A method, system, and storage medium for self-balancing loads in mobile shelving units.

Abstract

This invention provides a method, system, and storage medium for self-balancing of mobile shelving loads, relating to the field of intelligent warehousing equipment technology. The method includes: acquiring multi-source real-time data of the mobile shelving, including time-series data, spatial stress data, and environmental parameter data; using an LSTM model to predict the future load data of the mobile shelving within a preset future time period based on the time-series data, spatial stress data, and environmental parameter data; generating an adjustment strategy for the mobile shelving based on the future load data; and controlling the mobile shelving to self-balance according to the adjustment strategy. This invention optimizes the self-balancing control of the support arms by real-time monitoring, accurate prediction, and dynamic adjustment of the mobile shelving load, thereby significantly improving the storage efficiency of the mobile shelving.

Description

Technical Field

[0001] This invention relates to the field of intelligent warehousing equipment technology, and more specifically, to a method, system, and storage medium for self-balancing loads on mobile shelving units. Background Technology

[0002] As a core storage device, the load management of intelligent mobile shelving directly affects storage security and space utilization. To ensure that the mobile shelving can achieve a dynamic balance between storage density and shelving load, it is necessary to adjust the spacing between the shelves in real time according to the load conditions.

[0003] In related technologies, discretely arranged pressure sensors are typically used to monitor the load on the mobile shelving unit, and the spacing between shelves is adjusted manually according to preset rules based on the load. However, since pressure sensors cannot provide complete coverage of the mobile shelving unit, the load detection is incomplete. Furthermore, over-reliance on preset rules introduces delays in adjustments, ultimately making it difficult for the mobile shelving unit to handle sudden load changes. This can cause the support arms of the mobile shelving unit to vibrate, seriously affecting its safety. Summary of the Invention

[0004] The problem addressed by this invention is how to improve the storage efficiency of mobile shelving units.

[0005] To address the aforementioned problems, this invention provides a method, system, and storage medium for self-balancing loads in mobile shelving units.

[0006] In a first aspect, the present invention provides a method for self-balancing the load of mobile shelving units, comprising:

[0007] Acquire multi-source real-time data of the mobile shelving unit, including time-series data, spatial stress data, and environmental parameter data of the mobile shelving unit.

[0008] Using an LSTM model, the future load data of the mobile shelving unit within a preset future time period is predicted based on the time series data, the spatial stress data, and the environmental parameter data.

[0009] Based on the future load data, an adjustment strategy for the mobile shelving unit is generated;

[0010] According to the adjustment strategy, the mobile shelving unit is controlled to self-balance.

[0011] Optionally, acquiring multi-source real-time data of the mobile shelving unit includes:

[0012] From the database of the mobile shelving unit, obtain the file access timestamp, operation duration, and operation interval period of the mobile shelving unit within a preset historical time period, and use the file access timestamp, operation duration, and operation interval period as the time series data;

[0013] By installing stress sensors on the mobile shelving unit, the mobile shelving unit is monitored in real time to obtain the strain value and strain gradient tensor of the mobile shelving unit, and the strain value and strain gradient tensor are used as the spatial stress data.

[0014] The temperature, humidity, vibration acceleration, and light intensity of the environment in which the mobile shelving unit is located are obtained through environmental sensors, and the temperature, humidity, vibration acceleration, and light intensity are used as environmental parameter data.

[0015] Optionally, the step of using an LSTM model to predict the future load data of the mobile shelving unit within a preset future time period based on the time series data, the spatial stress data, and the environmental parameter data includes:

[0016] The time series data, spatial stress data, and environmental parameter data are aligned according to a preset time base using interpolation. The aligned time series data, spatial stress data, and environmental parameter data are then subjected to feature engineering and normalization processes to obtain the input data for the LSTM model.

[0017] The input data is fed into the LSTM model for prediction to obtain the load probability heatmap, deformation risk index and adjustment priority sequence of the compact shelving within the preset future time period.

[0018] The load probability heatmap, the deformation risk index, and the adjustment priority sequence are used as the future load data.

[0019] Optionally, the step of inputting the input data into the LSTM model for prediction to obtain the load probability heatmap, deformation risk index, and adjustment priority sequence of the compact shelving within the preset future time period includes:

[0020] The input data is fed into the LSTM model, and features are extracted through the hidden layers of the LSTM model to obtain spatiotemporal features;

[0021] Attention is allocated to the spatiotemporal features through the attention mechanism layer of the LSTM model to obtain key spatiotemporal node features;

[0022] The LSTM model's fusion layer performs dimensionality reduction and feature fusion on the key spatiotemporal node features to obtain the load probability heatmap, deformation risk index, and adjustment priority sequence for the preset future time period.

[0023] Optionally, generating an adjustment strategy for the mobile shelving unit based on the future load data includes:

[0024] The adjustment path of the support arm of the mobile shelving unit is planned based on the load probability heat map to obtain the future path plan of the support arm;

[0025] The support mode of the support arm is determined based on the deformation risk index.

[0026] The adjustment order of the frame is determined according to the adjustment priority sequence;

[0027] The adjustment strategy for the mobile shelving unit is generated based on the support mode, the adjustment sequence, and the future path planning of the support arm corresponding to each shelving unit.

[0028] Optionally, the step of planning the adjustment path of the support arm of the mobile shelving unit based on the load probability heatmap to obtain the future path planning of the support arm includes:

[0029] Based on the load probability heatmap, the high load probability area of ​​the mobile shelving unit is determined;

[0030] Based on the high-load probability area, determine the target position of the frame of the mobile shelving unit;

[0031] The target position of the support arm is obtained by combining the path planning algorithm with the target position and the constraint conditions of the frame corresponding to the support arm.

