A warehouse location allocation method and system for intelligent warehousing

Abstract

The application discloses a kind of intelligent warehousing's library location allocation method and system, comprising: according to the incentive generated by impact contact library location when handling equipment accesses, the vibration response signal of library location is collected by vibration sensing unit;Signal is normalized in amplitude, extract the main frequency value and damping ratio in power spectral density function and / or frequency response function, obtain the frequency characteristics of library location support structure;According to the comparison of the frequency characteristics and historical baseline characteristic information, obtain the residual stiffness information, and determine the loadable level of library location;The weight of the goods to be stored in the warehouse and the loadable level of each library location are obtained, and the library location is allocated accordingly, so that the total weight of goods in each library location does not exceed the upper limit corresponding to its loadable level.The application does not need additional inspection equipment, uses daily access operation to sense the residual stiffness of library location support structure in real time, dynamically adjusts the library location allocation strategy, solves the safety hazard and warehouse space waste problem caused by structural stiffness degradation, and realizes the collaborative optimization of safety and efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent warehousing, and in particular to a method and system for allocating storage locations in intelligent warehousing. Background Technology

[0002] In intelligent warehousing systems, storage location allocation is a core element determining warehouse operational efficiency and safety. Traditional storage location allocation methods primarily optimize based on static or semi-static attributes such as cargo turnover rate, relevance, volume, and weight. For example, frequently outbound goods are allocated to storage locations closest to the shipping outlet (turnover rate-oriented), goods that are frequently outbound simultaneously are allocated to adjacent storage locations (relevance-based storage), or load balancing is achieved by evenly distributing cargo weight. These methods all rely on a common assumption: the racking structure has an infinite lifespan and a constant load-bearing capacity, and the safe load limit for each storage location depends solely on design parameters, independent of usage history and accumulated damage.

[0003] However, the load-bearing capacity of shelving is dynamic in actual operation. During long-term service, shelving can experience structural performance degradation due to various reasons: alternating loads from frequent storage and retrieval operations lead to metal fatigue accumulation; accidental collisions with handling equipment may cause millimeter-level local dents or bends in the uprights or beams; bolts at connecting nodes gradually loosen under vibration; and uneven settlement of the foundation causes the shelving to tilt slowly. These minor damages are often difficult to detect with the naked eye in the early stages, but they significantly reduce the actual residual stiffness of the corresponding warehouse location's support structure. When the residual stiffness of a warehouse location decreases, its actual safe load capacity is lower than the design rating. If the warehouse management system continues to allocate heavy goods to that location according to the initial rated load, it will accelerate the spread of damage and may also trigger local instability or even a chain reaction of collapses.

[0004] Another common problem with existing warehouse location allocation methods is that allocation decisions are completely decoupled from the structural health of the shelving. Warehouse management systems can only obtain information such as "whether the location is vacant" and "design load capacity of the location," but cannot perceive the current remaining stiffness of the supporting structure of each location. To avoid safety risks, warehouse managers often adopt a conservative strategy, disabling all locations in areas suspected of being damaged, resulting in a large amount of idle and wasted storage space; or they simply ignore structural health issues and continue to use the warehouse according to the initial design parameters, creating potential safety hazards.

[0005] Furthermore, traditional methods for obtaining the health status of shelving structures (such as periodic manual inspections and offline non-destructive testing) require stopping warehouse operations, consuming significant time and manpower costs, and cannot achieve real-time, online status awareness. More importantly, these offline inspection methods do not have any data interaction with the storage location allocation system, and the inspection results cannot be automatically converted into the basis for adjusting storage location allocation strategies. Summary of the Invention

[0006] In view of the aforementioned deficiencies of the prior art, the technical problem to be solved by the present invention is to provide a method and system for intelligent warehousing location allocation, which aims to sense the remaining stiffness of the storage location support structure in real time to dynamically optimize the allocation of storage locations, eliminate safety hazards and improve the utilization rate of warehousing.

[0007] To achieve the above objectives, the first aspect of the present invention discloses a method for allocating storage locations in an intelligent warehouse, the method comprising: Step S1: Based on the impact excitation generated when the handling equipment performs storage and retrieval operations and impacts the corresponding storage location, the vibration response signal generated by the storage location in response to the impact excitation is collected by a vibration sensing unit deployed on the rack structure and corresponding to the storage location. Step S2: Perform amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal; extract the power spectral density function and / or frequency response function of the vibration response signal, and extract frequency characteristic parameters including at least the dominant frequency value and damping ratio from the power spectral density function or frequency response function to obtain the frequency characteristics of the support structure corresponding to the warehouse location in the current state; Step S3: Based on the comparison results between the frequency characteristics and the pre-stored historical baseline characteristic information, obtain the remaining stiffness information of the supporting structure corresponding to the storage location; based on the remaining stiffness information, determine the load-bearing capacity level of the storage location; wherein, the historical baseline characteristic information is used to characterize the correspondence between the frequency characteristics and the remaining stiffness of the supporting structure, and the load-bearing capacity level is associated with the upper limit of the weight of goods that can be stored. Step S4: Obtain the weight information of the goods to be stored and the carrying capacity level of each storage location in the warehousing environment; according to the weight information of the goods to be stored and the carrying capacity level of each storage location, allocate storage locations for the goods to be stored so that the total weight of the goods in each storage location after allocation does not exceed the weight limit corresponding to its carrying capacity level.

[0008] Optionally, in step S3, the remaining stiffness information of the support structure corresponding to the reservoir location is obtained based on the comparison result between the frequency characteristics and the pre-stored historical baseline characteristic information. This specifically includes: The frequency features are input into a pre-trained convolutional neural network model, which has learned the correspondence between the frequency features represented by the historical baseline features and the residual stiffness. The convolutional neural network model outputs the residual stiffness value of the support structure corresponding to the storage location, and simultaneously outputs the damage type identification result and the corresponding confidence level. The damage type includes one or more of the following: local buckling, loosening of connection nodes, column bending deformation, and foundation settlement.

