Dynamic configuration method for food cold chain transportation resources

Abstract

The invention discloses a dynamic configuration method for food cold-chain transportation resources, which comprises the following steps: constructing a distributed cold-chain intelligent agent network, inputting a pre-trained LSTM model, outputting low, medium and high three-level risk signals, and when the risk signal is low, adjusting the vehicle speed to a preset energy-saving interval and reducing the refrigeration power by 10-15%; and when the risk signal is medium, switching to a standby transportation path according to a real-time road condition, calculating contribution degree weights of food enterprises, transportation parties and consumers based on a Shapley value algorithm, allocating transportation resources, and calling social idle refrigerator car resources through a block chain platform. Through cooperative use of real-time data perception and the LSTM risk assessment model, the effect of dynamically identifying the risk level of the transportation environment is achieved, the problem of emergency response delay caused by static weight distribution in the prior art is solved, and it is ensured that temperature-sensitive materials can still maintain stable temperature control when abnormities suddenly occur.

Description

Technical Field

[0001] The present invention relates to the field of food cold chain transportation, and in particular to a method for dynamically configuring food cold chain transportation resources. Background Art

[0002] Food cold chain transportation is a critical link in ensuring food safety and quality, requiring efficient coordination of multiple resources, including transport vehicles, cold storage nodes, and temperature control equipment. With the rapid development of industries like fresh food e-commerce and the pharmaceutical cold chain, demand for cold chain logistics has exploded, driving ever-increasing requirements for transportation efficiency, temperature control accuracy, and resource utilization.

[0003] Currently, the cold chain transportation industry has widely adopted Internet of Things (IoT) technologies, such as on-board temperature and humidity sensors, GPS positioning systems, and cloud-based data management platforms, to enable real-time monitoring of the transportation process. Furthermore, some companies have experimented with implementing intelligent scheduling algorithms (such as route optimization and energy management) to improve transportation efficiency.

[0004] For example, a method for dynamic configuration of food cold chain transportation resources disclosed in the authorization announcement number CN108090711B includes S1, establishing a food cold chain transportation service network model; S2, balancing the multiple demands of food enterprises, dynamic scheduling of transportation resources, and consumers based on the scale of the transportation service network, the size of transportation demand and its physical distribution, the saturation of transportation resources relative to demand, and the cost structure, and assigning different weights to the distribution of transportation resources in the food cold chain transportation service network based on solitary waves; S3, according to the solitary wave operation law equation of the food cold chain transportation service network flow; using the orthogonal phase queuing theory, according to the different weights of different types of transportation resources, the food cold chain transportation service network is adaptively and dynamically configured with transportation resources. After adopting the above technical solution, the beneficial effects of the present invention are: it can effectively ensure food safety and reduce the loss of food due to deterioration during transportation.

[0005] Although the above patent realizes the dynamic weight allocation of transportation resources through the solitary wave model and orthogonal phase queuing theory, and can optimize the resource utilization of the cold chain transportation network to a certain extent, its static weight setting method is difficult to adapt to sudden environmental changes (such as extreme weather and traffic anomalies), resulting in insufficient dynamic response capabilities.

[0006] Specifically, the weight allocation of this patent is based on preset rules and is not dynamically adjusted in combination with real-time risk data. When a transport vehicle suddenly breaks down or encounters extreme weather, the fixed weight cannot prioritize the scheduling of key resources (such as spare cold storage and emergency vehicles). The differences in sensitivity of different temperature-sensitive materials (such as vaccines and ordinary fresh food) to timeliness and temperature have not been quantified, which may lead to high-value materials and low-priority materials competing for resources with the same weight. Summary of the Invention

[0007] The purpose of the present invention is to provide a method for dynamically configuring food cold chain transportation resources to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A method for dynamically configuring food cold chain transportation resources comprises the following steps:

[0010] Step S1: Build a distributed cold chain agent network, deploy edge computing devices at transport vehicles and cold storage nodes to form multiple agents, which exchange location, temperature, and equipment status data in real time through the MQTT protocol;

[0011] Step S2: Dynamic risk assessment. Each agent collects data on ambient temperature, humidity, traffic conditions, and equipment energy consumption, inputs the data into a pre-trained LSTM model, and outputs three risk levels: low, medium, and high.