[0032] The future path planning is obtained based on the initial position and the target position of the support arm.

[0033] Optionally, determining the support mode of the support arm based on the deformation risk index includes:

[0034] The support mode is determined based on the relationship between the deformation risk index and the preset risk threshold;

[0035] Wherein, when the deformation risk index is greater than or equal to the preset risk threshold, the support mode is set to rigid mode;

[0036] When the deformation risk index is less than the preset risk threshold, the support mode is set to compliant mode, and the damping coefficient of the mobile shelving is set according to preset requirements.

[0037] Optionally, controlling the mobile shelving to self-balance according to the adjustment strategy includes:

[0038] The mobile shelving unit is switched according to the support mode.

[0039] The frame is controlled to adjust the angle of the corresponding support arms according to the adjustment sequence and the future path planning until all the support arms of the mobile shelving unit reach the target position.

[0040] Secondly, the present invention provides a self-balancing load system for mobile shelving units, comprising:

[0041] The monitoring unit is used to acquire multi-source real-time data of the mobile shelving unit, including time-series data, spatial stress data, and environmental parameter data of the mobile shelving unit.

[0042] The prediction unit is used to predict, based on the time series data, the spatial stress data, and the environmental parameter data, the future load data of the mobile shelving unit within a preset future time period using an LSTM model.

[0043] An analysis unit is used to generate an adjustment strategy for the mobile shelving unit based on the future load data.

[0044] A control unit is used to control the mobile shelving unit to perform self-balancing according to the adjustment strategy.

[0045] Thirdly, the present invention provides a computer-readable storage medium, including a computer-readable storage medium storing a computer program and a processor, wherein the computer program is read and executed by the processor to implement the self-balancing load method of the mobile shelving unit as described above.

[0046] The mobile shelving load self-balancing method, system, and storage medium of this invention achieve comprehensive monitoring of the mobile shelving's status by acquiring multi-source real-time data, including time-series data, spatial stress data, and environmental parameter data. This provides accurate data support for subsequent prediction and adjustment, enabling the system to promptly perceive the mobile shelving's status and providing a data foundation for improving its storage performance. It ensures that adjustment strategies are based on accurate information. Furthermore, the LSTM model is used to analyze and predict the multi-source real-time data, obtaining future load data for the mobile shelving within a preset future time period. This allows for the early determination of the future load conditions and potential risks, transforming traditional passive response into proactive prevention. This provides a crucial basis for generating scientifically sound adjustment strategies and is a core element in improving storage performance, effectively preventing structural problems caused by abnormal loads. Finally, adjustment strategies are generated based on the future load data, and the mobile shelving is controlled to self-balance accordingly, achieving dynamic and refined management of the mobile shelving and ensuring its reliable operation under various working conditions. According to the adjustment strategy, the linear motor driven by the electronic control system adjusts the angle of the support arm, enabling the mobile shelving to self-balance. This achieves dynamic balance adjustment of the mobile shelving, enhancing its stability and safety, improving space utilization, reducing the need for safety redundancy space, ensuring the safety and reliability of stored items, effectively improving the stability and safety of the mobile shelving, preventing structural damage caused by uneven load, ensuring the safety and reliability of stored items, increasing the storage density and space value of the mobile shelving, achieving a virtuous balance between storage efficiency and safety, and effectively improving the overall storage effect of the mobile shelving. Attached Figure Description

[0047] Figure 1 This is a flowchart of the self-balancing load method for mobile shelving units according to an embodiment of the present invention;

[0048] Figure 2 This is a structural block diagram of the self-balancing load system for mobile shelving units according to an embodiment of the present invention. Detailed Implementation

[0049] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0050] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0051] As used herein, the term "comprising" and its variations are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; and the term "optionally" means "optional embodiments". Definitions for other terms will be given in the description below.

[0052] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0053] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0054] Combination Figure 1 As shown, the present invention provides a method for self-balancing the load of mobile shelving units, comprising:

[0055] Acquire multi-source real-time data of the mobile shelving unit, including time-series data, spatial stress data, and environmental parameter data of the mobile shelving unit.

[0056] Specifically, by collecting time-series data of the mobile shelving (such as file access timestamps, operation duration, and operation intervals), spatial stress data (strain values ​​and strain gradient tensors obtained with the help of stress sensors), and environmental parameter data (including temperature, humidity, vibration acceleration, and light intensity), the system achieves comprehensive and multi-dimensional real-time status monitoring of the mobile shelving. This provides a detailed and accurate data foundation for subsequent load prediction and adjustment strategy generation, enabling the system to perceive the current status of the mobile shelving in a timely and accurate manner. This ensures that the formulated adjustment strategies are based on the latest and most reliable information, thus laying a solid data foundation for improving the storage performance of the mobile shelving.

[0057] By using an LSTM model, based on the time series data, the spatial stress data, and the environmental parameter data, the future load data of the mobile shelving unit within a preset future time period is predicted.

