Article state prediction method and system in cold-chain logistics based on artificial intelligence

Abstract

The invention relates to the technical field of logistics transportation management, and discloses an article state prediction method and system in cold-chain logistics based on artificial intelligence. According to the method, state data of articles in cold-chain logistics are collected through installed monitoring equipment, a sample set is constructed, meanwhile, the state data of the articles in the sample set are processed in a data processing mode, and after processing is completed, an article state prediction model is constructed based on the obtained state data of the processed articles. Meanwhile, current position information and position information of multiple destinations in cold-chain logistics are collected in real time, a path is planned based on a path planning algorithm, the path distance is calculated, the arrival time of an article is predicted, the obtained predicted arrival time of the article is input into the article state prediction model, and the state of the article is judged. And the accuracy of article state prediction in cold-chain logistics is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics and transportation management, and specifically to an artificial intelligence-based method and system for predicting the status of items in cold chain logistics. Background Art

[0002] With the rapid development of artificial intelligence technology, supply chain logistics companies are facing more and more problems, especially in the cold chain logistics transportation process. Cold chain transportation has been widely used to conveniently ensure the freshness and safety of food, but due to the lack of monitoring of the transportation process, the cold chain transportation system still has a series of problems.

[0003] The existing Chinese public invention patent CN117217645A, this method collects cold chain transportation data collected by sensors preset on cold chain turnover boxes; whenever passing through a routing node, the cold chain transportation data is transmitted to a logistics management system, and the location, status and usage of the cold chain turnover box are tracked and managed; when an abnormality is detected in the cold chain transportation data, the distribution route and plan of the goods are optimized; but since the transmission and analysis are only performed when passing through a routing node, the prediction and judgment capabilities are insufficient and have great limitations. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the shortcomings of the prior art, the present invention provides an artificial intelligence-based method and system for predicting the status of items in cold chain logistics, which has the advantages of accurate prediction, real-time, and high efficiency, and solves the problem of imperfect cold chain transportation system due to lack of monitoring of the transportation process.

[0006] (II) Technical solution

[0007] In order to solve the above technical problem that the cold chain transportation system is imperfect due to the lack of monitoring of the transportation process, the present invention provides the following technical solutions:

[0008] This embodiment discloses a method for predicting the status of items in cold chain logistics based on artificial intelligence, which specifically includes the following steps:

[0009] S1. Collect the status data of items in cold chain logistics through installed monitoring equipment, and build a sample set based on the collected status data of items;

[0010] S2. Processing the status data of the items in the sample set by a data processing method to obtain the status data of the processed items;

[0011] S3, constructing an item status prediction model based on the obtained processed item status data;

[0012] S4, collect the current location information and multiple destination location information in the cold chain logistics in real time, plan the path based on the path planning algorithm, calculate the path distance and predict the arrival time of the items;

[0013] S5. Input the predicted item arrival time into the item status prediction model to determine the item status.

[0014] The present invention collects status data of items in cold chain logistics through installed monitoring equipment and constructs a sample set. At the same time, the status data of the items in the sample set are processed by a data processing method. After the processing is completed, an item status prediction model is constructed based on the status data of the processed items. At the same time, the current position information and multiple destination location information in the cold chain logistics are collected in real time. The path is planned based on the path planning algorithm, the path distance is calculated, and the arrival time of the item is predicted. The predicted arrival time of the item is input into the item status prediction model to judge the status of the item, thereby improving the accuracy of the item status prediction in cold chain logistics.

[0015] Preferably, the collecting of the status data of the items in the cold chain logistics by the installed monitoring equipment includes: item data collection, transportation process speed collection, and item displacement photography collection;

[0016] Item data collection: collect data information on the temperature around the monitoring equipment and the temperature of items;

[0017] Object displacement shooting and collection: Set the shooting and collection interval, and collect information on the displacement of objects during cold chain logistics transportation based on the set shooting and collection interval;

[0018] Speed ​​collection during transportation: Real-time collection of vehicle speed and speed changes during cold chain transportation.

[0019] The present invention collects the status data of items in cold chain logistics by collecting item data, collecting the speed during transportation, and photographing and collecting the displacement of items, thereby ensuring the accuracy of the item status data collection.

