Community intelligent express management method and system

Abstract

The invention relates to the technical field of intelligent express management, and discloses a community intelligent express management method and system, and the method comprises the steps: obtaining the historical travel behavior data, express pickup behavior data and address information of community residents, constructing a multi-express cabinet deployment space model, and outputting a delivery cabinet candidate set; determining a target cabinet number; pushing warehousing state data to residents; a long-short-term memory network is adopted for temperature prediction, and temperature is regulated and controlled in combination with parcel temperature control requirements. Compared with an express cabinet distribution method based on a static address or a fixed path in the prior art, the technical problem that efficient cabinet grid scheduling and environment adaptation are difficult to realize particularly under the conditions that multiple express cabinets exist in a community and resident travel habits are diversified is solved. A better express delivery decision is realized, so that the problems of mistaken delivery, delayed delivery or improper temperature control of the express delivery are avoided, and the delivery accuracy and the package safety are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent express delivery management, and in particular to a community intelligent express delivery management method and system. Background Art

[0002] With the rapid development of terminal logistics systems, smart express lockers are becoming a key infrastructure for improving delivery efficiency and alleviating labor pressure. However, current mainstream express locker systems generally rely on a "point-to-point distribution and static matching" delivery mechanism. Their scheduling strategies rely primarily on fixed address binding or courier experience, making it difficult to accurately adapt to the personalized needs and dynamic pickup behaviors of different residents. This leads to high rates of misdelivery, low delivery efficiency, and a poor pickup experience. For example, in real-world community scenarios, the following situations often occur: multiple express lockers are located at different entrances and exits, but the express delivery system fails to intelligently select based on residents' daily travel routes; different residents have significantly different pickup frequencies and preferred time periods, and the system cannot dynamically predict residents' active hours, resulting in missed deliveries and delayed pickups; and express lockers in different buildings within the same community have severely uneven utilization rates, with some lockers being permanently full while others remain idle and wasted. Furthermore, with the growing demand for cold chain delivery and fresh food packages, most existing express locker systems only offer passive temperature control for special express parcels requiring temperature or humidity control, lacking proactive, predictive adjustment mechanisms. Therefore, there is an urgent need for an intelligent express management method that integrates resident travel pattern modeling, express behavior preference analysis, package attribute identification, spatial collaborative scheduling, and temperature control prediction and adjustment. It can achieve high responsiveness and low error rate under multi-source information perception and complex environmental conditions, as well as full-process optimization of express management with intelligent temperature control, so as to improve the intelligence level, operational efficiency and user satisfaction of the express service system in the community environment. Summary of the Invention

[0003] In response to the above-mentioned technical deficiencies, the purpose of the present invention is to propose a community intelligent express delivery management method, which aims to solve the technical problem that the express delivery cabinet allocation method in the existing technology is mostly based on static addresses or fixed paths, especially when there are multiple express delivery cabinets in the community and residents have diverse travel habits, it is difficult to achieve efficient cabinet scheduling and environmental adaptation.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a community intelligent express delivery management method,

[0005] The community intelligent express delivery management method includes:

[0006] Step S10: Obtain historical travel behavior data, express pickup behavior data, and community resident address information of community residents, pre-build and train a multi-express locker deployment space model, use the historical travel behavior data, express pickup behavior data, and community resident address information as input to the multi-express locker deployment space model, and output a candidate set of lockers;

[0007] Step S20: Automatically collect basic attribute data and extended demand data of the incoming package through QR code recognition or RFID chip recognition, and determine the target cabinet number from the candidate cabinet set based on the basic attribute data and extended demand data;

[0008] Step S30: Send an open control instruction to the target cabinet number and monitor the switch status of the target cabinet number in real time. When the switch status of the target cabinet number is monitored to be closed, send the package storage status data to the community residents in real time via email or mobile phone APP;

[0009] Step S40: collecting an ambient temperature data set in real time, and performing temperature prediction calculation based on the ambient temperature data set in combination with a long short-term memory network method to obtain a temperature error term;

[0010] Step S50: Perform PID control of the temperature control system in the community smart cabinet according to the temperature error term and the expanded demand data.

[0011] Preferably, in step S10, the historical travel behavior data include the average daily travel time period of residents, the travel frequency of residents, the coordinates of the entrances and exits commonly used by residents, the time density distribution of residents entering and leaving the community, and the characteristics of residents' transportation modes; the express pickup behavior data include the frequency of residents' pickup, the peak pickup time period of residents, the location preference of residents' pickup cabinets, the delay time of residents' pickup, and abnormal pickup behavior; the express delivery address information of residents includes the building number of the residents, the geographical relative position between the residents and the express cabinet, and the accessible route data.

