the state description information of each target medium, including: For each time period, historical statistics of media consumption are calculated based on the target media status description information, and the media range of the media storage device in each time period is determined according to the preset media storage device management rules. Generate a preset number of particles, and based on the media storage device's media range in each time period, generate the initial position and initial velocity of each particle under compliance conditions.

4. The method according to claim 3, characterized in that, Based on the media range of the storage device in each time period, and under compliance conditions, the initial position and initial velocity of each particle are generated, including: Based on the media range of the media storage device in each time period, the initial position and initial velocity of each particle are generated; Determine the compliance of the initial position and initial velocity of each particle. When the position of the target particle after updating according to the initial velocity exceeds the medium range, update the initial velocity of the target particle.

5. The method according to claim 2, characterized in that, The fitness value of each particle is calculated based on its motion information using a fitness function, including: The corresponding media addition cost, media shortage rate, and media backlog rate are calculated based on the preset media storage device management rules and the motion information of each particle. The particle fitness value of each particle is calculated based on the medium addition cost, medium shortage rate, and medium accumulation rate of each particle.

6. The method according to claim 1, characterized in that, Based on the current system time and the ideal media storage capacity of each media storage device in each time period, determine the media management strategy for each target media storage device within the target area, including: Based on the current system time and the ideal media storage capacity of each media storage device in each time period, each media storage device is identified as a first-class media storage device that adds supplementary media and a second-class media storage device that can transfer out media. Obtain the current processing device in each type I media storage device, and obtain the target ideal media storage capacity of the current processing device at the current system time; If the real-time media storage capacity of the current processing device at the current system time is less than the target ideal media storage capacity, then calculate the media addition cost when directly adding media to the current processing device. Calculate the minimum media scheduling cost when scheduling media to the current processing device using at least one Class II media storage device; If the minimum media scheduling cost is less than the media addition cost, then generate a media scheduling strategy that matches the minimum media scheduling cost for the current processing device. If the minimum media scheduling cost is greater than or equal to the media addition cost, then a media addition strategy is generated for the current processing device.

7. The method according to claim 1, characterized in that, The collaborative media management method between media storage devices further includes: The obtained operating parameters of the media storage device are input into a pre-built fault prediction model to obtain fault prediction results; Based on the fault prediction results, the corresponding fault warning is triggered.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the collaborative media management method between media storage devices according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the collaborative media management method between media storage devices as described in any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the collaborative media management method between media storage devices according to any one of claims 1-7.

Images