[0020] Preferably, the step of processing the status data of the items in the sample set by a data processing method to obtain the processed status data of the items comprises the following steps:

[0021] The status data of the items in the sample set are standardized by using a data standardization method;

[0022] The standardized calculation formula is as follows:

[0023]

[0024] in, represents the state data of the items in the sample set after standardization, x represents the state data of the items in the sample set, μ represents the mean of the state data of the items in the sample set, and σ represents the standard deviation of the state data of the items in the sample set.

[0025] The present invention standardizes the status data of items in a sample set through a data standardization method, thereby reducing the problem of inaccurate prediction of item status in cold chain logistics due to data fluctuations and improving the reliability of item status data.

[0026] Preferably, the step of constructing an item status prediction model based on the obtained item status data after processing comprises the following steps:

[0027] S31, constructing a temperature prediction model based on the obtained state data of the processed object;

[0028] S32, constructing an object displacement prediction model based on the obtained processed object status data;

[0029] S33. Setting an object state prediction model includes a temperature prediction model and an object displacement prediction model.

[0030] Preferably, the step of constructing a temperature prediction model based on the obtained state data of the processed object comprises the following steps:

[0031] Building a temperature prediction model based on the temperature of the surrounding area of ​​the monitoring device and the temperature of the object in the obtained state data of the processed object;

[0032]

[0033] Among them, T(t) represents the temperature of the object at time t, H represents the temperature around the monitoring device, α represents the cooling coefficient, T0 represents the initial temperature of the object, and t0 represents the initial time.

[0034] Preferably, constructing the object displacement prediction model based on the obtained state data of the processed object comprises the following steps:

[0035] Building an item displacement prediction model based on the item displacement amount and the transportation process speed in the obtained state data of the processed item;

[0036] V t =V t-1 +at;

[0037]

[0038] Among them, V t represents the transport process speed at time t, V t-1 represents the transport process speed at t-1, a represents the acceleration, S trepresents the predicted displacement of the item at time t, S0 represents the displacement at the initial time, and ΔS represents the displacement change;

[0039] An acceleration threshold range is set. When the acceleration of the transport vehicle during transportation is not within the set acceleration threshold range, it indicates that the item has overturned. When the acceleration of the transport vehicle during transportation is within the set acceleration threshold range, it indicates that the item is intact.

[0040] The present invention constructs a temperature prediction model and an item displacement prediction model through the processed item status data, and judges the item status by setting a threshold range and collecting the acceleration of the transport vehicle in real time during transportation, thereby improving the accuracy of item status prediction in cold chain logistics.

[0041] Preferably, the real-time collection of current location information and multiple destination location information in cold chain logistics, planning a path based on a path planning algorithm, calculating the path distance, and predicting the arrival time of items include the following steps:

[0042] S41, initializing algorithm parameters and performing chromosome encoding on location information in cold chain logistics;

[0043] S42, obtaining an initial population based on a conservation algorithm;

[0044] The initial population is obtained based on the conservation algorithm. The specific steps are as follows:

[0045] S421, enable saturated transport calculation, record all routes from the current location information to multiple destination location information, and obtain a set of all saturated paths in cold chain logistics;

[0046] S422, summarizing the saturated path set in S421, merging any two paths in the saturated path set, calculating the savings value of the new path formed after the merger compared with the sum of the paths before the merger, and arranging the savings values ​​in descending order;

[0047] Savings value = total cost of the path before merging - total cost of the new path after merging;

[0048] Furthermore, the path total cost F calculation process includes the following steps:

[0049] Total path cost F = transportation cost f + transportation distance cost;

[0050] The calculation formula of transportation cost f is as follows;

[0051] f = cg;

[0052] Among them, f represents the fixed cost of the transport vehicle, c represents the fixed loss cost of the transport vehicle per kilometer, and g represents the number of kilometers traveled by the transport vehicle during the transportation of the goods;

[0053] The formula for calculating the transportation distance cost is as follows:

[0054] Set the coordinates of any two points in the path to: A(x1,y1), B(x1,y1)

[0055]

[0056] Among them, d AB Indicates the transportation distance between points A and B in the path;

[0057] S423, by setting the average speed of the transport vehicle, analyzing whether the two paths that can be merged meet the set node path completion time, if the completion time of the merged new path meets the set node path completion time, then merge the two paths into a new path and execute S424, if not, cancel the merger and repeat S423;

[0058] S424, deriving the chromosome code of the saturated path set in the cold chain logistics, setting the chromosome code at this time as the parent chromosome code, and setting the set of the parent chromosome codes as the initial population;