[0012] Step S3: Anti-interference strategy execution. When the risk signal is low, the vehicle speed is adjusted to a preset energy-saving range and the cooling power is reduced by 10%-15%. When the risk signal is medium, an alternative transportation route is switched based on real-time road conditions. The alternative route is provided by the AutoNavi Map API. When the risk signal is high, an emergency cold storage transfer protocol is initiated, which includes inventory retrieval and reservation of the nearest cold storage.

[0013] Step S4, multi-agent resource allocation: Calculate the contribution weights of food companies, transporters, and consumers based on the Shapley value algorithm, and allocate transportation resources. The calculation formula for the contribution weights is:

[0014]

[0015] Step S5, flexible resource scheduling: call on idle refrigerated truck resources in the society through the blockchain platform. The platform deploys smart contracts, and the contract code includes resource retrieval, token payment and breach of contract penalty logic.

[0016] In the present invention, in step S1, the edge computing device includes a vehicle-mounted terminal and a cold storage control terminal, the vehicle-mounted terminal has a built-in GPS module and a temperature sensor, and the cold storage control terminal is connected to the storage door switch and the refrigeration compressor.

[0017] In the present invention, in step S2, the training data of the LSTM model includes historical temperature fluctuation records, traffic delay events and equipment failure logs, the input data is normalized to the interval [0,1], and the output layer uses the Softmax function to classify the risk level.

[0018] In the present invention, in step S3, the execution steps of the emergency cold storage transfer protocol include:

[0019] Step S301: The agent sends an inventory query request to the surrounding cold storage, and the cold storage returns the available storage capacity and temperature range;

[0020] In step S302, the agent selects a cold storage with a distance of ≤5km and a matching temperature, and completes the reservation through blockchain payment tokens.

[0021] In the present invention, in step S4, when the contribution weight is calculated, a risk compensation coefficient λ is introduced. When the risk level is high, λ=1.5; when the risk level is medium, λ=1.2; when the risk level is low, λ=1.0.

[0022] A food cold chain transportation resource dynamic configuration system, comprising:

[0023] Multiple intelligent terminals are deployed in transport vehicles and cold storages to execute the method;

[0024] Cloud server, which communicates with the intelligent terminal and stores the LSTM model and Shapley value algorithm program;

[0025] Blockchain nodes run smart contracts to manage elastic resource scheduling.

[0026] A computer-readable storage medium stores a computer program, which implements the steps of the method when executed by a processor.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] 1. This invention achieves the effect of dynamically identifying the risk level of the transportation environment by combining real-time data perception with the LSTM risk assessment model. It solves the emergency response delay problem caused by static weight allocation in the existing technology and ensures that temperature-sensitive materials can maintain stable temperature control in the event of sudden abnormalities.

[0029] 2. This invention achieves the effect of quantifying multi-party contributions and flexible resource scheduling through the coordinated execution of the Shapley value algorithm and blockchain smart contracts, solving the problem of rigid resource allocation caused by fixed rules in traditional methods, and realizing automatic priority allocation of high-value materials and transparent settlement of token incentives. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a three-layer architecture diagram of the system of the present invention;

[0031] Figure 2 This is a flow chart of the dynamic configuration method of the present invention;

[0032] Figure 3 This is a data processing flow chart of the LSTM model of the present invention;

[0033] Figure 4 This is a sequence diagram of the emergency cold storage transfer protocol of the present invention;

[0034] Figure 5 This is the state diagram for calculating the Shapley value of the present invention. DETAILED DESCRIPTION

[0035] 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.

[0036] Example 1

[0037] This invention addresses the problem of inefficient resource allocation caused by environmental fluctuations, traffic anomalies, and equipment failures during cold chain transportation. It achieves dynamic risk response and optimal resource allocation through distributed intelligent agent collaborative decision-making and blockchain elastic scheduling. Because it involves edge computing, deep learning, and game theory algorithms, it relies on hardware architecture to achieve the following technical goals:

[0038] Real-time data perception: Collect multi-dimensional environmental parameters through vehicle-mounted terminals and cold storage control terminals;

[0039] Intelligent risk assessment: Utilizes LSTM models to achieve low-latency (<500ms) risk level classification;

[0040] Decentralized resource scheduling: Based on the Shapley value algorithm and smart contracts, multi-party contribution quantification and token incentives are achieved.