[0058] Specifically, based on collected multi-source real-time data, the powerful time-series data processing and prediction capabilities of the LSTM model are utilized to accurately predict the load situation of the mobile shelving within a preset future time period, obtaining future load data such as load probability heatmaps, deformation risk indices, and adjustment priority sequences. The LSTM model transforms traditional passive response management into proactive preventative management, enabling the system to anticipate upcoming load change trends and potential risks for the mobile shelving. This method, which allows for advance planning of countermeasures, provides a crucial basis for generating scientifically sound and forward-looking adjustment strategies. It is the core element of the intelligent prediction process, effectively preventing structural problems caused by abnormal load changes, thereby improving the stability and storage security of the mobile shelving.

[0059] An adjustment strategy for the mobile shelving unit is generated based on the future load data.

[0060] Specifically, based on future load data predicted by the LSTM model, including load probability heatmaps, deformation risk indices, and adjustment priority sequences, an adjustment strategy matching the future load conditions of the mobile shelving unit is generated. By rationally planning the adjustment path of the support arms, determining the support mode, and clarifying the adjustment sequence of the shelving unit, it is ensured that the mobile shelving unit can make accurate and effective adjustments before load changes. This dynamic adjustment strategy generation based on predictive data not only improves the response speed and adaptability of the mobile shelving unit to load changes, but also realizes refined and dynamic management of the mobile shelving unit. It is an important link connecting the prediction and actual control links, directly related to the stable operation and reliability of the mobile shelving unit under various working conditions, and provides clear and scientific guidance for subsequent actual control operations.

[0061] According to the adjustment strategy, the mobile shelving unit is controlled to self-balance.

[0062] Specifically, by using an electronic control system to drive a linear motor to adjust the angle of the support arms, the mobile shelving unit achieves self-balancing adjustment, putting into practice the adjustment strategies developed in previous steps based on data monitoring and intelligent prediction. Through this real-time, precise dynamic balancing adjustment, the mobile shelving unit effectively enhances its stability and safety, reduces the risk of mechanical wear and structural damage caused by uneven load distribution, and extends the equipment's lifespan. Simultaneously, reasonable self-balancing adjustment also helps improve the space utilization rate of the mobile shelving unit, reduces unnecessary safety redundancy space reservations, and increases storage density and space value while ensuring the safety and reliability of stored items. This achieves a virtuous cycle of storage efficiency and safety, significantly improving the overall storage effect and performance of the mobile shelving unit.

[0063] The self-balancing load method for mobile shelving units of this invention achieves comprehensive monitoring of the shelving unit's status by acquiring multi-source real-time data, including time-series data, spatial stress data, and environmental parameter data. This provides accurate data support for subsequent prediction and adjustment, enabling the system to promptly perceive the shelving unit's status and providing a data foundation for improving storage performance. It ensures that adjustment strategies are based on accurate information. Furthermore, the LSTM model is used to analyze and predict the multi-source real-time data, obtaining future load data for the shelving unit within a preset future time period. This allows for the early determination of the shelving unit's future load conditions and potential risks, transforming traditional passive response into proactive prevention. This provides a crucial basis for generating scientifically sound adjustment strategies and is a core element in improving storage performance, effectively preventing structural problems caused by abnormal loads. Finally, adjustment strategies are generated based on the future load data, and the shelving unit is controlled to self-balance accordingly, achieving dynamic and refined management of the shelving unit and ensuring its reliable operation under various working conditions. According to the adjustment strategy, the linear motor driven by the electronic control system adjusts the angle of the support arm, enabling the mobile shelving to self-balance. This achieves dynamic balance adjustment of the mobile shelving, enhancing its stability and safety, improving space utilization, reducing the need for safety redundancy space, ensuring the safety and reliability of stored items, effectively improving the stability and safety of the mobile shelving, preventing structural damage caused by uneven load, ensuring the safety and reliability of stored items, increasing the storage density and space value of the mobile shelving, achieving a virtuous balance between storage efficiency and safety, and effectively improving the overall storage effect of the mobile shelving.

[0064] Optionally, acquiring multi-source real-time data of the mobile shelving unit includes:

[0065] From the database of the mobile shelving unit, obtain the file access timestamp, operation duration, and operation interval period of the mobile shelving unit within a preset historical time period, and use the file access timestamp, operation duration, and operation interval period as the time series data;

[0066] By installing stress sensors on the mobile shelving unit, the mobile shelving unit is monitored in real time to obtain the strain value and strain gradient tensor of the mobile shelving unit, and the strain value and strain gradient tensor are used as the spatial stress data.

[0067] The temperature, humidity, vibration acceleration, and light intensity of the environment in which the mobile shelving unit is located are obtained through environmental sensors, and the temperature, humidity, vibration acceleration, and light intensity are used as environmental parameter data.

[0068] Specifically, the system retrieves file access timestamps, operation durations, and operation intervals within a preset historical time period from the mobile shelving's database, treating this as time-series data. For example, operation durations and intervals are arranged sequentially according to file storage timestamps. This data reflects the dynamic changes in file access. Simultaneously, stress sensors mounted on the mobile shelving monitor the strain values ​​and strain gradient tensors in real time, acquiring spatial stress data. The stress sensors have a sampling frequency of 100Hz and a spatial resolution of 10cm×10cm grid, accurately reflecting the stress distribution within the shelving. Spatial stress data from multiple different times can be correlated with file storage timestamps. Furthermore, environmental sensors acquire environmental parameters such as temperature, humidity, vibration acceleration, and light intensity. This data helps assess the impact of the external environment on the mobile shelving's load. Environmental parameter data can be collected in real time using MEMS sensors, and environmental parameter data from multiple different times can be correlated with file storage timestamps. This comprehensive monitoring of the mobile shelving's status provides precise data support for subsequent prediction and adjustments.