[0012] Preferably, in step S10, the structure of the multi-express locker deployment space model specifically includes:

[0013] The feature fusion and input layer receives and integrates historical travel behavior data, express delivery pickup behavior data, and community residents' address information, uniformly encoding them to form a high-dimensional behavior feature vector. Travel behavior and pickup behavior are encoded using time series normalization, and address information is topologically connected to the cabinet layout through spatial coordinate mapping.

[0014] The spatial perception encoding layer is used to extract the spatial accessibility feature vector and walking path accessibility feature vector between the community building node and the express locker node from the high-dimensional behavior feature vector;

[0015] The behavior time aggregation layer is used to extract high-frequency active time window feature vectors from high-dimensional behavior feature vectors;

[0016] The cabinet preference matching layer is used to extract the resident preference feature vector from the high-dimensional behavior feature vector;

[0017] The feature vector fusion layer is used to fuse the spatial accessibility feature vector, the walking path accessibility feature vector, the high-frequency active time window feature vector, and the resident preference feature vector to obtain an intermediate fused feature vector;

[0018] The delivery cabinet candidate set output layer is used to calculate the multi-objective delivery score based on the intermediate fusion feature vector using a multi-factor optimization method, and output the delivery cabinet candidate set based on the multi-objective delivery score.

[0019] Preferably, in step S30, before the step of sending the opening control instruction to the target cabinet number, the method further includes:

[0020] Preset facial database template, collect static facial image of the user, and compare the collected static facial image with the preset facial database template for first-factor identity verification;

[0021] Real-time collection of dynamic facial images of users, and performing second-factor identity verification based on dynamic facial images combined with liveness detection method;

[0022] When the first authentication and the second authentication are passed, an opening control instruction is sent to the target cabinet number.

[0023] Preferably, in step S20, the basic attribute data includes package size information, package weight data, package barcode, RFID chip information, package delivery address information and package type information; the extended requirement data includes temperature control requirements, fragility requirements and special safety requirements.

[0024] Preferably, in step S20, the step of determining the target cabinet number from the candidate cabinet set according to the basic attribute data and the extended demand data specifically includes:

[0025] Obtaining package size information and package weight data from the basic attribute data, matching the cabinet size with the candidate cabinet set based on the package size information and package weight data, and obtaining a first target cabinet number set;

[0026] Obtaining fragility requirements and special security requirements from the expanded demand data, performing cabinet and box security matching based on the fragility requirements and special security requirements, and obtaining a second target cabinet and box number set;

[0027] The temperature control requirements are obtained from the extended demand data, and the cabinet temperatures in the second target cabinet number set are collected through the temperature sensor. The temperature is matched according to the temperature control requirements and the cabinet temperatures to obtain the target cabinet number.

[0028] Preferably, in step S40, the step of collecting an ambient temperature data set in real time and performing temperature prediction calculation based on the ambient temperature data set in combination with a long short-term memory network method to obtain a temperature error term specifically includes:

[0029] Preset a temperature long-short-term memory prediction network, obtain a historical ambient temperature data set, and pre-train the temperature long-short-term memory prediction network using the historical ambient temperature data set;

[0030] Collect ambient temperature data sets and current temperature data in real time, use the ambient temperature data sets as input to the pre-trained temperature long-short-term memory prediction network, and output the predicted temperature;

[0031] The temperature error term is calculated based on the predicted temperature and current temperature data;

[0032] Among them, during the pre-training process of the long short-term memory prediction network, the mean square error term and the first-order derivative difference term are introduced as the loss function used for training; among them, the mean square error term is used to measure the static deviation between the predicted value and the true value; the first-order derivative difference term is used to suppress the sudden change of the predicted value.

[0033] The present invention also provides a community intelligent express delivery management system comprising:

[0034] The resident behavior modeling module is used to obtain historical travel behavior data, express delivery pickup behavior data, and community resident address information of community residents, pre-build and train a multi-express locker deployment space model. The historical travel behavior data, express delivery pickup behavior data, and community resident address information are used as inputs to the multi-express locker deployment space model, and the output is a set of candidate locker boxes.