[0059] S43, calculating the fitness of the population through the fitness function;

[0060] Each monomer is set to represent a set of saturated paths, and each set of saturated paths represents the path of the transport vehicle after traversing multiple destination location information from the current location;

[0061] Set the fitness function:

[0062] Among them, Fit(z) represents the fitness of monomer z, and F represents the total cost of the monomer in the objective function;

[0063] S44, genetic operator selection operation;

[0064] Assume the probability of a monomer being selected is p(x i );

[0065]

[0066] Among them, I is the population size, Fit(z i ) is the fitness of the ith monomer;

[0067] S45, chromosome crossover operation;

[0068] A crossover point is randomly selected in the chromosome encoding of the saturated path set in cold chain logistics, and the chromosome encoding fragments between the crossover points are interacted to form a new chromosome encoding, thereby forming a new population;

[0069] Remove duplicate nodes in order and ensure that the chromosome coding length is consistent to form daughter chromosomes;

[0070] S46, chromosome mutation operation;

[0071] The mutation process is to replace the gene fragment of the parent chromosome to produce a chromosome code different from the parent; set the mutation probability p to mutate the parent chromosome code, thereby generating a new generation of daughter chromosomes;

[0072] The fitness of all offspring chromosomes is calculated. If the offspring is better than the parent, the parent chromosome code is replaced. Otherwise, if the parent is better, the mutation fails and the parent is selected to continue the iteration operation.

[0073] S47, output the optimal solution to the problem;

[0074] Set the maximum number of iterations Y and the iteration threshold N. Set the hybrid genetic algorithm to terminate when the chromosome code output after N iterations remains unchanged or when the hybrid genetic algorithm reaches the set maximum number of iterations, and use the chromosome code output for the last time as the optimal path.

[0075] S48, predicting the arrival time of the item based on the output optimal path and the set average speed;

[0076]

[0077] Among them, t 到 represents the predicted arrival time of the item, Σd represents the total distance of the optimal path, V 均 Indicates the set average speed.

[0078] The present invention collects the current location information and multiple destination location information in cold chain logistics in real time, records all possible paths through saturated transport calculation, optimizes all possible paths by setting fitness functions, outputs the optimal path, calculates the optimal path distance and predicts the arrival time of items, thereby ensuring accurate prediction of the status of items.

[0079] Preferably, the step of inputting the obtained predicted item arrival time into an item status prediction model to determine the item status comprises the following steps:

[0080] Set the item temperature threshold range, input the predicted item arrival time into the item status prediction model, and calculate the temperature of the item at the time of arrival. When the temperature of the item at the time of arrival is not within the set item temperature threshold range, it indicates that the item is damaged. When the temperature of the item at the time of arrival is within the set item temperature threshold range, it indicates that the item is intact.

[0081] The present invention sets a temperature threshold range for an item, inputs the predicted item arrival time into an item state prediction model, and determines the item state by calculating the predicted temperature, thereby improving the real-time performance of item state prediction.

[0082] The embodiment also discloses an artificial intelligence-based item status prediction system in cold chain logistics, comprising: a data collection module, a model building module, a path planning module, and an item status judgment module;

[0083] The data acquisition module is used to collect status data of items in cold chain logistics and transmit the collected status data to the model building module;

[0084] The model building module is used to build an item status prediction model according to the received status data;

[0085] The path planning module is used to collect the transport vehicle location information and the destination location information to perform path planning and calculate the arrival time;

[0086] The object status judgment module is used to set an object temperature threshold and an acceleration threshold to judge the object status.

[0087] (III) Beneficial effects

[0088] Compared with the prior art, the present invention provides an artificial intelligence-based method and system for predicting the status of items in cold chain logistics, which has the following beneficial effects:

[0089] 1. The invention collects the status data of items in the cold chain logistics through the installed monitoring equipment and constructs a sample set. At the same time, the status data of the items in the sample set are processed by data processing. After the processing is completed, an item status prediction model is constructed based on the status data of the processed items. At the same time, the current position information and multiple destination location information in the cold chain logistics are collected in real time. The path is planned based on the path planning algorithm, the path distance is calculated, and the arrival time of the item is predicted. The predicted arrival time of the item is input into the item status prediction model to judge the status of the item, thereby improving the accuracy of the item status prediction in the cold chain logistics.