[0069] In a preferred embodiment of the present invention, the time window of the time series data in the input data of the LSTM model is the past 24 hours, and the sampling frequency is 1Hz, which includes the operation duration and operation interval period during file access. Spatial stress data is then collected through 16 FBG sensors, with strain values ​​of [50, 55, 60,..., 100]με, to simulate the strain values ​​of the 16 sensors at a certain moment. The strain gradient tensor is... , The technical specifications for spatial profitability data can be a 10cm×10cm spatial resolution grid. For environmental parameters, the temperature and humidity are 25℃ and 40%RH, the vibration acceleration is 0.5g, and the light intensity is 300 lux. These data can be acquired in real time using MEMS sensors.

[0070] In this optional embodiment, by acquiring time-series data such as file access timestamps, operation durations, and operation intervals, the dynamic patterns of file access can be understood, providing a time-dimensional basis for predicting future load changes. Real-time monitoring of spatial stress data such as strain values ​​and strain gradient tensors allows for precise understanding of the stress distribution within the shelving unit, enabling the timely identification of potential stress concentration areas. Simultaneously, acquiring environmental parameter data such as temperature, humidity, vibration acceleration, and light intensity helps assess the impact of the external environment on the mobile shelving unit, further improving prediction accuracy. This comprehensive data collection ensures that the system can promptly and accurately perceive the real-time status of the mobile shelving unit, enabling subsequent adjustment strategies to be formulated based on the latest and most reliable information, thereby effectively improving the storage performance, stability, and security of the mobile shelving unit.

[0071] Optionally, the step of using an LSTM model to predict the future load data of the mobile shelving unit within a preset future time period based on the time series data, the spatial stress data, and the environmental parameter data includes:

[0072] The time series data, spatial stress data, and environmental parameter data are aligned according to a preset time base using interpolation. The aligned time series data, spatial stress data, and environmental parameter data are then subjected to feature engineering and normalization processes to obtain the input data for the LSTM model.

[0073] The input data is fed into the LSTM model for prediction to obtain the load probability heatmap, deformation risk index and adjustment priority sequence of the compact shelving within the preset future time period.

[0074] The load probability heatmap, the deformation risk index, and the adjustment priority sequence are used as the future load data.

[0075] Specifically, firstly, interpolation is used to align the time-series data, spatial stress data, and environmental parameter data to ensure all data have a unified time reference, with errors controlled within 10ms, providing a synchronized data foundation for subsequent analysis. Next, feature engineering and normalization are performed on the aligned data, extracting features such as the low-frequency component of the Fourier transform of the operation frequency (0-0.1Hz), and the first and second moments (distribution dispersion) of the strain field. The data is then normalized to obtain a data format suitable for LSTM model input. Afterward, the processed data is input into a dual-channel LSTM model for prediction. The time-series channel processes dynamic features, while the spatial channel processes real-time stress distribution data. An attention mechanism is used to focus on key nodes, ultimately outputting a load probability heatmap for the next 15 minutes (resolution 10cm×10cm, probability value 0-1), a deformation risk index (scalar value 0-1, 0.3 as the warning threshold), and an adjustment priority sequence (frame unit numbers arranged in descending order of risk level). For example, the model can predict the load probability distribution of each area of ​​the mobile shelving unit in the future, identify high-risk areas and priority adjustment units.

[0076] In a preferred embodiment of the invention, normalization can be achieved by first time-aligning the time series data, spatial stress data, and environmental parameter data using interpolation, ensuring that all data sources are on a unified time reference with errors controlled within less than 10 milliseconds, thus achieving data synchronization. This process is crucial for integrating information from different sensors and data sources, allowing the model to compare and analyze different types of data at the same point in time. After alignment, feature engineering is performed on these data to extract key features such as the low-frequency components (0-0.1Hz) of the Fourier transform of the operation frequency and the first and second moments of the strain field. These features help reveal the patterns and trends behind the data. Subsequently, these features are normalized using the following formula.

[0077] ;

[0078] Using the formula above, x represents the value of the original data point. The normalized data points are scaled to the interval [-1, 1], where μ represents the mean and σ represents the standard deviation. Normalization ensures that different features are on the same scale, which helps improve the training efficiency and prediction accuracy of the LSTM model. After these steps, the resulting data will be used as input to the LSTM model to predict the future load of the mobile shelving unit.

[0079] In another preferred embodiment of the invention, the temporal dynamic features include the moving average of the operation frequency over the past hour and the load change rate over adjacent 15 minutes, used to reflect the trend of storage and retrieval operation intensity, with weights of 0.18 and 0.15, respectively. Spatial stress data includes the ratio of the maximum principal stress to the material yield strength and the area ratio of high strain gradient regions, used to assess the risk of plastic deformation and identify local stress concentrations, with weights of 0.22 and 0.19, respectively. Environmental parameter data includes the temperature-humidity-stress correlation coefficient and the cumulative value of vibration energy, used to quantify the impact of the environment on the mechanical properties of materials and to determine the impact of external disturbances on structural stability, with weights of 0.12 and 0.08, respectively. Time series data includes the time-delay cross-correlation between operation frequency and strain change, used to reveal the relationship between human operation and structural response, with a weight of 0.06. The above data together constitute the input of the LSTM model, used to predict the future load of the mobile shelving unit to achieve self-balancing control.