[0035] The package information recognition and cabinet matching module is used to automatically collect the basic attribute data and extended demand data of the incoming package through QR code recognition or RFID chip recognition, and determine the target cabinet number from the candidate cabinet set based on the basic attribute data and extended demand data;

[0036] The delivery control and status feedback module is used to send an opening control instruction to the target cabinet number and monitor the switch status of the target cabinet number in real time. When the switch status of the target cabinet number is detected to be closed, the package entry status data is sent to community residents in real time via email or mobile phone APP;

[0037] The temperature prediction module is used to collect the ambient temperature data set in real time, and calculate the temperature error term based on the ambient temperature data set combined with the long short-term memory network method;

[0038] The temperature control PID adjustment module is used to perform PID control of the temperature control system in the community smart cabinet based on the temperature error term and expanded demand data.

[0039] The present invention also provides a computer program product, including a community intelligent express delivery management program, which implements the community intelligent express delivery management method when executed by a processor.

[0040] The beneficial effect of the present invention is that compared with the express cabinet allocation method in the prior art which is mostly based on static addresses or fixed paths, especially when there are multiple express cabinets in the community and residents have diverse travel habits, it is difficult to achieve the technical problem of efficient cabinet scheduling and environmental adaptation. Since the present application uses behavior prediction modeling and multi-objective cabinet matching strategy, it realizes better express delivery decision-making, thereby avoiding the problems of express misdelivery, delayed delivery or improper temperature control, and improving delivery accuracy and package safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 This is a flow chart of a first embodiment of a community intelligent express delivery management method of the present invention.

[0043] Figure 2 This is a schematic diagram of equipment for a community intelligent express delivery management method of the present invention. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] Example 1: Figure 1 2 is a flow chart of the first embodiment of the community intelligent express delivery management method of the present invention, which provides the first embodiment of the community intelligent express delivery management method of the present invention.

[0046] In a first embodiment, the community intelligent express delivery management method includes:

[0047] Step S10: Obtain historical travel behavior data, express pickup behavior data, and community resident address information of community residents, pre-build and train a multi-express locker deployment space model, use the historical travel behavior data, express pickup behavior data, and community resident address information as input to the multi-express locker deployment space model, and output a candidate set of lockers;

[0048] It should be noted that in step S10, the structure of the multi-express locker deployment space model specifically includes:

[0049] The feature fusion and input layer receives and integrates historical travel behavior data, express delivery pickup behavior data, and community residents' address information, uniformly encoding them to form a high-dimensional behavior feature vector. Travel behavior and pickup behavior are encoded using time series normalization, and address information is topologically connected to the cabinet layout through spatial coordinate mapping.

[0050] The spatial perception encoding layer is used to extract the spatial accessibility feature vector and walking path accessibility feature vector between the community building node and the express locker node from the high-dimensional behavior feature vector;

[0051] It should be noted that the high-dimensional behavior feature vector is a unified numerical vector generated by uniformly structured encoding of historical travel behavior data related to express delivery pickup, express delivery pickup behavior data, and community resident address information, taking a single community resident as a unit. For example: ;

[0052] in, is the high-dimensional behavior feature vector, Score the frequency of pickup activity for 24 hours a day. is the historical accessibility score of k express lockers, Score the residents' historical preferences for k cabinets, are the geometric coordinates of the building, Auxiliary binary codes are used to represent special pickup needs. The 24-hour pickup frequency vector, the historical accessibility scores of k express lockers, the historical preference scores of residents for k lockers, the geometric coordinates of the building, and the auxiliary binary codes used to represent special pickup needs are all directly or indirectly obtained (using mathematical statistical analysis methods) from historical travel behavior data of community residents, express pickup behavior data, and community address information.

[0053] It is understandable that the historical accessibility score is used to describe the accessibility characteristics of a resident from the building where they live to each express locker during actual historical pickup processes, mainly considering their travel path, walking time, and pickup success rate. This historical accessibility score is obtained by obtaining the recorded path trajectory or access control time of the resident during actual pickup from historical travel behavior data and express pickup behavior data, and combining it with the locker location data to analyze whether the resident has frequently used a certain channel to reach a specific express locker. If a resident has successfully reached a locker multiple times from the same path in the past, and the walking path is short and the walking time is low, the accessibility score of the locker is higher.

[0054] For example, there is an accessible pedestrian passage between Building 5 in the East District where resident A lives and the East Gate cabinet, and more than 70% of A's pickup records in the past month were from this cabinet. The system will count the average travel time and number of passes recorded on this path, generate a corresponding accessibility score, and encode it into its high-dimensional features.

[0055] The historical preference score reflects a resident's subjective preference for a specific parcel locker. It is automatically derived from historical travel and parcel pickup data, including the resident's pickup frequency, locker selection rate, and user feedback. This score is derived by calculating the percentage of pickups made at each locker over a period of time (e.g., 30 days). This score is then normalized based on the feedback provided by the resident after each pickup (e.g., ratings, complaints, and timely pickup). For example, if resident B has a historical preference for atrium locker B, 60% of parcels were delivered to this locker over the past month, and the average rating across all pickups is "satisfactory," this locker will have a higher historical preference score than other lockers.