[0009] Optionally, step S2 involves amplitude normalization of the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal. Specifically, this includes: Obtain the weight of the goods and the fork movement speed of the handling equipment corresponding to this storage and retrieval operation; Based on the preset weight-speed-impact force mapping relationship, determine the theoretical impact force reference value corresponding to the weight of the goods and the speed of the fork movement; Dividing the amplitude of the collected vibration response signal by the theoretical impact force reference value yields the normalized vibration response signal waveform, thus providing a comparable amplitude benchmark for the waveforms collected under different cargo weights and handling speeds.

[0010] Optionally, in step S3, determining the load-bearing capacity level of the storage location based on the remaining stiffness information specifically includes: The obtained remaining stiffness information is compared with multiple pre-stored stiffness threshold intervals; If the remaining stiffness falls within the first threshold range, the storage location is determined to be of a high load-bearing capacity, allowing the storage of any goods not exceeding the rated load. If the remaining stiffness falls within the second threshold range, the storage location is determined to be of medium load-bearing capacity, and only goods whose weight does not exceed the first percentage threshold of the rated load are allowed to be stored. If the remaining stiffness falls within the third threshold range, the storage location is determined to be of low load-bearing capacity, and only goods weighing no more than the second percentage threshold of the rated load are allowed to be stored, and a maintenance prompt is triggered. If the remaining stiffness falls within the fourth threshold range, the storage location is determined to be a no-load level, prohibiting the storage of any goods, and an emergency maintenance work order is triggered. The first threshold interval, the second threshold interval, the third threshold interval, and the fourth threshold interval are divided in descending order of remaining stiffness, and the endpoint values ​​of each interval are determined according to a preset ratio based on the initial remaining stiffness of the storage location.

[0011] Optionally, in step S1, the start time of the impact excitation is triggered in the following manner: An acceleration trigger sensor is installed on the handling equipment. When the forks of the handling equipment come into contact with the shelves or pallets of the storage location, the acceleration trigger sensor generates a trigger signal. The vibration sensing unit switches from a low-power standby state to a high-speed acquisition state according to the trigger signal, and acquires the vibration response signal at a first sampling rate higher than the normal sampling rate. The acquisition time covers at least a first preset time before the impact occurs to a second preset time after the impact occurs.

[0012] The second aspect of this invention discloses a smart warehousing storage location allocation system, the system comprising: a signal acquisition module, a signal processing module, a storage location availability confirmation module, and a storage location allocation module; The signal acquisition module is used to acquire the vibration response signal of the storage location in response to the impact excitation generated by the impact contact between the handling equipment and the corresponding storage location when the handling equipment performs storage and retrieval operations, through a vibration sensing unit deployed on the rack structure and corresponding to the storage location. The signal processing module is used to perform amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal; extract the power spectral density function and / or frequency response function of the vibration response signal, and extract frequency characteristic parameters including at least the dominant frequency value and damping ratio from the power spectral density function or frequency response function to obtain the frequency characteristics of the support structure corresponding to the warehouse location in the current state; The storage location load-bearing capacity confirmation module is used to obtain the remaining stiffness information of the supporting structure corresponding to the storage location based on the comparison result between the frequency characteristics and the pre-stored historical baseline characteristic information; and to determine the load-bearing capacity level of the storage location based on the remaining stiffness information; wherein, the historical baseline characteristic information is used to characterize the correspondence between the frequency characteristics and the remaining stiffness of the supporting structure, and the load-bearing capacity level is associated with the upper limit of the weight of goods that can be stored. The storage location allocation module is used to obtain the weight information of the goods to be stored and the carrying capacity level of each storage location in the warehousing environment; and to allocate storage locations for the goods to be stored according to the weight information of the goods to be stored and the carrying capacity level of each storage location, so that the total weight of the goods in each storage location after allocation does not exceed the weight limit corresponding to its carrying capacity level.

[0013] Optionally, when the storage location bearing capacity confirmation module obtains the remaining stiffness information of the support structure corresponding to the storage location based on the comparison results between the frequency characteristics and pre-stored historical baseline characteristic information, it specifically includes: The frequency features are input into a pre-trained convolutional neural network model, which has learned the correspondence between the frequency features represented by the historical baseline features and the residual stiffness. The convolutional neural network model outputs the residual stiffness value of the support structure corresponding to the storage location, and simultaneously outputs the damage type identification result and the corresponding confidence level. The damage type includes one or more of the following: local buckling, loosening of connection nodes, column bending deformation, and foundation settlement.

[0014] Optionally, the signal processing module performs amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal, specifically including: Obtain the weight of the goods and the fork movement speed of the handling equipment corresponding to this storage and retrieval operation; Based on the preset weight-speed-impact force mapping relationship, determine the theoretical impact force reference value corresponding to the weight of the goods and the speed of the fork movement; Dividing the amplitude of the collected vibration response signal by the theoretical impact force reference value yields the normalized vibration response signal waveform, thus providing a comparable amplitude benchmark for the waveforms collected under different cargo weights and handling speeds.

[0015] Optionally, the storage location bearing capacity confirmation module, when performing the process of determining the bearing capacity level of the storage location based on the remaining stiffness information, specifically includes: The obtained remaining stiffness information is compared with multiple pre-stored stiffness threshold intervals; If the remaining stiffness falls within the first threshold range, the storage location is determined to be of a high load-bearing capacity, allowing the storage of any goods not exceeding the rated load. If the remaining stiffness falls within the second threshold range, the storage location is determined to be of medium load-bearing capacity, and only goods whose weight does not exceed the first percentage threshold of the rated load are allowed to be stored. If the remaining stiffness falls within the third threshold range, the storage location is determined to be of low load-bearing capacity, and only goods weighing no more than the second percentage threshold of the rated load are allowed to be stored, and a maintenance prompt is triggered. If the remaining stiffness falls within the fourth threshold range, the storage location is determined to be a no-load level, prohibiting the storage of any goods, and an emergency maintenance work order is triggered. The first threshold interval, the second threshold interval, the third threshold interval, and the fourth threshold interval are divided in descending order of remaining stiffness, and the endpoint values ​​of each interval are determined according to a preset ratio based on the initial remaining stiffness of the storage location.