[0080] In this optional embodiment, data alignment and preprocessing transform multi-source heterogeneous data into a model-recognizable format, ensuring data quality and consistency and laying the foundation for accurate prediction. A dual-channel LSTM model combined with an attention mechanism fully mines the spatiotemporal features and key information in the data, effectively improving prediction accuracy and reliability. Pre-obtained load probability heatmaps, deformation risk indices, and adjustment priority sequences, among other future load data, enable the system to anticipate load change trends and potential risks in the mobile shelving, transforming traditional passive response into proactive prevention. This provides crucial information for generating scientifically sound adjustment strategies, effectively improving the storage performance of the mobile shelving, enhancing its stability and security, reducing structural problems caused by abnormal loads, optimizing space utilization, and achieving a healthy balance between storage efficiency and security.

[0081] Optionally, the step of inputting the input data into the LSTM model for prediction to obtain the load probability heatmap, deformation risk index, and adjustment priority sequence of the compact shelving within the preset future time period includes:

[0082] The input data is fed into the LSTM model, and features are extracted through the hidden layers of the LSTM model to obtain spatiotemporal features;

[0083] Attention is allocated to the spatiotemporal features through the attention mechanism layer of the LSTM model to obtain key spatiotemporal node features;

[0084] The LSTM model's fusion layer performs dimensionality reduction and feature fusion on the key spatiotemporal node features to obtain the load probability heatmap, deformation risk index, and adjustment priority sequence for the preset future time period.

[0085] Specifically, this step inputs the processed input data into the LSTM model. Feature extraction from the hidden layers allows for in-depth mining of spatiotemporal features within the data, capturing dynamic trends in load changes and stress distribution patterns. The attention mechanism layer focuses on key spatiotemporal node features, increasing the model's attention to important information and enhancing prediction accuracy. Dimensionality reduction and feature fusion in the fusion layer further optimize feature representation, removing redundancy and highlighting key information. The resulting load probability heatmap, deformation risk index, and adjustment priority sequence can reveal potential future load risks and areas requiring focused attention for the mobile shelving. In a preferred embodiment of the invention, feature extraction is performed on the data through the model's hidden layers (2 layers × 128 units for the temporal channel, 1 layer × 64 units for the spatial channel), yielding spatiotemporal features containing time-series trends and spatial stress distribution. These features reflect the dynamic patterns of load changes and structural stress patterns in the mobile shelving. Next, the attention mechanism layer, a four-head attention mechanism, allocates attention to the spatiotemporal features, identifying and highlighting key spatiotemporal node features, such as time periods of drastic future load changes or spatial areas of stress concentration. Finally, through a fusion layer of 256, 128, and 64 dimensions, i.e., a fully connected network, the key spatiotemporal node features are dimensionality reduced and features are fused to remove redundant information and enhance the expressive power and discriminative power of the features. This results in a load probability heatmap for a preset future time period, with a resolution of 10cm×10cm, probability values ​​of 0-1 floating-point type, deformation risk index of scalar value 0-1, and an adjustment priority sequence, providing accurate predictive information for the generation of subsequent adjustment strategies.

[0086] In this optional embodiment, by accurately predicting the future load of the mobile shelving, structural problems caused by abnormal loads can be effectively prevented, the stability and safety of the mobile shelving can be improved, mechanical wear can be reduced, service life can be extended, and space utilization can be optimized to achieve a good balance between storage efficiency and safety.

[0087] Optionally, generating an adjustment strategy for the mobile shelving unit based on the future load data includes:

[0088] The adjustment path of the support arm of the mobile shelving unit is planned based on the load probability heat map to obtain the future path plan of the support arm;

[0089] The support mode of the support arm is determined based on the deformation risk index.

[0090] The adjustment order of the frame is determined according to the adjustment priority sequence;

[0091] The adjustment strategy for the mobile shelving unit is generated based on the support mode, the adjustment sequence, and the future path planning of the support arm corresponding to each shelving unit.

[0092] Specifically, firstly, the adjustment path of the mobile shelving unit's support arms is planned based on the load probability heatmap. Based on the grid cells of the load probability heatmap and the load probability of the corresponding time slice, high-risk areas are identified, and future paths are planned for the support arms to ensure they move to areas where future loads will increase to provide better support. Secondly, the support mode is determined based on the deformation risk index. In a preferred embodiment of the invention, the support mode for adjusting the support arms can be determined according to the magnitude of the deformation risk index. Finally, the shelving unit adjustment sequence is determined based on the adjustment priority sequence, and the shelving unit numbers can be arranged in descending order of risk level according to the adjustment priority sequence, thereby providing a clear adjustment priority order for the actuator. Combining this information, the system generates a mobile shelving unit adjustment strategy that includes support modes, adjustment sequences, and future path planning to achieve precise and dynamic load self-balancing.

[0093] In this optional embodiment, the adjustment path of the support arm is planned in advance using a load probability heatmap, enabling the support arm to be moved to areas where the load will increase in the future, providing timely and effective support and preventing structural damage caused by uneven load. The support mode is determined based on the deformation risk index, allowing a compliant mode to be used to reduce energy consumption in low-risk situations and a rigid mode to ensure structural stability in high-risk situations, achieving a balance between energy saving and safety. The adjustment sequence is determined according to the adjustment priority sequence, ensuring that high-risk areas are addressed first, improving adjustment efficiency and avoiding resource waste. In summary, this step, by generating a scientific and reasonable adjustment strategy, effectively improves the stability and safety of the mobile shelving, reduces mechanical wear, extends service life, optimizes space utilization, reduces the need for safety redundancy space, achieves a virtuous cycle of storage efficiency and safety, and significantly improves the overall storage effect of the mobile shelving.