[0056] It should be noted that the spatial accessibility feature vector and the walking path accessibility feature vector are mainly obtained based on the geometric coordinates of the building and the historical accessibility score in the high-dimensional behavior feature vector. Among them, the spatial accessibility feature vector is mainly obtained by calculating the straight-line distance between the geometric coordinates of the building and the smart express cabinet, and the walking path accessibility feature vector is obtained by normalizing the historical accessibility score.

[0057] The behavior time aggregation layer is used to extract high-frequency active time window feature vectors from high-dimensional behavior feature vectors;

[0058] It should be noted that the high-frequency active time window feature vector is mainly obtained by the periodic heat sliding window aggregation method based on the 24-hour pickup frequency score in the high-dimensional behavior feature vector, specifically including: Using the 24-dimensional time vector for 30 consecutive days , divide the window into 2-hour units, and generate 12 sliding windows in total; calculate the average active frequency in each window to get the pickup activity score , extract the time period with the highest score, and combine it with the standard deviation score of the past week to evaluate its volatility, forming the time window stability feature. The output result is: high-frequency active time window encoding vector, where each element represents the resident's tendency to pick up the item in that time period. The higher the value, the more likely it is to pick up the item within that period.

[0059] The cabinet preference matching layer is used to extract the resident preference feature vector from the high-dimensional behavior feature vector;

[0060] It should be noted that the resident preference feature vector is obtained by using the behavioral retrospective scoring method based on the historical preference scores of residents for k cabinets in the high-dimensional behavioral feature vector. The calculation of the historical preference score specifically includes:

[0061] Establish a historical interaction matrix between residents and cabinets If residents In the cabinet Pick up, then , otherwise it is 0; introduce the interaction quality factor Each interaction is based on the time delay, whether there is a complaint, and satisfaction feedback as the interaction quality factor Assignment, calculate the residents according to the following formula For the first Historical preference score for each cabinet: ,in, Score historical preferences, is the set of interaction history time points, is the stabilization regularization term. Among them, the interaction quality factor It can be determined by the following methods: "whether the package is picked up in time (for example, picking up within 24 hours is recorded as 1, and overtime is recorded as 0.5 or 0), whether a complaint is made about the counter (complaints are recorded as negative weights such as -1), and the express delivery integrity / resident satisfaction score (the value range is 0 to 5)". The value is taken as a positive value, and the range is generally [0, 1] or [-1, 1]. The higher the value, the higher the resident's satisfaction with the counter. The specific value can be a weighted average or discrete score. To stabilize the regularization term, the value is a constant, such as 1 or 2, in order to avoid When it is very small, the score is too extreme or unstable, usually in the range of [0.5,5]; is a set of interactive historical time points, used to represent the residents Historically, in the cabinet The number of times a cabinet has been used to retrieve items can be directly obtained from system logs, access control clock-in records, cabinet code scanning records, etc.

[0062] This formula is a typical weighted average and regularization strategy for scoring, and belongs to the empirical scoring method in behavioral modeling. The numerator is the sum of the quality of all interactions between residents and the cabinets, reflecting the "intensity" of preference, and the denominator is the number of interactions + the regularization term, reflecting the "credibility" of preference. The overall result is a stable, interpretable, and updateable personalized scoring indicator used to indicate whether residents tend to use a certain cabinet and the quality of their usage experience.

[0063] The feature vector fusion layer is used to fuse the spatial accessibility feature vector and the walking path accessibility feature vector, the high-frequency active time window feature vector and the resident preference feature vector to obtain an intermediate fusion feature vector; the delivery cabinet candidate set output layer is used to calculate the multi-objective delivery score based on the intermediate fusion feature vector using a multi-factor optimization method, and output the delivery cabinet candidate set based on the multi-objective delivery score.

[0064] It should be noted that the intermediate fusion feature vector is a weighted process that uses a multi-channel attention mechanism to process different types of input features. Each input feature is first dimensionally unified through a fully connected layer, and then a weight is calculated for each feature through the attention mechanism. Finally, all features are weighted and fused to form an intermediate fusion feature vector. Each component feature vector in the intermediate fusion feature vector is automatically matched with the relevant information relative to the express delivery cabinet to generate a multi-target delivery score. A multi-target delivery score threshold is preset, and the set of corresponding express delivery cabinet numbers with a multi-target delivery score greater than the multi-target delivery score threshold is output as the delivery cabinet candidate set.