[0016] Optionally, the system further includes a signal acquisition triggering module, which, when the signal acquisition module is working, is specifically used for: When the forks of the handling equipment come into contact with the shelves or pallets of the storage location, an acceleration trigger sensor installed on the handling equipment generates a trigger signal; the vibration sensing unit switches from a low-power standby state to a high-speed acquisition state according to the trigger signal, and acquires the vibration response signal at a first sampling rate higher than the normal sampling rate, with the acquisition time covering at least a first preset time before the impact to a second preset time after the impact.

[0017] The beneficial effects of this invention are as follows: 1. This invention utilizes the inherent impact of storage and retrieval operations in warehousing as an excitation source. Vibration response signals are collected by vibration sensing units deployed on the shelves, and parameters such as the dominant frequency and damping ratio are extracted from the frequency domain characteristics. The remaining stiffness information of the support structure of each storage location is obtained by comparing it with historical baselines. The entire process is carried out synchronously with daily storage and retrieval operations, without the need for dedicated excitation equipment or offline testing by stopping warehouse operations, significantly reducing the time and labor costs of structural health monitoring. 2. Traditional storage location allocation relies solely on static design load, failing to detect stiffness degradation. This leads to either risky overloading causing safety hazards or conservative prohibition of storage resulting in wasted space. This invention converts remaining stiffness into a load-bearing capacity level and uses it as the core constraint for storage location allocation. This invention automatically matches the load-bearing capacity level based on the weight of the goods, ensuring that storage locations with degraded stiffness only store light goods or prohibit storage of goods, while healthy storage locations can still be fully utilized, maximizing warehouse utilization while ensuring structural safety. 3. Early damage such as local buckling, loose connections, and foundation settlement is difficult to detect with strain gauges or tilt sensors, but it causes measurable changes in frequency domain characteristics. This invention, through frequency domain analysis and comparison with historical baselines, can identify anomalies when the damage is still at the micrometer level and invisible to the naked eye, and proactively adjust the storage location allocation strategy to prevent cascading failures caused by accumulated damage at the source, significantly improving the intrinsic safety level of the warehousing system. 4. This invention not only outputs residual stiffness and load-bearing capacity, but also further identifies damage types and generates structured maintenance work orders. After repair, a re-inspection confirms stiffness recovery, and the storage location level automatically increases. This closed-loop mechanism keeps the shelving in a controlled and healthy state, extending the overall service life of the shelving and reducing the total life-cycle maintenance cost.

[0018] In summary, this invention proposes an intelligent warehouse location allocation method based on dynamic excitation structural response recognition. By utilizing the impact generated during daily storage and retrieval operations as the excitation source, it collects the vibration response signals of the shelving, extracts frequency domain features, and compares them with historical baselines to obtain the residual stiffness information of the supporting structure of each location. This allows for dynamic classification of load-bearing capacity levels, guiding the WMS system to achieve optimal location allocation that balances safety and efficiency. This method solves the technical challenges of existing technologies, such as the decoupling of location allocation from structural health status, difficulty in early identification of micro-damage, and the coexistence of safety hazards and space waste. It offers significant advantages, including no need for additional inspections, early warning, closed-loop control, and ease of deployment, providing an innovative solution for the safe operation and refined management of intelligent warehousing systems. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a method for allocating storage locations in an intelligent warehouse, provided by a specific embodiment of the present invention. Figure 2This is a schematic diagram of the structure of an intelligent warehousing location allocation system provided in a specific embodiment of the present invention. Detailed Implementation

[0020] This invention discloses a method and system for intelligent warehousing location allocation. Those skilled in the art can refer to the content of this document and appropriately modify the technical details to implement it. It should be particularly noted that all similar substitutions and modifications are obvious to those skilled in the art and are considered to be included in this invention. The methods and applications of this invention have been described through preferred embodiments. Those skilled in the art can obviously modify or appropriately change and combine the methods and applications described herein without departing from the content, spirit, and scope of this invention to implement and apply the technology of this invention.

[0021] This invention provides a method for allocating storage locations in an intelligent warehouse, such as... Figure 1 As shown, the method includes: Step S1: Based on the impact excitation generated when the handling equipment performs storage and retrieval operations and impacts the corresponding storage location, the vibration response signal generated by the storage location in response to the impact excitation is collected by a vibration sensing unit deployed on the rack structure and corresponding to the storage location.

[0022] It should be noted that in step S1, when the handling equipment (such as a stacker crane, AGV, shuttle, etc.) performs goods storage and retrieval operations, its forks will physically contact the rack beams or pallets of the target storage location, generating an impact excitation at that instant. This invention utilizes this inherent operational impact as the excitation source for structural health monitoring, eliminating the need for additional specialized equipment such as force hammers or vibrators. To collect vibration signals, this invention pre-deploys vibration sensing units corresponding to each storage location on the rack structure (e.g., installing miniature accelerometers at the nodes of the uprights or below the beams of each storage location). When the forks of the handling equipment contact the storage location, the storage location and its supporting structure will generate weak mechanical vibrations due to the impact, i.e., a vibration response signal. The vibration sensing unit captures this signal at a high sampling rate (e.g., 1000Hz), recording the waveform of the entire process from the occurrence of the impact to the decay of the vibration. The key to this step is to combine routine operations with inspection actions, making each storage and retrieval a structural check-up, thereby continuously acquiring raw data reflecting the health status of the shelves without interrupting warehouse operations or adding extra time costs.

[0023] In this specific embodiment, in step S1, the start time of the impact excitation is triggered in the following manner: An acceleration trigger sensor is installed on the handling equipment. When the forks of the handling equipment come into contact with the shelves or pallets in the storage location, the acceleration trigger sensor generates a trigger signal. The vibration sensing unit switches from a low-power standby state to a high-speed acquisition state according to the trigger signal. It acquires the vibration response signal at a first sampling rate higher than the normal sampling rate. The acquisition time covers at least the first preset time before the impact occurs to the second preset time after the impact occurs.