[0094] Optionally, the step of planning the adjustment path of the support arm of the mobile shelving unit based on the load probability heatmap to obtain the future path planning of the support arm includes:

[0095] Based on the load probability heatmap, the high load probability area of ​​the mobile shelving unit is determined;

[0096] Based on the high-load probability area, determine the target position of the frame of the mobile shelving unit;

[0097] The target position of the support arm is obtained by combining the path planning algorithm with the target position and the constraint conditions of the frame corresponding to the support arm.

[0098] The future path planning is obtained based on the initial position and the target position of the support arm.

[0099] Specifically, firstly, high-load-probability areas of the mobile shelving unit are identified based on a load probability heatmap. For example, the shelving unit is divided into 10cm × 10cm grid cells based on the load probability heatmap, and a load probability (0-1) is provided for each cell in the corresponding time slice. By analyzing these probability values, areas where the future load may increase can be located. Next, these high-load-probability areas are determined as target positions for the shelving unit so that the support arms can provide more effective support. Then, using path planning algorithms, such as the A* algorithm or the artificial potential field method, combined with the target position and the constraints of the support arms (such as physical structural limitations, kinematic parameters, etc.), the target position of the support arms is calculated. The support arms need to reach the target position within a specified time with an allowed speed and acceleration, while avoiding collisions with other components of the mobile shelving unit. Finally, based on the initial position of the support arms and the calculated target position, a future path plan for the support arms is generated to ensure that the support arms can move to the target position safely and efficiently.

[0100] In this optional embodiment, by identifying high-load probability areas and planning the adjustment path of the support arms, the system can move the support arms to the areas requiring reinforced support before the actual load increases, thereby effectively avoiding structural damage caused by uneven load and enhancing the stability and safety of the mobile shelving unit. Advance path planning also reduces mechanical wear and extends the service life of the mobile shelving unit.

[0101] Optionally, determining the support mode of the support arm based on the deformation risk index includes:

[0102] The support mode is determined based on the relationship between the deformation risk index and the preset risk threshold;

[0103] Wherein, when the deformation risk index is greater than or equal to the preset risk threshold, the support mode is set to rigid mode;

[0104] When the deformation risk index is less than the preset risk threshold, the support mode is set to compliant mode, and the damping coefficient of the mobile shelving is set according to preset requirements.

[0105] Specifically, the deformation risk index is first compared with a preset risk threshold to determine the support mode of the support arm. The deformation risk index is calculated based on the predicted maximum strain over the next 15 minutes, the current real-time maximum strain, and the environmental coupling coefficient. When the deformation risk index is greater than or equal to the preset risk threshold, the mobile shelving unit is in a high-risk state. In this case, the support mode is set to rigid mode, i.e., the electromagnetic clutch locks the support arm, the magnetorheological damper increases viscosity, and the vibration suppression rate of the shelving unit increases to enhance structural stability and prevent structural damage caused by load changes. Conversely, when the deformation risk index is less than the preset risk threshold, the support mode is set to compliant mode. In this case, the support arm is fine-tuned at a preset speed, the magnetorheological damper maintains a low viscosity state to reduce energy consumption and maintain structural flexibility. Simultaneously, the damping coefficient of the mobile shelving unit is set according to preset requirements to ensure a balance between energy saving and stability in a low-risk state.

[0106] In a preferred embodiment of the present invention, assuming the current preset risk threshold is 0.35, exceeding the preset risk threshold of 0.3, the system marks it as high risk. The rate of change of the preset risk threshold over the last 30 seconds is calculated, finding Δshort = 0.02 / s, exceeding the preset risk threshold of 0.01 / s, triggering a rapid response protocol. Analysis of the predictive heatmap reveals that the preset risk threshold will continue to exceed 0.28 within the next 5 minutes. The current relative humidity is 30%RH, exceeding 20%RH; therefore, the threshold is increased by 5%, and the new preset risk threshold is adjusted to 0.33. The vibration energy Ev is detected at 0.3 g²·s, below the threshold of 0.5 g²·s, therefore no adjustment to the preset risk threshold is needed.

[0107] Ultimately, the support mode and corresponding actions of the support arm are determined based on a combination of the Deformation Risk Index (DRI) and the Time Duration Factor (TPF). When the DRI value exceeds 0.3 and the TPF is greater than 0.25, the system determines a high risk, immediately switches to rigid mode, and triggers an audible and visual alarm to ensure the stability of the mobile shelving structure. If the DRI value exceeds 0.3 but the TPF is less than or equal to 0.25, the system determines a transient high risk, at which point the compliant mode is enhanced, increasing the damping coefficient by 50% to improve stability. When the DRI value is less than or equal to 0.3 but the rate of change of DRI Δshort is greater than 0.01 / s, the system considers the risk to be rising, pre-tightening the electromagnetic clutch in preparation for locking at any time to deal with potential risks. In all other cases, the system determines a low risk, maintaining the compliant mode to ensure the normal operation of the mobile shelving. This condition-based decision-making process ensures that the mobile shelving can adopt the most appropriate support mode under different risk levels, thereby improving the safety and reliability of the entire system.

[0108] In this optional embodiment, the support mode is flexibly adjusted according to the actual risk status of the mobile shelving unit, achieving a balance between energy saving and safety. Switching to rigid mode under high-risk conditions significantly enhances the structural stability of the mobile shelving unit, effectively preventing structural damage caused by sudden load changes or stress concentration, and ensuring the safety and reliability of stored items. Maintaining compliant mode under low-risk conditions meets the basic structural support requirements while reducing energy consumption and extending the service life of mechanical components.