[0090] 2. The invention collects the status data of items in cold chain logistics by collecting item data, collecting the speed of the transportation process, and photographing and collecting the displacement of items, thereby ensuring the accuracy of the item status data collection.

[0091] 3. The invention standardizes the status data of items in the sample set through a data standardization method, thereby reducing the problem of inaccurate prediction of item status in cold chain logistics due to data fluctuations and improving the reliability of item status data.

[0092] 4. The invention constructs a temperature prediction model and an item displacement prediction model through the status data of the processed items. At the same time, the item status is judged by setting a threshold range and collecting the acceleration of the transport vehicle in real time during transportation, thereby improving the accuracy of item status prediction in cold chain logistics.

[0093] 5. The invention collects the current location information and multiple destination location information in the cold chain logistics in real time, records all possible paths through saturated transport calculation, optimizes all possible paths by setting fitness functions, outputs the optimal path, calculates the optimal path distance and predicts the arrival time of items, thereby ensuring accurate prediction of the item status.

[0094] 6. The invention sets a temperature threshold range for the item, inputs the predicted item arrival time into the item status prediction model, and determines the item status by calculating the predicted temperature, thereby improving the real-time performance of item status prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 It is a schematic diagram of the structure of the item status prediction process in cold chain logistics of the present invention. DETAILED DESCRIPTION

[0096] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0097] Example 1

[0098] See also Figure 1 This embodiment discloses a method for predicting the status of items in cold chain logistics based on artificial intelligence, which specifically includes the following steps:

[0099] S1. Collect the status data of items in cold chain logistics through installed monitoring equipment, and build a sample set based on the collected status data of items;

[0100] The status data of items in cold chain logistics are collected through installed monitoring equipment, including: item data collection, transportation process speed collection, and item displacement photography collection;

[0101] Item data collection: collect data information on the temperature around the monitoring equipment and the temperature of items;

[0102] Object displacement shooting and collection: Set the shooting and collection interval, and collect information on the displacement of objects during cold chain logistics transportation based on the set shooting and collection interval;

[0103] Speed ​​collection during transportation: real-time collection of vehicle speed and speed changes during cold chain transportation;

[0104] S2. Processing the status data of the items in the sample set by a data processing method to obtain the status data of the processed items;

[0105] Processing the status data of the items in the sample set by a data processing method to obtain the status data of the processed items includes the following steps:

[0106] The status data of the items in the sample set are standardized by using a data standardization method;

[0107] The standardized calculation formula is as follows:

[0108]

[0109] in, represents the state data of the items in the sample set after standardization, x represents the state data of the items in the sample set, μ represents the mean of the state data of the items in the sample set, and σ represents the standard deviation of the state data of the items in the sample set;

[0110] S3, constructing an item status prediction model based on the obtained processed item status data;

[0111] Building an item status prediction model based on the obtained processed item status data includes the following steps:

[0112] S31, constructing a temperature prediction model based on the obtained state data of the processed object;

[0113] Building a temperature prediction model based on the temperature of the surrounding area of ​​the monitoring device and the temperature of the object in the obtained state data of the processed object;

[0114]

[0115] Where T(t) represents the temperature of the object at time t, H represents the temperature around the monitoring device, α represents the cooling coefficient, T0 represents the initial temperature of the object, and t0 represents the initial time;

[0116] S32, constructing an object displacement prediction model based on the obtained processed object status data;

[0117] Building an item displacement prediction model based on the item displacement amount and the transportation process speed in the obtained state data of the processed item;

[0118] V t =V t-1 +at;

[0119]

[0120] Among them, V t represents the transport process speed at time t, V t-1 represents the transport process speed at t-1, a represents the acceleration, S t represents the predicted displacement of the item at time t, S0 represents the displacement at the initial time, and ΔS represents the displacement change;

[0121] Furthermore, an acceleration threshold range is set. When the acceleration of the transport vehicle during transportation is not within the set acceleration threshold range, it indicates that the item has overturned. When the acceleration of the transport vehicle during transportation is within the set acceleration threshold range, it indicates that the item is intact.