[0065] It is understandable that by deeply integrating behavioral characteristics, spatial layout and time preferences, the spatial model not only has static geographic computing capabilities, but can also dynamically adapt to changes in user individual behavior, thereby realizing a multi-factor comprehensive cabinet recommendation mechanism of behavior-driven + geographic perception + time synchronization, which is significantly better than the traditional single delivery matching method that relies on address or distance.

[0066] It should be understood that traditional systems typically deliver packages to a single locker based solely on the resident's address or the courier's proximity, making it difficult to fully adapt to the diverse behaviors of residents within a community and the current situation of multiple lockers coexisting. In contrast, the multi-locker deployment model designed in this invention offers the following technical advantages: it can automatically match lockers with high hit rates based on users' actual usage paths and daily active times; it can dynamically select the most suitable multiple alternative lockers in a community with multiple lockers, improving hit rates and distributing load; and it supports subsequent algorithm modules in high-availability delivery path planning and scheduling decisions.

[0067] For example, in a simulation test of a large residential community, resident A lived in the east unit of Building 5. She left home at 8 a.m. and returned home at 9 p.m. Her historical pickup times were concentrated around 8 p.m. on Fridays, giving her a preference for the east gate locker. By building a spatial model for multiple lockers, the system identified her as having the shortest distance to the east gate locker, a smoother route, a high degree of overlap in active hours, and a higher preference score. During a delivery, the model prioritized the east gate locker as a candidate. When the east gate locker was full, the model recommended the north gate locker, which had a similar route, as a second choice. This improved the hit rate by approximately 31.4% in actual testing, and reduced the average delivery path by 23.2%.

[0068] Step S20: Automatically collect basic attribute data and extended demand data of the incoming package through QR code recognition or RFID chip recognition, and determine the target cabinet number from the candidate cabinet set based on the basic attribute data and extended demand data;

[0069] It should be noted that in step S20, the step of determining the target cabinet number from the candidate cabinet set based on the basic attribute data and the extended demand data specifically includes: obtaining the package size information and the package weight data from the basic attribute data, matching the cabinet size in the candidate cabinet set based on the package size information and the package weight data, and obtaining a first target cabinet number set; obtaining the fragility requirements and the special safety requirements from the extended demand data, matching the cabinet security based on the fragility requirements and the special safety requirements, and obtaining a second target cabinet number set; obtaining the temperature control requirements from the extended demand data, collecting the cabinet temperature in the second target cabinet number set through a temperature sensor, performing temperature matching based on the temperature control requirements and the cabinet temperature, and obtaining the target cabinet number.

[0070] In step S30, by matching multi-dimensional data (size, weight, safety, temperature control, etc.), the most suitable cabinet is accurately selected for parcel storage. This ensures that parcels receive the appropriate storage space based on their size and weight, while also providing additional protection based on expanded requirements (such as temperature control and safety), mitigating risks during parcel storage. Furthermore, through real-time temperature monitoring, the system ensures that parcels with special requirements (such as temperature-controlled storage) receive the optimal storage environment.

[0071] For example, consider a package measuring 40 × 30 × 20 cm, weighing 3 kg, and marked as fragile requiring cold chain storage. First, by matching the dimensions and weight, we select cabinets that match the package size, obtaining the first target cabinet number set. Next, based on the package's fragility requirements, we select cabinets with reinforced designs, obtaining the second target cabinet number set. Finally, a temperature sensor measures the temperature of the second target cabinet. Assuming the current cabinet temperature is 4°C and the package's temperature control requirement is between 2°C and 8°C, we determine that the cabinet meets the temperature control requirements and ultimately select it as the target cabinet number, ensuring the package's safe storage.

[0072] Step S30: Send an open control instruction to the target cabinet number and monitor the switch status of the target cabinet number in real time. When the switch status of the target cabinet number is monitored to be closed, send the package storage status data to the community residents in real time via email or mobile phone APP;

[0073] It should be understood that this approach allows users to receive timely and accurate notifications on the storage status of their parcels after they are stored, improving user experience and system transparency. Instant notifications via email or app ensure real-time feedback on the parcel's storage status, avoiding inconvenience caused by delayed information.