[0109] Optionally, controlling the mobile shelving to self-balance according to the adjustment strategy includes:

[0110] The mobile shelving unit is switched according to the support mode.

[0111] The frame is controlled to adjust the angle of the corresponding support arms according to the adjustment sequence and the future path planning until all the support arms of the mobile shelving unit reach the target position.

[0112] Specifically, switching the support mode of the mobile shelving unit according to the support mode is the first step in controlling its self-balancing. Specifically, when the deformation risk index reaches or exceeds a preset risk threshold (DRI ≥ 0.3), the system switches the support mode of the mobile shelving unit to rigid mode. In this mode, the electromagnetic clutch locks the support arm, and the viscosity of the magnetorheological damper increases to 50 N·s / mm, thereby increasing the vibration suppression rate of the shelving unit to over 90% and ensuring structural stability. Conversely, if the deformation risk index is below the preset risk threshold (DRI < 0.3), a compliant mode is adopted. In this mode, the support arm is fine-tuned at a speed of 0.1 mm / s, and the magnetorheological damper maintains a low viscosity state (5 N·s / mm) to reduce energy consumption and maintain structural flexibility. Next, the system controls the corresponding support arm of the shelving unit to adjust its angle according to the adjustment sequence and future path planning. This process is based on an adjustment strategy generated from the load probability heatmap, deformation risk index, and adjustment priority sequence, ensuring that each support arm can move to the target position according to the predetermined sequence and path. For example, the system may prioritize adjusting the support arms in areas with a high probability of load to ensure these areas receive timely support. Driven by an electronic control system, the linear motors gradually adjust the angle of the support arms until all support arms reach their target positions, completing the self-balancing process of the mobile shelving unit.

[0113] In this optional embodiment, the dynamic balance adjustment of the mobile shelving unit is achieved through precise control of the support arm angle adjustment, significantly improving its stability and safety. The mode switching mechanism ensures that in high-risk conditions, the mobile shelving unit can quickly switch to rigid mode, enhancing structural stability, preventing structural damage caused by sudden load changes, and ensuring the safe and reliable storage of items. In low-risk conditions, a compliant mode is used, which meets support requirements while reducing energy consumption and extending the service life of mechanical components. Angle adjustment is performed according to the adjustment sequence and future path planning, ensuring orderly movement of the support arms, avoiding mutual interference, improving adjustment efficiency, and reducing adjustment time. This precise control method also reduces mechanical wear, extends the service life of the mobile shelving unit, optimizes space utilization, reduces waste caused by safety redundancy space, and achieves a virtuous cycle of storage efficiency and safety, significantly improving the overall storage effect of the mobile shelving unit and enabling it to maintain efficient and stable operation under various working conditions.

[0114] Combination Figure 2 As shown, the present invention also provides a self-balancing load system for mobile shelving units, comprising:

[0115] The monitoring unit is used to acquire multi-source real-time data of the mobile shelving unit, including time-series data, spatial stress data, and environmental parameter data of the mobile shelving unit.

[0116] The prediction unit is used to predict, based on the time series data, the spatial stress data, and the environmental parameter data, the future load data of the mobile shelving unit within a preset future time period using an LSTM model.

[0117] An analysis unit is used to generate an adjustment strategy for the mobile shelving unit based on the future load data.

[0118] A control unit is used to control the mobile shelving unit to perform self-balancing according to the adjustment strategy.

[0119] The self-balancing load system of the mobile shelving unit of the present invention has the same advantages over the prior art as the aforementioned self-balancing load system of the mobile shelving unit, and will not be repeated here.

[0120] The present invention also provides a computer-readable storage medium, including a computer-readable storage medium storing a computer program and a processor, wherein the computer program is read and executed by the processor to implement the self-balancing load method of the mobile shelving unit as described above.

[0121] The computer-readable storage medium of the present invention has the same advantages over the prior art as the above-mentioned self-balancing load method for mobile shelving compared to the prior art, and will not be repeated here.

[0122] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for self-balancing the load of mobile shelving units, characterized in that, include: Acquire multi-source real-time data of the mobile shelving unit, including time-series data, spatial stress data, and environmental parameter data of the mobile shelving unit; specifically, this includes: obtaining the file access timestamp, operation duration, and operation interval period of the mobile shelving unit within a preset historical time period from the database of the mobile shelving unit, and using the file access timestamp, operation duration, and operation interval period as the time-series data; By installing stress sensors on the mobile shelving unit, the mobile shelving unit is monitored in real time to obtain the strain value and strain gradient tensor of the mobile shelving unit, and the strain value and strain gradient tensor are used as the spatial stress data. The temperature, humidity, vibration acceleration, and light intensity of the environment in which the mobile shelving unit is located are obtained through environmental sensors, and the temperature, humidity, vibration acceleration, and light intensity are used as environmental parameter data. Using an LSTM model, based on the time series data, the spatial stress data, and the environmental parameter data, the future load data of the mobile shelving unit within a preset future time period is predicted; specifically, this includes: inputting the input data into the LSTM model for prediction to obtain a load probability heatmap, deformation risk index, and adjustment priority sequence of the mobile shelving unit within the preset future time period. The load probability heatmap, the deformation risk index, and the adjustment priority sequence are used as the future load data. Based on the future load data, an adjustment strategy for the mobile shelving unit is generated, including: planning the adjustment path of the support arm of the mobile shelving unit based on the load probability heatmap, and obtaining the future path plan of the support arm; The support mode of the support arm is determined based on the deformation risk index. Specifically, this includes: determining the support mode based on the relationship between the deformation risk index and the preset risk threshold; Wherein, when the deformation risk index is greater than or equal to the preset risk threshold, the support mode is set to rigid mode; When the deformation risk index is less than the preset risk threshold, the support mode is set to compliant mode, and the damping coefficient of the mobile shelving is set according to preset requirements. The adjustment order of the frame is determined according to the adjustment priority sequence; The adjustment strategy for the mobile shelving unit is generated based on the support mode, the adjustment sequence, and the future path planning of the support arm corresponding to each shelving unit. According to the adjustment strategy, the mobile shelving unit is controlled to perform self-balancing; specifically, this includes switching the mobile shelving unit to a mode according to the support mode. The frame is controlled to adjust the angle of the corresponding support arms according to the adjustment sequence and the future path planning until all the support arms of the mobile shelving unit reach the target position.