[0122] S33, setting an item state prediction model including a temperature prediction model and an item displacement prediction model;

[0123] S4, collect the current location information and multiple destination location information in the cold chain logistics in real time, plan the path based on the path planning algorithm, calculate the path distance and predict the arrival time of the items;

[0124] Real-time collection of current location information and multiple destination location information in cold chain logistics, planning the path based on the path planning algorithm, calculating the path distance, and predicting the arrival time of items include the following steps:

[0125] S41, initializing algorithm parameters and performing chromosome encoding on location information in cold chain logistics;

[0126] S42, obtaining an initial population based on a conservation algorithm;

[0127] The initial population is obtained based on the conservation algorithm. The specific steps are as follows:

[0128] S421, enable saturated transport calculation, record all routes from the current location information to multiple destination location information, and obtain a set of all saturated paths in cold chain logistics;

[0129] S422, summarizing the saturated path set in S421, merging any two paths in the saturated path set, calculating the savings value of the new path formed after the merger compared with the sum of the paths before the merger, and arranging the savings values ​​in descending order;

[0130] Savings value = total cost of the path before merging - total cost of the new path after merging;

[0131] Furthermore, the path total cost F calculation process includes the following steps:

[0132] Total path cost F = transportation cost f + transportation distance cost;

[0133] The calculation formula of transportation cost f is as follows;

[0134] f = cg;

[0135] Among them, f represents the fixed cost of the transport vehicle, c represents the fixed loss cost of the transport vehicle per kilometer, and g represents the number of kilometers traveled by the transport vehicle during the transportation of the goods;

[0136] The formula for calculating the transportation distance cost is as follows:

[0137] Set the coordinates of any two points in the path to: A(x1,y1), B(x1,y1)

[0138]

[0139] Among them, d AB Indicates the transportation distance between points A and B in the path;

[0140] S423, by setting the average speed of the transport vehicle, analyzing whether the two paths that can be merged meet the set node path completion time, if the completion time of the merged new path meets the set node path completion time, then merge the two paths into a new path and execute S424, if not, cancel the merger and repeat S423;

[0141] S424, deriving the chromosome code of the saturated path set in the cold chain logistics, setting the chromosome code at this time as the parent chromosome code, and setting the set of the parent chromosome codes as the initial population;

[0142] S43, calculating the fitness of the population through the fitness function;

[0143] Each monomer is set to represent a set of saturated paths, and each set of saturated paths represents the path of the transport vehicle after traversing multiple destination location information from the current location;

[0144] Set the fitness function:

[0145] Among them, Fit(z) represents the fitness of monomer z, and F represents the total cost of the monomer in the objective function;

[0146] S44, genetic operator selection operation;

[0147] Assume the probability of a monomer being selected is p(x i );

[0148]

[0149] Among them, I is the population size, Fit(z i ) is the fitness of the ith monomer;

[0150] S45, chromosome crossover operation;

[0151] A crossover point is randomly selected in the chromosome encoding of the saturated path set in cold chain logistics, and the chromosome encoding fragments between the crossover points are interacted to form a new chromosome encoding, thereby forming a new population;

[0152] Furthermore, duplicate nodes are removed in order, and the chromosome coding length is ensured to be consistent to form daughter chromosomes;

[0153] S46, chromosome mutation operation;

[0154] The mutation process is to replace the gene fragment of the parent chromosome to produce a chromosome code different from the parent; set the mutation probability p to mutate the parent chromosome code, thereby generating a new generation of daughter chromosomes;

[0155] Furthermore, the fitness of all offspring chromosomes is calculated. If the offspring is better than the parent, the parent chromosome code is replaced. Otherwise, if the parent is better, it means that the mutation has failed, and the parent is selected to continue the iteration operation.

[0156] S47, output the optimal solution to the problem;

[0157] Set the maximum number of iterations Y and the iteration threshold N. Set the hybrid genetic algorithm to terminate when the chromosome code output after N iterations remains unchanged or when the hybrid genetic algorithm reaches the set maximum number of iterations, and use the chromosome code output for the last time as the optimal path.

[0158] S48, predicting the arrival time of the item based on the output optimal path and the set average speed;

[0159]

[0160] Among them, t 到 represents the predicted arrival time of the item, Σd represents the total distance of the optimal path, V 均 Indicates the set average speed;

[0161] S5. Input the predicted item arrival time into the item status prediction model to determine the item status;

[0162] Set the item temperature threshold range, input the predicted item arrival time into the item status prediction model, calculate the temperature of the item at the time of arrival, when the temperature of the item at the time of arrival is not within the set item temperature threshold range, it indicates that the item is damaged, when the temperature of the item at the time of arrival is within the set item temperature threshold range, it indicates that the item is intact;

[0163] Example 2

[0164] This embodiment also discloses an artificial intelligence-based item status prediction system in cold chain logistics, including: a data acquisition module, a model building module, a path planning module, and an item status judgment module;

[0165] The data acquisition module is used to collect status data of items in cold chain logistics and transmit the collected status data to the model building module;

[0166] The model building module is used to build an item status prediction model according to the received status data;

[0167] The path planning module is used to collect the transport vehicle location information and the destination location information for path planning and calculate the arrival time;

[0168] The object status judgment module is used to set an object temperature threshold and an acceleration threshold to judge the object status.