[0041] Based on the above technical requirements, this embodiment adopts Figure 1 The "edge-cloud-chain" three-layer architecture shown:

[0042] Edge layer: Deploy vehicle terminals (including GPS, temperature and humidity sensors) and cold storage control terminals (connected to the door switch and refrigeration compressor) to transmit data in real time through the MQTT protocol;

[0043] Cloud service layer: Runs the LSTM model (input normalized to [0,1]) and the Shapley value algorithm to output risk level and contribution weight;

[0044] Blockchain layer: Ethereum private chain nodes manage smart contracts, execute token payments (ERC-20 standard) and cold storage resource retrieval.

[0045] like Figure 2 As shown, the specific process of the dynamic configuration method implemented on this architecture includes:

[0046] Step S1: Build a distributed cold chain agent network, deploy edge computing devices at transport vehicles and cold storage nodes to form multiple agents, which exchange location, temperature, and equipment status data in real time through the MQTT protocol.

[0047] In step S1, the edge computing device includes a vehicle-mounted terminal and a cold storage control terminal. The vehicle-mounted terminal has a built-in GPS module and a temperature sensor, and the cold storage control terminal is connected to a door switch and a refrigeration compressor.

[0048] Step S2: Dynamic risk assessment. Each agent collects environmental temperature, humidity, traffic status, and equipment energy consumption data, inputs them into the pre-trained LSTM model, and outputs three-level risk signals: low, medium, and high.

[0049] In step S2, the training data of the LSTM model includes historical temperature fluctuation records, traffic delay events, and equipment failure logs. The input data is normalized to the interval [0, 1], and the output layer uses the Softmax function to classify the risk level.

[0050] like Figure 3 As shown, the LSTM anomaly detection model formula is as follows:

[0051] f t =σ(W f ·[h t-1 ,x t ]·+b f )

[0052] i t =σ(W i ·[h t-1 ,x t ]+b i )

[0053]

[0054] O t =σ(W O ·[h t-1 ,x t ]+b o )

[0055] h t =O t ⊙tanh(C t )

[0056] Among them, x t represents input data (temperature, humidity, traffic delay index, energy consumption anomaly, normalized to [0,1]), h t represents the hidden state, C tIndicates the cell state, W f , W i , W c , W O represents the weight matrix, b f , b i , b C , b o Represents the bias term, σ represents the Sigmoid activation function, and ⊙ represents element-wise multiplication.

[0057] Step S3: Anti-interference strategy execution: When the risk signal is low, the vehicle speed is adjusted to the preset energy-saving range and the cooling power is reduced by 10%-15%. When the risk signal is medium, the vehicle switches to an alternative transportation route based on the real-time road conditions. The alternative route is provided by the AutoNavi Map API. When the risk signal is high, the emergency cold storage transfer protocol is initiated, which includes inventory retrieval and reservation of the nearest cold storage.

[0058] like Figure 4 As shown, in step S3, the execution steps of the emergency cold storage transfer protocol include:

[0059] Step S301: The agent sends an inventory query request to the surrounding cold storage, and the cold storage returns the available storage capacity and temperature range;

[0060] In step S302, the agent selects a cold storage with a distance of ≤5km and a matching temperature, and completes the reservation through blockchain payment tokens.

[0061] Step S4: Multi-agent resource allocation: Calculate the contribution weights of food companies, transporters, and consumers based on the Shapley value algorithm and allocate transportation resources. The calculation formula for the contribution weights is:

[0062]

[0063] in, represents participant i, S represents the alliance subset, v(S) represents the benefit function of alliance S, λ R Represents the risk compensation coefficient.

[0064] like Figure 5 As shown, in step S4, when the contribution weight is calculated, the risk compensation coefficient λ is introduced. When the risk level is high, λ=1.5; when the risk level is medium, λ=1.2; when the risk level is low, λ=1.

[0065] Step S5: Flexible resource scheduling: The idle refrigerated truck resources in the society are called through the blockchain platform. The platform deploys smart contracts, and the contract code includes resource retrieval, token payment and breach of contract penalty logic.

[0066] In step S5, the smart contract token incentive calculation formula is as follows:

[0067]

[0068] Among them, the basic reward represents the preset value, and the cold storage capacity is 10m 3 The base unit is 100km.

[0069] Example 2

[0070] Suppose a cold chain company transports frozen meat at -20°C from City A to City B (1500 km away) and encounters the following anomaly:

[0071] Heavy rain on Section C (humidity 95%, temperature suddenly rose to -18.5°C);

[0072] The vehicle's refrigeration compressor is faulty (energy consumption fluctuates by ±20%).