[0024] It should be noted that this embodiment specifically defines the triggering method for the initial moment of impact excitation in step S1 and the operating mode of the vibration sensing unit. To achieve accurate event synchronization, an acceleration trigger sensor is installed at the root of the forks of the handling equipment. This sensor continuously monitors the acceleration changes of the forks. When the forks make physical contact with the rack beams or pallets in the storage location, a steep acceleration pulse is generated (the rise time is typically less than 5 milliseconds). The acceleration trigger sensor captures this pulse and immediately generates a trigger signal. Under normal conditions, the vibration sensing unit is in a low-power standby state (the sampling rate can be reduced to tens of hertz, used only for environmental vibration monitoring or self-testing) to save energy and reduce data volume. Once the trigger signal is received, it immediately switches to a high-speed acquisition state, starting to record the vibration response signal at a first sampling rate higher than the normal sampling rate (e.g., not less than 1000Hz). The acquisition duration setting must ensure that the entire process from the onset of the impact to the complete decay of the vibration is captured: covering the first preset duration before the impact and the second preset duration after the impact. This event-driven working mode achieves precise synchronization of access triggering and triggering acquisition, which not only ensures the integrity and validity of the detection data, but also minimizes the power consumption and computational burden of the system.

[0025] Step S2: Normalize the amplitude of the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal; extract the power spectral density function and / or frequency response function of the vibration response signal, and extract frequency characteristic parameters, including at least the dominant frequency value and damping ratio, from the power spectral density function or frequency response function to obtain the frequency characteristics of the support structure corresponding to the warehouse location in the current state.

[0026] It should be noted that the core task of step S2 is to transform the original vibration waveform into frequency domain characteristics that can characterize the structural stiffness. Because the weight of the goods and the fork speed of the handling equipment vary with each storage and retrieval operation, the resulting impact energy will differ significantly—the waveform amplitude generated by heavy goods and high-speed impacts is much greater than that of light goods and low-speed impacts. To eliminate the impact of this difference on subsequent analysis, this invention first performs amplitude normalization processing on the vibration response signal. Specifically, the system acquires the actual weight of the goods and the fork speed of the operation, calculates the theoretical impact force reference value based on the pre-calibrated "weight-velocity-impact force mapping relationship," and then divides the amplitude value of each sampling point of the measured waveform by this reference value, so that regardless of the actual working conditions, the normalized waveform amplitude can reflect the dynamic characteristics of the rack structure itself and is independent of the excitation intensity. Subsequently, a fast Fourier transform is performed on the normalized waveform to obtain the power spectral density function (reflecting the energy distribution of different frequency components) and / or the frequency response function (reflecting the structure's response characteristics to a unit excitation). From these frequency domain functions, the system automatically extracts key feature parameters, including at least the dominant frequency (the frequency corresponding to the peak of the power spectrum, reflecting the structural stiffness) and the damping ratio (peak bandwidth, reflecting energy dissipation capacity). These parameters together constitute the frequency characteristics of the support structure corresponding to the storage location at the current moment. Unlike directly using the peak value or energy of the time domain waveform, frequency domain features are extremely sensitive to local stiffness changes and can exhibit significant drift even at the micro-damage stage (such as millimeter-level dents or minor bolt loosening), thus forming the technical basis for early damage identification.

[0027] In this specific embodiment, step S2 involves amplitude normalization of the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal. Specifically, this includes: Obtain the weight of the goods and the fork movement speed of the handling equipment corresponding to this storage and retrieval operation; Based on the preset weight-velocity-impact force mapping relationship, determine the theoretical impact force reference value corresponding to the weight of the goods and the speed of the fork movement; Dividing the amplitude of the collected vibration response signal by the theoretical impact force reference value yields the normalized vibration response signal waveform, enabling the waveforms collected under different cargo weights and handling speeds to have comparable amplitude benchmarks.

[0028] It should be noted that this embodiment provides a detailed refinement of the amplitude normalization process in step S2. Since the waveform amplitude of the vibration response signal is positively correlated with the weight of the goods and the fork speed of the handling equipment during the current storage and retrieval operation—the heavier the goods and the faster the speed, the greater the impact force and the higher the waveform amplitude—if the waveforms under different operating conditions are directly compared, the waveform of a healthy shelf under light goods and low-speed storage and retrieval may have similar amplitudes to that of a damaged shelf under heavy goods and high-speed storage and retrieval, leading to misjudgment. To eliminate this interference, this solution first obtains the weight of the goods and the fork speed of the current operation from the WMS or field sensors; then, it queries the pre-calibrated "weight-speed-impact force mapping relationship" to calculate the theoretical impact force reference value under this operating condition; finally, it divides the amplitude value of each sampling point in the measured waveform by this reference value to achieve amplitude normalization. After this processing, regardless of how the actual operating conditions change, the amplitude of the normalized waveform is decoupled from the excitation intensity and only reflects the dynamic response characteristics of the shelf structure itself, thus providing a unified comparison benchmark for waveforms collected under different storage locations, times, and operating conditions.

[0029] Step S3: Based on the comparison results between the frequency characteristics and the pre-stored historical baseline characteristics, obtain the remaining stiffness information of the support structure corresponding to the reservoir location; based on the remaining stiffness information, determine the load-bearing capacity level of the reservoir location.

[0030] Among them, historical baseline feature information is used to characterize the correspondence between frequency characteristics and the remaining stiffness of the supporting structure, and the load-bearing capacity level is associated with the upper limit of the weight of goods that can be stored.