[0169] Although 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 the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for predicting the status of items in cold chain logistics based on artificial intelligence, characterized in that: The following steps are involved: S1. Collect the status data of items in cold chain logistics through installed monitoring equipment, and build a sample set based on the collected status data of items; S2. Processing the status data of the items in the sample set by a data processing method to obtain the status data of the processed items; S3, constructing an item status prediction model based on the obtained processed item status data; S4, collect the current location information and multiple destination location information in the cold chain logistics in real time, plan the path based on the path planning algorithm, calculate the path distance and predict the arrival time of the items; S5. Input the predicted item arrival time into the item status prediction model to determine the item status.

2. According to claim 1, a method for predicting the state of items in cold chain logistics based on artificial intelligence is characterized in that: The state data of items in the cold chain logistics are collected by installing monitoring equipment, including: item data collection, transportation process speed collection, and item displacement photography collection; Item data collection: collect data information on the temperature around the monitoring equipment and the temperature of items; Object displacement shooting and collection: Set the shooting and collection interval, and collect information on the displacement of objects during cold chain logistics transportation based on the set shooting and collection interval; Speed ​​collection during transportation: Real-time collection of vehicle speed and speed changes during cold chain transportation.

3. The method for predicting the state of items in cold chain logistics based on artificial intelligence according to claim 1 is characterized in that: The state data of the items in the sample set are processed by the data processing method to obtain the state data of the items after processing. The following steps are involved: The status data of the items in the sample set are standardized by using a data standardization method; The standardized calculation formula is as follows: x＝ x-μ ； σ Among them, x represents the state data of the items in the sample set after standardization, x represents the state data of the items in the sample set, μ represents the mean of the state data of the items in the sample set, and σ represents the standard deviation of the state data of the items in the sample set.

4. The method for predicting the state of items in cold chain logistics based on artificial intelligence according to claim 1 is characterized in that: The method of constructing an item status prediction model based on the obtained processed item status data comprises the following steps: S31, constructing a temperature prediction model based on the obtained state data of the processed object; S32, constructing an object displacement prediction model based on the obtained processed object status data; S33. Setting an object state prediction model includes a temperature prediction model and an object displacement prediction model.

5. The method for predicting the state of items in cold chain logistics based on artificial intelligence according to claim 4 is characterized in that: The step of constructing a temperature prediction model based on the obtained state data of the processed object comprises the following steps: Building a temperature prediction model based on the temperature of the surrounding area of ​​the monitoring device and the temperature of the object in the obtained state data of the processed object; Among them, T(t) represents the temperature of the object at time t, H represents the temperature around the monitoring device, α represents the cooling coefficient, T0 represents the initial temperature of the object, and t0 represents the initial time.

6. The method for predicting the state of items in cold chain logistics based on artificial intelligence according to claim 4 is characterized in that: The method of constructing the object displacement prediction model based on the obtained processed object status data comprises the following steps: Building an item displacement prediction model based on the item displacement amount and the transportation process speed in the obtained state data of the processed item; V t =V t-1 +at; Among them, V t represents the transport process speed at time t, V t-1 represents the transport process speed at t-1, a represents the acceleration, S t represents the predicted displacement of the item at time t, S0 represents the displacement at the initial time, and ΔS represents the displacement change; An acceleration threshold range is set. When the acceleration of the transport vehicle during transportation is not within the set acceleration threshold range, it indicates that the item has overturned. When the acceleration of the transport vehicle during transportation is within the set acceleration threshold range, it indicates that the item is intact.