[0074] For example, suppose a user's package is deposited in compartment 101 by a courier. A switch sensor detects that the compartment switches from open to closed within three seconds. At this point, incoming status data is generated containing the following information: Package ID: 12345; Compartment Number: 101; Incoming Time: April 28, 2023, 14:30:00. Next, using the smptlib library, an email is sent to the user with the following content: Subject: Package Incoming Notification; Body: Your package (ID: 12345) has been successfully deposited in compartment 101. The deposit time is April 28, 2023, 14:30:00. You can pick it up at any time. Simultaneously, the same incoming status information is pushed to the user via the app to ensure accurate and timely communication.

[0075] Step S40: collecting an ambient temperature data set in real time, and performing temperature prediction calculation based on the ambient temperature data set in combination with a long short-term memory network method to obtain a temperature error term;

[0076] It should be noted that in step S40, the step of collecting the ambient temperature data set in real time, and performing temperature prediction and calculation based on the ambient temperature data set in combination with the long short-term memory network method to obtain the temperature error term specifically includes: presetting a temperature long short-term memory prediction network, obtaining a historical ambient temperature data set, and pre-training the temperature long short-term memory prediction network through the historical ambient temperature data set; collecting the ambient temperature data set and the current temperature data in real time, using the ambient temperature data set as the input of the pre-trained temperature long short-term memory prediction network, and outputting the predicted temperature; calculating the temperature error term based on the predicted temperature and the current temperature data; wherein, in the process of pre-training the long short-term memory prediction network, the mean square error term and the first-order derivative difference term are introduced as the loss function used for training; wherein, the mean square error term is used to measure the static deviation between the predicted value and the true value; and the first-order derivative difference term is used to suppress the sudden change of the predicted value.

[0077] It's understandable that by combining the LSTM method with environmental sensor data, it's possible to perceive and predict temperature trends within community express lockers, thereby identifying potential temperature anomalies in advance and improving the foresight and accuracy of temperature adjustments. Unlike traditional control methods that rely solely on current temperature values, this method incorporates a temperature trend prediction mechanism, effectively avoiding issues like over-regulation or delayed regulation.

[0078] It should be understood that compared to traditional single-point PID feedback control methods, this embodiment, by introducing a predictive drive mechanism and a robust optimization loss function, effectively reduces control errors caused by sudden environmental changes (such as sunlight, high outdoor temperatures, and cabinet door openings), thereby enhancing the temperature control system's adaptability and control accuracy in highly variable environments. Furthermore, by introducing a first-order derivative control term, the model can more smoothly fit the actual temperature curve, avoiding oscillation or overcompensation at "inflection points."

[0079] For example, in a simulation experiment, the system collected the past 48 hours of temperature data from a parcel locker as training samples. Under strong summer sunlight, the temperature inside the locker fluctuated dramatically. Traditional MSE training resulted in frequent fluctuations and large errors in the LSTM model output. By introducing the first-order derivative difference term, the model's prediction curve became smoother and more closely aligned with the actual temperature trend. In a 24-hour rolling forecast, the average prediction error decreased from ±1.7°C to ±0.8°C, and the prediction response latency was reduced by approximately 34%. This temperature error term was incorporated into the subsequent PID control, further optimizing the response efficiency and energy stability of the locker temperature control system.

[0080] Step S50: Perform PID control of the temperature control system in the community smart cabinet according to the temperature error term and the expanded demand data.

[0081] It's understandable that by introducing a predicted temperature error term to drive the PID control mechanism, the system can proactively execute adjustments based on predicted trends, avoiding the traditional "passive response" problem based on current temperature, thereby improving adjustment response speed and energy efficiency. Furthermore, expanding the incorporation of demand data into control parameter selection enables differentiated and precise temperature control, meeting the requirements of diverse package storage environments.

[0082] It should be understood that compared to the "constant temperature deadband control" or "single PID adjustment" methods used in conventional temperature-controlled cabinets, this embodiment transforms PID control from a "single-target, fixed-parameter" approach to a dynamic, closed-loop system with multiple objectives by integrating prediction errors with the personalized needs of packages. This solution can effectively mitigate temperature control instability caused by delayed or overreacting adjustments, especially when faced with factors such as large diurnal temperature swings and frequent cabinet door openings during actual deployment.

[0083] For example, in a community smart cabinet environment simulation experiment, a batch of fresh food packages required the cabinet temperature to be maintained at 2°C to 8°C. A "steady-state PID parameter template" was preset for this type of package. Under strong sunlight at noon, the predicted temperature was 9.2°C, while the current temperature was 8.3°C, resulting in a system-identified error term of 0.9°C. The PID controller calculated the cooling power command, driving the cooling module to operate in advance to prevent further temperature increases. The temperature difference during the control process was stabilized within ±0.5°C. Compared with traditional PID control without prediction, the temperature control response was approximately 11 minutes earlier, and the cold start frequency was reduced by approximately 27%, effectively improving temperature control accuracy and energy consumption control.