[0031] It should be noted that step S3 is responsible for comparing the real-time frequency characteristics obtained in step S2 with the pre-stored historical baseline characteristic information to quantitatively assess the remaining stiffness of the warehouse support structure and classify its load-bearing capacity level accordingly. The historical baseline characteristic information is a "health fingerprint" collected and statistically analyzed through multiple standard access operations when the warehouse is first put into use and the structure is confirmed to be intact. Essentially, it characterizes the correspondence between the frequency characteristics of the warehouse in a healthy state and the remaining stiffness of the support structure (e.g., the higher the dominant frequency and the smaller the damping ratio, the greater the remaining stiffness). During operation, the system compares the real-time frequency characteristics with the baseline: if the dominant frequency does not shift significantly and the damping ratio does not increase significantly, it indicates that the structure is in good condition and the remaining stiffness is close to the initial value; if the dominant frequency shifts significantly downward and the damping ratio increases, it indicates that the stiffness has degraded, and the greater the drift, the more severe the degradation. Based on this comparison result, the system calculates the quantitative remaining stiffness information (e.g., expressing the current stiffness as a percentage of the initial stiffness). Subsequently, the load-bearing capacity of the storage location is determined based on the remaining stiffness information—different stiffness ranges correspond to different maximum allowable weight limits for stored goods. For example, high-stiffness areas can store heavy goods, medium-stiffness areas can only store light goods, and low-stiffness areas prohibit storage and trigger an alarm.

[0032] In this specific embodiment, step S3 obtains the remaining stiffness information of the support structure corresponding to the reservoir location based on the comparison results between the frequency characteristics and the pre-stored historical baseline characteristic information, specifically including: The frequency features are input into a pre-trained convolutional neural network model, which has learned the correspondence between the frequency features represented by historical baseline features and the residual stiffness. The convolutional neural network model outputs the residual stiffness value of the support structure corresponding to the storage location, and simultaneously outputs the damage type identification result and the corresponding confidence level. The damage type includes one or more of the following: local buckling, loosening of connection nodes, column bending deformation, and foundation settlement.

[0033] It should be noted that this embodiment further defines a preferred implementation method for obtaining residual stiffness information in step S3: inputting frequency features into a pre-trained convolutional neural network model. This model has completed its learning task offline, and its training data comes from in-depth mining of historical baseline feature information—specifically, by artificially simulating various typical damages on a health shelf, collecting corresponding frequency features and measuring actual residual stiffness, forming a labeled dataset of frequency features, residual stiffness, and damage type. The convolutional neural network, with its powerful nonlinear mapping capability, can automatically extract high-dimensional features from input parameters such as the dominant frequency and damping ratio, and directly output the residual stiffness value, while also outputting the damage type classification result and its confidence level. Compared to traditional numerical calculation formulas, the neural network does not require pre-calibrating the physical equations for each type of damage, making it more adaptable to complex working conditions; simultaneously, the output confidence level can serve as a basis for subsequent decisions (for example, when the confidence level is below a threshold, the system can switch to an alternative calculation method). The damage type identification results can also be used to generate accurate maintenance work orders, significantly improving maintenance efficiency.

[0034] In this specific embodiment, step S3, based on the remaining stiffness information, determines the load-bearing capacity level of the storage location, specifically including: The obtained residual stiffness information is compared with multiple pre-stored stiffness threshold intervals; If the remaining stiffness falls within the first threshold range, the storage location is determined to be of high load-bearing capacity, allowing the storage of any goods not exceeding the rated load. If the remaining stiffness falls within the second threshold range, the storage location is determined to be of medium load-bearing capacity, and only goods whose weight does not exceed the first percentage threshold of the rated load are allowed to be stored. If the remaining stiffness falls within the third threshold range, the storage location is determined to be of low load-bearing level, and only goods whose weight does not exceed the second percentage threshold of the rated load are allowed to be stored, and a maintenance prompt is triggered. If the remaining stiffness falls within the fourth threshold range, the storage location is determined to be a no-load level, prohibiting the storage of any goods, and an emergency maintenance work order is triggered. The first threshold interval, the second threshold interval, the third threshold interval, and the fourth threshold interval are divided in descending order of remaining stiffness, and the endpoint values ​​of each interval are determined according to the initial remaining stiffness of the storage location according to a preset ratio.

[0035] It should be noted that this embodiment specifies the detailed rules for determining the load-bearing capacity level based on the remaining stiffness information in step S3. The system pre-stores multiple stiffness threshold intervals, which are divided into four intervals from high to low remaining stiffness. The initial remaining stiffness of each storage location in a healthy state is used as a baseline value, and the endpoints of the threshold intervals are determined according to a preset proportion of the initial stiffness (e.g., the first interval is ≥95% initial stiffness, the second interval is 85%-95%, the third interval is 70%-85%, and the fourth interval is <70%). After comparing the obtained real-time residual stiffness values ​​to determine which interval they fall into, the system automatically performs the corresponding level determination: falling into the first interval is determined to be a high load-bearing level, allowing the storage of any goods not exceeding the rated load; falling into the second interval is determined to be a medium load-bearing level, allowing only the storage of goods whose weight does not exceed the first percentage threshold of the rated load (e.g., 70%); falling into the third interval is determined to be a low load-bearing level, allowing only the storage of goods whose weight does not exceed the second percentage threshold of the rated load (e.g., 40%), and triggering a maintenance prompt; falling into the fourth interval is determined to be a prohibited level, prohibiting the storage of any goods, and immediately triggering an emergency maintenance work order. This tiered strategy achieves a gradual balance between safety and efficiency: slightly degraded storage locations can still store light goods to avoid space waste, while severely degraded storage locations are strictly prohibited from storing goods to eliminate safety risks.

[0036] Step S4: Obtain the weight information of the goods to be put into storage and the carrying capacity level of each storage location in the warehousing environment; based on the weight information of the goods to be put into storage and the carrying capacity level of each storage location, allocate storage locations for the goods to be put into storage so that the total weight of the goods in each storage location after allocation does not exceed the weight limit corresponding to its carrying capacity level.