7. The method for predicting the state of items in cold chain logistics based on artificial intelligence according to claim 1 is characterized in that: The real-time collection of current location information and multiple destination location information in cold chain logistics, planning a path based on a path planning algorithm, calculating the path distance, and predicting the arrival time of items include the following steps: S41, initializing algorithm parameters and performing chromosome encoding on location information in cold chain logistics; S42, obtaining an initial population based on a conservation algorithm; The initial population is obtained based on the conservation algorithm. The specific steps are as follows: S421, enable saturated transport calculation, record all routes from the current location information to multiple destination location information, and obtain a set of all saturated paths in cold chain logistics; S422, summarizing the saturated path set in S421, merging any two paths in the saturated path set, calculating the savings value of the new path formed after the merger compared with the sum of the paths before the merger, and arranging the savings values ​​in descending order; Savings value = total cost of the path before merging - total cost of the new path after merging; Furthermore, the path total cost F calculation process includes the following steps: Total path cost F = transportation cost f + transportation distance cost; The calculation formula of transportation cost f is as follows; f = cg; Among them, f represents the fixed cost of the transport vehicle, c represents the fixed loss cost of the transport vehicle per kilometer, and g represents the number of kilometers traveled by the transport vehicle during the transportation of the goods; The formula for calculating the transportation distance cost is as follows: Set the coordinates of any two points in the path to: A(x1,y1), B(x1,y1) Among them, d AB Indicates the transportation distance between points A and B in the path; S423, by setting the average speed of the transport vehicle, analyzing whether the two paths that can be merged meet the set node path completion time, if the completion time of the merged new path meets the set node path completion time, then merge the two paths into a new path and execute S424, if not, cancel the merger and repeat S423; S424, deriving the chromosome code of the saturated path set in the cold chain logistics, setting the chromosome code at this time as the parent chromosome code, and setting the set of the parent chromosome codes as the initial population; S43, calculating the fitness of the population through the fitness function; Each monomer is set to represent a set of saturated paths, and each set of saturated paths represents the path of the transport vehicle after traversing multiple destination location information from the current location; Set the fitness function: Among them, Fit(z) represents the fitness of monomer z, and F represents the total cost of the monomer in the objective function; S44, genetic operator selection operation; Assume the probability of a monomer being selected is p(x i ); Among them, I is the population size, Fit(z i ) is the fitness of the ith monomer; S45, chromosome crossover operation; A crossover point is randomly selected in the chromosome encoding of the saturated path set in cold chain logistics, and the chromosome encoding fragments between the crossover points are interacted to form a new chromosome encoding, thereby forming a new population; Remove duplicate nodes in order and ensure that the chromosome coding length is consistent to form daughter chromosomes; S46, chromosome mutation operation; The mutation process is to replace the gene fragment of the parent chromosome to produce a chromosome code different from the parent; set the mutation probability p to mutate the parent chromosome code, thereby generating a new generation of daughter chromosomes; The fitness of all offspring chromosomes is calculated. If the offspring is better than the parent, the parent chromosome code is replaced. Otherwise, if the parent is better, the mutation fails and the parent is selected to continue the iteration operation. S47, output the optimal solution to the problem; Set the maximum number of iterations Y and the iteration threshold N. Set the hybrid genetic algorithm to terminate when the chromosome code output after N iterations remains unchanged or when the hybrid genetic algorithm reaches the set maximum number of iterations, and use the chromosome code output for the last time as the optimal path. S48, predicting the arrival time of the item based on the output optimal path and the set average speed; Among them, t 到 represents the predicted arrival time of the item, ∑d represents the total distance of the optimal path, V 均 Indicates the set average speed.

8. The method for predicting the state of items in cold chain logistics based on artificial intelligence according to claim 1 is characterized in that: The obtained predicted item arrival time is input into the item status prediction model, and the item status is determined, comprising the following steps: Set the item temperature threshold range, input the predicted item arrival time into the item status prediction model, and calculate the temperature of the item at the time of arrival. When the temperature of the item at the time of arrival is not within the set item temperature threshold range, it indicates that the item is damaged. When the temperature of the item at the time of arrival is within the set item temperature threshold range, it indicates that the item is intact.

9. A system for implementing the method for predicting the state of items in cold chain logistics based on artificial intelligence according to any one of claims 1 to 8, characterized in that: include: Data collection module, model building module, path planning module and item status judgment module; The data acquisition module is used to collect status data of items in cold chain logistics and transmit the collected status data to the model building module; The model building module is used to build an item status prediction model according to the received status data; The path planning module is used to collect the transport vehicle location information and the destination location information to perform path planning and calculate the arrival time; The object status judgment module is used to set an object temperature threshold and an acceleration threshold to judge the object status.

Images