[0073] The specific steps of this embodiment are as follows:

[0074] Step S1:

[0075] The vehicle agent (ID: V002) reports the following data: humidity (0.95), temperature deviation (0.025), and energy consumption anomaly (0.2).

[0076] D cold storage (ID: W004) returns the available capacity (10m 3 , -22℃~-15℃).

[0077] Step S2:

[0078] LSTM outputs high risk (probability [0.1, 0.3, 0.6]), triggering the emergency protocol.

[0079] Step S3:

[0080] Start cold storage transfer: search for cold storage (W004) with a distance of ≤5km;

[0081] Smart contract payment 80Token Reserve cold storage.

[0082] Step S4:

[0083] Shapley value calculation (λ R =1.5):

[0084] Transporter contribution: 4500 × 1.5 = 6750 (value of avoided cargo damage);

[0085] Contribution of cold storage: 3000×1.5=4500 (providing emergency storage).

[0086] result:

[0087] The frozen meat was temporarily stored in cold storage D for 2 hours and the temperature returned to -20°C;

[0088] The total cost savings is 800 yuan (the original goods loss is estimated to be 1,200 yuan), and the weights of all parties are distributed transparently.

[0089] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0090] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A dynamic configuration method for food cold chain transportation resources, characterized in that: The following steps are involved: Step S1: Build a distributed cold chain agent network, deploy edge computing devices at transport vehicles and cold storage nodes to form multiple agents, which exchange location, temperature, and equipment status data in real time through the MQTT protocol; Step S2: Dynamic risk assessment. Each agent collects data on ambient temperature, humidity, traffic conditions, and equipment energy consumption, inputs the data into a pre-trained LSTM model, and outputs three risk levels: low, medium, and high. Step S3: Anti-interference strategy execution. When the risk signal is low, the vehicle speed is adjusted to a preset energy-saving range and the cooling power is reduced by 10%-15%. When the risk signal is medium, an alternative transportation route is switched to based on real-time road conditions. The alternative route is provided by the AutoNavi Map API. When the risk signal is high, the emergency cold storage transfer protocol is activated, which includes inventory retrieval and reservation of the nearest cold storage; Step S4, multi-agent resource allocation: Calculate the contribution weights of food companies, transporters, and consumers based on the Shapley value algorithm, and allocate transportation resources. The calculation formula for the contribution weights is: Step S5, flexible resource scheduling: call on idle refrigerated truck resources in the society through the blockchain platform. The platform deploys smart contracts, and the contract code includes resource retrieval, token payment and breach of contract penalty logic.

2. The method for dynamically configuring food cold chain transportation resources according to claim 1, characterized in that: In step S1, the edge computing device includes a vehicle-mounted terminal and a cold storage control terminal. The vehicle-mounted terminal has a built-in GPS module and a temperature sensor, and the cold storage control terminal is connected to a door switch and a refrigeration compressor.

3. The method for dynamically configuring food cold chain transportation resources according to claim 1, characterized in that: In step S2, the training data of the LSTM model includes historical temperature fluctuation records, traffic delay events, and equipment failure logs. The input data is normalized to the interval [0, 1], and the output layer uses the Softmax function to classify the risk level.

4. The method for dynamically configuring food cold chain transportation resources according to claim 1, characterized in that: In step S3, the execution steps of the emergency cold storage transfer protocol include: Step S301: The agent sends an inventory query request to the surrounding cold storage, and the cold storage returns the available storage capacity and temperature range; In step S302, the agent selects a cold storage with a distance of ≤5km and a matching temperature, and completes the reservation through blockchain payment tokens.

5. The method for dynamically configuring food cold chain transportation resources according to claim 1, characterized in that: In step S4, when the contribution weight is calculated, a risk compensation coefficient λ is introduced. When the risk level is high, λ=1.5; when the risk level is medium, λ=1.2; when the risk level is low, λ=1.

0.

6. A food cold chain transportation resource dynamic configuration system, characterized in that: include: Multiple intelligent terminals, deployed in transport vehicles and cold storage, for executing the method according to any one of claims 1 to 5; Cloud server, which communicates with the intelligent terminal and stores the LSTM model and Shapley value algorithm program; Blockchain nodes run smart contracts to manage elastic resource scheduling.

7. A computer-readable storage medium storing a computer program, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images