2. The self-balancing method for mobile shelving loads according to claim 1, characterized in that, The step of using an LSTM model to predict the future load data of the mobile shelving unit within a preset future time period based on the time series data, the spatial stress data, and the environmental parameter data includes: The time series data, spatial stress data, and environmental parameter data are aligned according to a preset time base using interpolation. The aligned time series data, spatial stress data, and environmental parameter data are then subjected to feature engineering and normalization processes to obtain the input data for the LSTM model.

3. The self-balancing method for the load of mobile shelving units according to claim 2, characterized in that, The step of inputting the input data into the LSTM model for prediction to obtain the load probability heatmap, deformation risk index, and adjustment priority sequence of the compact shelving within the preset future time period includes: The input data is fed into the LSTM model, and features are extracted through the hidden layers of the LSTM model to obtain spatiotemporal features; Attention is allocated to the spatiotemporal features through the attention mechanism layer of the LSTM model to obtain key spatiotemporal node features; The LSTM model's fusion layer performs dimensionality reduction and feature fusion on the key spatiotemporal node features to obtain the load probability heatmap, deformation risk index, and adjustment priority sequence for the preset future time period.

4. The self-balancing method for mobile shelving units according to claim 1, characterized in that, The step of planning the adjustment path of the support arm of the mobile shelving unit based on the load probability heatmap to obtain the future path planning of the support arm includes: Based on the load probability heatmap, the high load probability area of ​​the mobile shelving unit is determined; Based on the high-load probability area, determine the target position of the frame of the mobile shelving unit; The target position of the support arm is obtained by combining the path planning algorithm with the target position and the constraint conditions of the frame corresponding to the support arm. The future path planning is obtained based on the initial position and the target position of the support arm.

5. A self-balancing load system for mobile shelving units, characterized in that, include: The monitoring unit is used to acquire multi-source real-time data of the mobile shelving unit, which includes time-series data, spatial stress data, and environmental parameter data of the mobile shelving unit. Specifically, it includes: acquiring the file access timestamp, operation duration, and operation interval period of the mobile shelving unit within a preset historical time period from the database of the mobile shelving unit, and using the file access timestamp, operation duration, and operation interval period as the time-series data. By installing stress sensors on the mobile shelving unit, the mobile shelving unit is monitored in real time to obtain the strain value and strain gradient tensor of the mobile shelving unit, and the strain value and strain gradient tensor are used as the spatial stress data. The temperature, humidity, vibration acceleration, and light intensity of the environment in which the mobile shelving unit is located are obtained through environmental sensors, and the temperature, humidity, vibration acceleration, and light intensity are used as environmental parameter data. The prediction unit is used to predict the future load data of the mobile shelving within a preset future time period by using an LSTM model based on the time series data, the spatial stress data, and the environmental parameter data; specifically, it includes: inputting the input data into the LSTM model for prediction to obtain a load probability heatmap, deformation risk index, and adjustment priority sequence of the mobile shelving within the preset future time period. The load probability heatmap, the deformation risk index, and the adjustment priority sequence are used as the future load data. An analysis unit is used to generate an adjustment strategy for the mobile shelving unit based on the future load data; including: planning the adjustment path of the support arm of the mobile shelving unit based on the load probability heat map, and obtaining the future path plan of the support arm; The support mode of the support arm is determined based on the deformation risk index. Specifically, this includes: determining the support mode based on the relationship between the deformation risk index and the preset risk threshold; Wherein, when the deformation risk index is greater than or equal to the preset risk threshold, the support mode is set to rigid mode; When the deformation risk index is less than the preset risk threshold, the support mode is set to compliant mode, and the damping coefficient of the mobile shelving is set according to preset requirements. The adjustment order of the frame is determined according to the adjustment priority sequence; The adjustment strategy for the mobile shelving unit is generated based on the support mode, the adjustment sequence, and the future path planning of the support arm corresponding to each shelving unit. The control unit is used to control the mobile shelving to perform self-balancing according to the adjustment strategy; specifically, it includes switching the mobile shelving mode according to the support mode. The frame is controlled to adjust the angle of the corresponding support arms according to the adjustment sequence and the future path planning until all the support arms of the mobile shelving unit reach the target position.

6. A computer-readable storage medium, characterized in that, It includes a computer-readable storage medium storing a computer program and a processor, the computer program being read and executed by the processor to implement the self-balancing load method for mobile shelving as described in any one of claims 1 to 4.

Images