[0037] It should be noted that in step S4, the system performs the final storage location allocation for the actual inbound task. First, it obtains the weight information of the goods to be inbound (which can be obtained through a weighbridge, conveyor belt scale, or manual input), and simultaneously reads the load-bearing capacity level of all available or vacant storage locations in the current warehousing environment from the database (this level is dynamically updated and stored in step S3). The core constraint of the allocation algorithm is that the total weight of goods allocated to each storage location (including the sum of the weight of existing goods in that location and the weight of newly inbound goods) must not exceed the weight limit corresponding to the load-bearing capacity level of that location. That is, if the remaining stiffness of a storage location has dropped to 60% of its rated load capacity, the system will not allocate any goods exceeding this limit to it. Under the premise of satisfying this constraint, the system can further combine other optimization objectives (such as shortest picking path, turnover rate matching, etc.) for comprehensive decision-making. Through this step, the present invention ensures that every item is placed in a storage location that is structurally capable of safely supporting it, while avoiding storage waste caused by conservatively disabling all degraded areas—because slightly degraded storage locations can still be used to store light goods. Thus, from impact excitation to frequency domain analysis, from stiffness assessment to grade classification, and then to storage location allocation under safety constraints, a complete, closed-loop, and deeply integrated intelligent warehousing storage location allocation method has been formed.

[0038] This invention utilizes the inherent impact of storage and retrieval operations as an excitation source. Vibration response signals are collected by vibration sensing units deployed on the shelves, and parameters such as the dominant frequency and damping ratio are extracted from the frequency domain characteristics. The remaining stiffness information of the support structure of each storage location is obtained by comparing it with historical baselines. The entire process is carried out synchronously with daily storage and retrieval operations, without the need for dedicated excitation equipment or offline testing that requires stopping warehouse operations, significantly reducing the time and labor costs of structural health monitoring.

[0039] Traditional warehouse space allocation relies solely on static design load capacity, failing to detect stiffness degradation. This leads to either reckless overloading causing safety hazards or conservative prohibition of storage resulting in wasted space. This invention converts remaining stiffness into a load-bearing capacity level, using it as the core constraint for warehouse space allocation. This invention automatically matches the load-bearing capacity level based on the weight of the goods, ensuring that warehouse spaces with degraded stiffness are used only for light goods or are prohibited from storage, while healthy warehouse spaces can still be fully utilized, maximizing warehouse utilization while ensuring structural safety.

[0040] Early damage such as local buckling, loose connections, and foundation settlement is difficult to detect with strain gauges or tilt sensors, but it can cause measurable changes in frequency domain characteristics. This invention, through frequency domain analysis and comparison with historical baselines, can identify anomalies when damage is still at the micrometer level and invisible to the naked eye. It can then proactively adjust storage location allocation strategies, preventing cascading failures caused by accumulated damage at the source, and significantly improving the intrinsic safety level of the storage system.

[0041] This invention not only outputs remaining stiffness and load-bearing capacity, but also further identifies damage types and generates structured maintenance work orders. After repairs are completed, stiffness recovery is confirmed through re-inspection, and the storage location level automatically increases. This closed-loop mechanism ensures that the shelving is always in a controlled and healthy state, extending the overall service life of the shelving and reducing total lifecycle maintenance costs.

[0042] In summary, this invention proposes an intelligent warehouse location allocation method based on dynamic excitation structural response recognition. By utilizing the impact generated during daily storage and retrieval operations as the excitation source, it collects the vibration response signal of the shelving, extracts frequency domain features, and compares them with historical baselines to obtain the remaining stiffness information of the supporting structure of each location. This allows for dynamic classification of load-bearing capacity levels, guiding the WMS system to achieve optimal location allocation that balances safety and efficiency. This method solves the technical challenges of existing technologies, such as the decoupling of location allocation from structural health status, difficulty in early identification of micro-damage, and the coexistence of safety hazards and space waste. It offers significant advantages, including no need for additional inspections, early warning, closed-loop control, and ease of deployment, providing an innovative solution for the safe operation and refined management of intelligent warehousing systems.

[0043] Based on the intelligent warehousing location allocation method provided above, this embodiment of the invention also provides an intelligent warehousing location allocation system, such as... Figure 2 As shown, the system includes: a signal acquisition module 201, a signal processing module 202, a storage location capacity confirmation module 203, and a storage location allocation module 204; The signal acquisition module 201 is used to collect the vibration response signal of the storage location in response to the impact excitation generated by the impact contact between the storage location and the storage location when the handling equipment performs storage and retrieval operations, through a vibration sensing unit deployed on the rack structure and corresponding to the storage location. The signal processing module 202 is used to perform amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal; extract the power spectral density function and / or frequency response function of the vibration response signal, and extract frequency characteristic parameters including at least the dominant frequency value and damping ratio from the power spectral density function or frequency response function to obtain the frequency characteristics of the support structure corresponding to the warehouse location in the current state; The storage location load-bearing capacity confirmation module 203 is used to obtain the remaining stiffness information of the support structure corresponding to the storage location based on the comparison results of frequency characteristics and pre-stored historical baseline characteristic information; and to determine the load-bearing capacity level of the storage location based on the remaining stiffness information; wherein, the historical baseline characteristic information is used to characterize the correspondence between frequency characteristics and the remaining stiffness of the support structure, and the load-bearing capacity level is associated with the upper limit of the weight of goods that can be stored. The storage location allocation module 204 is used to obtain the weight information of the goods to be stored and the carrying capacity level of each storage location in the warehousing environment; based on the weight information of the goods to be stored and the carrying capacity level of each storage location, the storage location is allocated to the goods to be stored so that the total weight of the goods in each storage location after allocation does not exceed the weight limit corresponding to its carrying capacity level.

[0044] Optionally, when the storage location bearing capacity confirmation module 203 obtains the remaining stiffness information of the support structure corresponding to the storage location based on the comparison results between frequency characteristics and pre-stored historical baseline characteristic information, it specifically includes: The frequency features are input into a pre-trained convolutional neural network model, which has learned the correspondence between the frequency features represented by historical baseline features and the residual stiffness. The convolutional neural network model outputs the residual stiffness value of the support structure corresponding to the storage location, and simultaneously outputs the damage type identification result and the corresponding confidence level. The damage type includes one or more of the following: local buckling, loosening of connection nodes, column bending deformation, and foundation settlement.