[0084] Embodiment 2: In addition, the present invention provides a community intelligent express delivery management system that utilizes the community intelligent express delivery management method described in the above embodiment to solve the technical problems of community intelligent express delivery management. Compared with the prior art, the beneficial effects of the community intelligent express delivery management system provided by the present invention are the same as those of the community intelligent express delivery management method described in the above embodiment. Other technical features of the community intelligent express delivery management system are the same as those disclosed in the above embodiment and are not further described here.

[0085] Example 3: The present invention provides a community intelligent express management device, please refer to Figure 2A community intelligent express delivery management device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the community intelligent express delivery management method described in the first embodiment. The community intelligent express delivery management device in this embodiment of the present invention may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The community intelligent express delivery management device is merely an example and should not limit the functionality and scope of use of the embodiments of the present invention. The community intelligent express delivery management device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of a community intelligent express delivery management device. The processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to the I / O interface 1006: an input device 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, hard disk, etc.; and a communication device 1009. The communication device 1009 allows the community intelligent express delivery management device to communicate with other devices wirelessly or wired to exchange data. While the figure illustrates a community intelligent express delivery management device with various systems, it should be understood that implementation or inclusion of all illustrated systems is not required. More or fewer systems may alternatively be implemented or included.

[0086] Example 4: The present invention also provides a computer program product, comprising a computer program. When executed by a processor, the computer program implements the steps of the aforementioned community intelligent express delivery management method. The computer program product provided by the present invention can solve the technical problem of community intelligent express delivery management. Compared with the prior art, the beneficial effects of the computer program product provided by the present invention are the same as those of the community intelligent express delivery management method provided by the aforementioned embodiment, and are not further elaborated here.

[0087] In particular, according to the embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present invention are performed.

[0088] It should be understood that the various parts disclosed in the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any appropriate manner in any one or more embodiments or examples.

[0089] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A community intelligent express delivery management method, characterized in that: Methods include: Step S10: Obtain historical travel behavior data, express pickup behavior data, and community resident address information of community residents, pre-build and train a multi-express locker deployment space model, use the historical travel behavior data, express pickup behavior data, and community resident address information as input to the multi-express locker deployment space model, and output a candidate set of lockers; Step S20: Automatically collect basic attribute data and extended demand data of the incoming package through QR code recognition or RFID chip recognition, and determine the target cabinet number from the candidate cabinet set based on the basic attribute data and extended demand data; Step S30: Send an open control instruction to the target cabinet number and monitor the switch status of the target cabinet number in real time. When the switch status of the target cabinet number is monitored to be closed, send the package storage status data to the community residents in real time via email or mobile phone APP; Step S40: collecting an ambient temperature data set in real time, and performing temperature prediction calculation based on the ambient temperature data set in combination with a long short-term memory network method to obtain a temperature error term; Step S50: Perform PID control of the temperature control system in the community smart cabinet according to the temperature error term and the expanded demand data.

2. A community intelligent express delivery management method according to claim 1, characterized in that: In step S10, the historical travel behavior data includes the average daily travel time period of residents, the frequency of residents' travel, the coordinates of residents' commonly used entrances and exits, the time density distribution of residents' entry and exit of the community, and the characteristics of residents' transportation methods; the express pickup behavior data includes the frequency of residents' pickup, the peak pickup time period of residents, the location preference of residents' pickup cabinets, the delay time of residents' pickup, and abnormal pickup behavior; the express delivery address information of residents includes the building number of the residents, the geographical relative position between the residents and the express cabinet, and the accessible route data.

3. A community intelligent express delivery management method according to claim 1, characterized in that: In step S10, the structure of the multi-express locker deployment space model specifically includes: The feature fusion and input layer receives and integrates historical travel behavior data, express delivery pickup behavior data, and community residents' address information, uniformly encoding them to form a high-dimensional behavior feature vector. Travel behavior and pickup behavior are encoded using time series normalization, and address information is topologically connected to the cabinet layout through spatial coordinate mapping. The spatial perception encoding layer is used to extract the spatial accessibility feature vector and walking path accessibility feature vector between the community building node and the express locker node from the high-dimensional behavior feature vector; The behavior time aggregation layer is used to extract high-frequency active time window feature vectors from high-dimensional behavior feature vectors; The cabinet preference matching layer is used to extract the resident preference feature vector from the high-dimensional behavior feature vector; The feature vector fusion layer is used to fuse the spatial accessibility feature vector, the walking path accessibility feature vector, the high-frequency active time window feature vector, and the resident preference feature vector to obtain an intermediate fused feature vector; The delivery cabinet candidate set output layer is used to calculate the multi-objective delivery score based on the intermediate fusion feature vector using a multi-factor optimization method, and output the delivery cabinet candidate set based on the multi-objective delivery score.