[0045] Optionally, the signal processing module 202 performs amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal, specifically including: Obtain the weight of the goods and the fork movement speed of the handling equipment corresponding to this storage and retrieval operation; Based on the preset weight-velocity-impact force mapping relationship, determine the theoretical impact force reference value corresponding to the weight of the goods and the speed of the fork movement; Dividing the amplitude of the collected vibration response signal by the theoretical impact force reference value yields the normalized vibration response signal waveform, enabling the waveforms collected under different cargo weights and handling speeds to have comparable amplitude benchmarks.

[0046] Optionally, the storage location bearing capacity confirmation module 203 determines the bearing capacity level of the storage location based on the remaining stiffness information, specifically including: The obtained residual stiffness information is compared with multiple pre-stored stiffness threshold intervals; If the remaining stiffness falls within the first threshold range, the storage location is determined to be of high load-bearing capacity, allowing the storage of any goods not exceeding the rated load. If the remaining stiffness falls within the second threshold range, the storage location is determined to be of medium load-bearing capacity, and only goods whose weight does not exceed the first percentage threshold of the rated load are allowed to be stored. If the remaining stiffness falls within the third threshold range, the storage location is determined to be of low load-bearing level, and only goods whose weight does not exceed the second percentage threshold of the rated load are allowed to be stored, and a maintenance prompt is triggered. If the remaining stiffness falls within the fourth threshold range, the storage location is determined to be a no-load level, prohibiting the storage of any goods, and an emergency maintenance work order is triggered. The first threshold interval, the second threshold interval, the third threshold interval, and the fourth threshold interval are divided in descending order of remaining stiffness, and the endpoint values ​​of each interval are determined according to the initial remaining stiffness of the storage location according to a preset ratio.

[0047] Optionally, the system also includes a signal acquisition triggering module, which, when the signal acquisition module 201 is working, is specifically used for: When the forks of the handling equipment come into contact with the shelves or pallets in the warehouse, an acceleration trigger sensor installed on the handling equipment generates a trigger signal. Based on the trigger signal, the vibration sensing unit switches from a low-power standby state to a high-speed acquisition state and acquires the vibration response signal at a first sampling rate higher than the normal sampling rate. The acquisition time covers at least the first preset time before the impact to the second preset time after the impact.

[0048] In summary, the intelligent warehouse location allocation method disclosed in this invention innovatively transforms the impact excitation generated by daily storage and retrieval operations into a natural source of information for structural health detection. By comparing the frequency domain analysis of the vibration response signal with historical baselines, the remaining stiffness of the supporting structure of each location is obtained in real time and quantified into a load-bearing level, which is directly coupled to the location allocation decision. This method achieves online perception of the structural health status at the location level and closed-loop control of the dynamic allocation strategy without requiring additional inspection shutdowns or dedicated excitation equipment. It fundamentally solves the contradiction between safety hazards and wasted storage space caused by neglecting stiffness degradation in traditional methods. This invention significantly improves the inherent safety level and space utilization of the warehousing system, reduces operation and maintenance costs, and provides core support for the evolution of intelligent warehousing towards structural health self-adaptation.

[0049] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0050] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0051] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for allocating storage locations in an intelligent warehouse, characterized in that, The method includes: Step S1: Based on the impact excitation generated when the handling equipment performs storage and retrieval operations and impacts the corresponding storage location, the vibration response signal generated by the storage location in response to the impact excitation is collected by a vibration sensing unit deployed on the rack structure and corresponding to the storage location. Step S2: Perform amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal; extract the power spectral density function and / or frequency response function of the vibration response signal, and extract frequency characteristic parameters including at least the dominant frequency value and damping ratio from the power spectral density function or frequency response function to obtain the frequency characteristics of the support structure corresponding to the warehouse location in the current state; Step S3: Based on the comparison results between the frequency characteristics and the pre-stored historical baseline characteristic information, obtain the remaining stiffness information of the supporting structure corresponding to the storage location; based on the remaining stiffness information, determine the load-bearing capacity level of the storage location; wherein, the historical baseline characteristic information is used to characterize the correspondence between the frequency characteristics and the remaining stiffness of the supporting structure, and the load-bearing capacity level is associated with the upper limit of the weight of goods that can be stored. Step S4: Obtain the weight information of the goods to be stored and the carrying capacity level of each storage location in the warehousing environment; according to the weight information of the goods to be stored and the carrying capacity level of each storage location, allocate storage locations for the goods to be stored so that the total weight of the goods in each storage location after allocation does not exceed the weight limit corresponding to its carrying capacity level.

2. The intelligent warehousing location allocation method according to claim 1, characterized in that, In step S3, the remaining stiffness information of the support structure corresponding to the reservoir location is obtained based on the comparison result between the frequency characteristics and the pre-stored historical baseline characteristics. Specifically, this includes: The frequency features are input into a pre-trained convolutional neural network model, which has learned the correspondence between the frequency features represented by the historical baseline features and the residual stiffness. The convolutional neural network model outputs the residual stiffness value of the support structure corresponding to the storage location, and simultaneously outputs the damage type identification result and the corresponding confidence level. The damage type includes one or more of the following: local buckling, loosening of connection nodes, column bending deformation, and foundation settlement.

3. The intelligent warehousing location allocation method according to claim 1, characterized in that, In step S2, the vibration response signal is normalized to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal. Specifically, this includes: Obtain the weight of the goods and the fork movement speed of the handling equipment corresponding to this storage and retrieval operation; Based on the preset weight-speed-impact force mapping relationship, determine the theoretical impact force reference value corresponding to the weight of the goods and the speed of the fork movement; Dividing the amplitude of the collected vibration response signal by the theoretical impact force reference value yields the normalized vibration response signal waveform, thus providing a comparable amplitude benchmark for the waveforms collected under different cargo weights and handling speeds.