4. A community intelligent express delivery management method according to claim 1, characterized in that: In step S30, before the step of sending the opening control instruction to the target cabinet number, the method further includes: Preset facial database template, collect static facial image of the user, and compare the collected static facial image with the preset facial database template for first-factor identity verification; Real-time collection of dynamic facial images of users, and performing second-factor identity verification based on dynamic facial images combined with liveness detection method; When the first authentication and the second authentication are passed, an opening control instruction is sent to the target cabinet number.

5. A community intelligent express delivery management method according to claim 1, characterized in that: In step S20, the basic attribute data includes package size information, package weight data, package barcode, RFID chip information, package delivery address information and package type information; the extended requirement data includes temperature control requirements, fragility requirements and special safety requirements.

6. A community intelligent express delivery management method according to claim 1, characterized in that: In step S20, the step of determining the target cabinet number from the candidate cabinet set according to the basic attribute data and the expanded demand data specifically includes: Obtaining package size information and package weight data from the basic attribute data, matching the cabinet size with the candidate cabinet set based on the package size information and package weight data, and obtaining a first target cabinet number set; Obtaining fragility requirements and special security requirements from the expanded demand data, performing cabinet and box security matching based on the fragility requirements and special security requirements, and obtaining a second target cabinet and box number set; The temperature control requirements are obtained from the extended demand data, and the cabinet temperatures in the second target cabinet number set are collected through the temperature sensor. The temperature is matched according to the temperature control requirements and the cabinet temperatures to obtain the target cabinet number.

7. A community intelligent express delivery management method according to claim 1, characterized in that: In step S40, the ambient temperature data set is collected in real time, and the temperature error term is obtained by performing temperature prediction calculation based on the ambient temperature data set in combination with the long short-term memory network method. Specifically, the steps include: Preset a temperature long-short-term memory prediction network, obtain a historical ambient temperature data set, and pre-train the temperature long-short-term memory prediction network using the historical ambient temperature data set; Collect ambient temperature data sets and current temperature data in real time, use the ambient temperature data sets as input to the pre-trained temperature long-short-term memory prediction network, and output the predicted temperature; The temperature error term is calculated based on the predicted temperature and current temperature data; Among them, during the pre-training process of the long short-term memory prediction network, the mean square error term and the first-order derivative difference term are introduced as the loss function used for training; among them, the mean square error term is used to measure the static deviation between the predicted value and the true value; the first-order derivative difference term is used to suppress the sudden change of the predicted value.

8. A community intelligent express delivery management system, applied to a community intelligent express delivery management method according to any one of claims 1 to 7, characterized in that: The community intelligent express delivery management system includes: The resident behavior modeling module is used to obtain historical travel behavior data, express delivery pickup behavior data, and community resident address information of community residents, pre-build and train a multi-express locker deployment space model. The historical travel behavior data, express delivery pickup behavior data, and community resident address information are used as inputs to the multi-express locker deployment space model, and the output is a set of candidate locker boxes. The package information recognition and cabinet matching module is used to automatically collect the basic attribute data and extended demand data of the incoming package through QR code recognition or RFID chip recognition, and determine the target cabinet number from the candidate cabinet set based on the basic attribute data and extended demand data; The delivery control and status feedback module is used to send an opening control instruction to the target cabinet number and monitor the switch status of the target cabinet number in real time. When the switch status of the target cabinet number is detected to be closed, the package entry status data is sent to community residents in real time via email or mobile phone APP; The temperature prediction module is used to collect the ambient temperature data set in real time, and perform temperature prediction calculation based on the ambient temperature data set combined with the long short-term memory network method to obtain the temperature error term; The temperature control PID adjustment module is used to perform PID control of the temperature control system in the community smart cabinet based on the temperature error term and expanded demand data.

9. A community intelligent express delivery management device, characterized in that: The community intelligent express management device includes: a memory, a processor, and a community intelligent express management program stored in the memory and executable on the processor. When the community intelligent express management program is executed by the processor, a community intelligent express management method according to any one of claims 1 to 7 is implemented.

10. A computer program product, characterized in that The computer program product includes a community intelligent express delivery management program, which, when executed by a processor, implements a community intelligent express delivery management method according to any one of claims 1 to 7.

Images