4. The intelligent warehousing location allocation method according to claim 1, characterized in that, Step S3, which determines the load-bearing capacity of the storage location based on the remaining stiffness information, specifically includes: The obtained remaining stiffness information is compared with multiple pre-stored stiffness threshold intervals; If the remaining stiffness falls within the first threshold range, the storage location is determined to be of a high load-bearing capacity, allowing the storage of any goods not exceeding the rated load. If the remaining stiffness falls within the second threshold range, the storage location is determined to be of medium load-bearing capacity, and only goods whose weight does not exceed the first percentage threshold of the rated load are allowed to be stored. If the remaining stiffness falls within the third threshold range, the storage location is determined to be of low load-bearing capacity, and only goods weighing no more than the second percentage threshold of the rated load are allowed to be stored, and a maintenance prompt is triggered. If the remaining stiffness falls within the fourth threshold range, the storage location is determined to be a no-load level, prohibiting the storage of any goods, and an emergency maintenance work order is triggered. The first threshold interval, the second threshold interval, the third threshold interval, and the fourth threshold interval are divided in descending order of remaining stiffness, and the endpoint values ​​of each interval are determined according to a preset ratio based on the initial remaining stiffness of the storage location.

5. The intelligent warehousing location allocation method according to claim 1, characterized in that, In step S1, the start time of the impact excitation is triggered in the following manner: An acceleration trigger sensor is installed on the handling equipment. When the forks of the handling equipment come into contact with the shelves or pallets of the storage location, the acceleration trigger sensor generates a trigger signal. The vibration sensing unit switches from a low-power standby state to a high-speed acquisition state according to the trigger signal, and acquires the vibration response signal at a first sampling rate higher than the normal sampling rate. The acquisition time covers at least a first preset time before the impact occurs to a second preset time after the impact occurs.

6. A smart warehousing location allocation system, characterized in that, The system includes: a signal acquisition module, a signal processing module, a storage location capacity confirmation module, and a storage location allocation module; The signal acquisition module is used to acquire the vibration response signal of the storage location in response to the impact excitation generated by the impact contact between the handling equipment and the corresponding storage location when the handling equipment performs storage and retrieval operations, through a vibration sensing unit deployed on the rack structure and corresponding to the storage location. The signal processing module is used to perform amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal; extract the power spectral density function and / or frequency response function of the vibration response signal, and extract frequency characteristic parameters including at least the dominant frequency value and damping ratio from the power spectral density function or frequency response function to obtain the frequency characteristics of the support structure corresponding to the warehouse location in the current state; The storage location load-bearing capacity confirmation module is used to obtain the remaining stiffness information of the supporting structure corresponding to the storage location based on the comparison result between the frequency characteristics and the pre-stored historical baseline characteristic information; and to determine the load-bearing capacity level of the storage location based on the remaining stiffness information; wherein, the historical baseline characteristic information is used to characterize the correspondence between the frequency characteristics and the remaining stiffness of the supporting structure, and the load-bearing capacity level is associated with the upper limit of the weight of goods that can be stored. The storage location allocation module is used to obtain the weight information of the goods to be stored and the carrying capacity level of each storage location in the warehousing environment; and to allocate storage locations for the goods to be stored according to the weight information of the goods to be stored and the carrying capacity level of each storage location, so that the total weight of the goods in each storage location after allocation does not exceed the weight limit corresponding to its carrying capacity level.

7. The intelligent warehousing location allocation system according to claim 6, characterized in that, When the reservoir location bearing capacity confirmation module performs a comparison based on the frequency characteristics and pre-stored historical baseline characteristics to obtain the remaining stiffness information of the support structure corresponding to the reservoir location, it specifically includes: The frequency features are input into a pre-trained convolutional neural network model, which has learned the correspondence between the frequency features represented by the historical baseline features and the residual stiffness. The convolutional neural network model outputs the residual stiffness value of the support structure corresponding to the storage location, and simultaneously outputs the damage type identification result and the corresponding confidence level. The damage type includes one or more of the following: local buckling, loosening of connection nodes, column bending deformation, and foundation settlement.

8. The intelligent warehousing location allocation system according to claim 6, characterized in that, The signal processing module specifically performs amplitude normalization processing on the vibration response signal to eliminate the influence of different cargo weights and handling speeds on the waveform amplitude of the vibration response signal, including: Obtain the weight of the goods and the fork movement speed of the handling equipment corresponding to this storage and retrieval operation; Based on the preset weight-speed-impact force mapping relationship, determine the theoretical impact force reference value corresponding to the weight of the goods and the speed of the fork movement; Dividing the amplitude of the collected vibration response signal by the theoretical impact force reference value yields the normalized vibration response signal waveform, thus providing a comparable amplitude benchmark for the waveforms collected under different cargo weights and handling speeds.

9. The intelligent warehousing location allocation system according to claim 6, characterized in that, The storage location bearing capacity confirmation module determines the bearing capacity level of the storage location based on the remaining stiffness information, specifically including: The obtained remaining stiffness information is compared with multiple pre-stored stiffness threshold intervals; If the remaining stiffness falls within the first threshold range, the storage location is determined to be of a high load-bearing capacity, allowing the storage of any goods not exceeding the rated load. If the remaining stiffness falls within the second threshold range, the storage location is determined to be of medium load-bearing capacity, and only goods whose weight does not exceed the first percentage threshold of the rated load are allowed to be stored. If the remaining stiffness falls within the third threshold range, the storage location is determined to be of low load-bearing capacity, and only goods weighing no more than the second percentage threshold of the rated load are allowed to be stored, and a maintenance prompt is triggered. If the remaining stiffness falls within the fourth threshold range, the storage location is determined to be a no-load level, prohibiting the storage of any goods, and an emergency maintenance work order is triggered. The first threshold interval, the second threshold interval, the third threshold interval, and the fourth threshold interval are divided in descending order of remaining stiffness, and the endpoint values ​​of each interval are determined according to a preset ratio based on the initial remaining stiffness of the storage location.

10. The intelligent warehousing location allocation system according to claim 6, characterized in that, The system also includes a signal acquisition triggering module, which, when the signal acquisition module is working, is specifically used for: When the forks of the handling equipment come into contact with the shelves or pallets of the storage location, an acceleration trigger sensor installed on the handling equipment generates a trigger signal; the vibration sensing unit switches from a low-power standby state to a high-speed acquisition state according to the trigger signal, and acquires the vibration response signal at a first sampling rate higher than the normal sampling rate, with the acquisition time covering at least a first preset time before the impact to a second preset time after the impact.

Images