Digital intelligent master control system for tobacco leaf purchasing

Abstract

The invention relates to the technical field of intelligent scheduling, and discloses a digital intelligent master control system for tobacco leaf purchasing, comprising: a tobacco grower information acquisition module for acquiring tobacco grower information of M tobacco growers; the equipment information acquisition module is used for acquiring equipment information of N tobacco processing equipment; the adaptation degree model building module is used for inputting equipment information of one tobacco processing equipment and tobacco grower information of one tobacco grower into an adaptation degree model, and an output value represents an adaptation degree value between the tobacco processing equipment and the tobacco grower; the tobacco grower distribution scheme generation module is used for generating a tobacco grower distribution scheme through a white whale optimization algorithm in combination with the fitness degree model; according to the method, the adaptation degree between the tobacco processing equipment and the tobacco growers is calculated through the neural network model, and the tobacco grower allocation scheme is automatically generated in combination with the white whale optimization algorithm, so that the utilization rate of equipment resources and human resources is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent scheduling technology, and more specifically, to a digital intelligent master control system for tobacco leaf purchasing. Background Art

[0002] In the traditional tobacco industry, the processing and preparation of tobacco leaves usually rely on labor-intensive operations, especially in the grading, screening and packaging of tobacco leaves. Managers often decide the number of tobacco farmers involved in the work and which equipment the farmers are assigned to based on past work experience and estimates of tobacco weight. However, the traditional manual management method mainly relies on personal work experience to judge the labor demand for tobacco leaf processing, which is subjective and uncertain. The lack of a scientific allocation strategy may lead to low utilization of equipment resources or human resources, or may lead to the failure to complete the preparation of tobacco products on time, affecting delivery time.

[0003] Therefore, there is an urgent need for a digital intelligent master control system for tobacco leaf purchasing to solve the above problems. Summary of the invention

[0004] The present invention provides a digital intelligent master control system for tobacco leaf purchasing, which solves the technical problem of XXXXX in the related technology.

[0005] The present invention provides a digital intelligent master control system for tobacco leaf purchasing, comprising:

[0006] A tobacco grower information collection module, which is used to collect tobacco grower information of M tobacco growers;

[0007] Tobacco farmer information includes: unique identifier, location, health status, years of work, assigned workload, and maximum workload;

[0008] An equipment information collection module, which is used to collect equipment information of N tobacco processing equipment;

[0009] Equipment information includes: unique identifier, location, current load, maximum processing capacity, energy consumption value and maintenance status;

[0010] A fitness model building module, which is used to input equipment information of a tobacco processing equipment and tobacco farmer information of a tobacco farmer into the fitness model, and the output value represents the fitness value between the tobacco processing equipment and the tobacco farmer;

[0011] The fit value between tobacco processing equipment and tobacco farmers ranges from 1 to 10;

[0012] A tobacco grower allocation plan generation module, which is used to generate a tobacco grower allocation plan by combining the White Whale optimization algorithm with the fitness model;

[0013] According to the tobacco farmer allocation plan, M tobacco farmers are assigned to N tobacco processing equipment.

[0014] Furthermore, M and N are both custom parameters.

[0015] Furthermore, the unique identifiers of tobacco processing equipment and tobacco farmers are represented by unique positive integer codes.

[0016] Furthermore, the locations of tobacco processing equipment and tobacco farmers are represented by longitude and latitude, the current load of the tobacco processing equipment is represented by the weight of tobacco that has been allocated, the maximum processing capacity is represented by the weight of tobacco processed per hour, the energy consumption value is represented by the power consumption value processed per hour, the maintenance status is represented by the number of historical maintenance times, the health status of tobacco farmers includes: healthy and unhealthy, represented by real number coding, the allocated workload is represented by the weight of tobacco that has been allocated, and the maximum workload is represented by the weight of tobacco processed per day.

[0017] Further, the fitness model includes: a first hidden layer, a second hidden layer and a classifier;

[0018] The first hidden layer includes: a first hidden unit and a second hidden unit;

[0019] The first hidden unit inputs equipment information of a tobacco processing equipment and outputs a first vector;

[0020] The second hidden unit inputs tobacco farmer information of a tobacco farmer and outputs a second vector;

[0021] The second hidden layer inputs the first vector and the second vector simultaneously and outputs the third vector;

[0022] The third vector is input into the classifier, and the classification space of the classifier represents the fitness value between the tobacco processing equipment and the tobacco farmers;

[0023] First, the similarity between the equipment information of a tobacco processing equipment and the tobacco farmer's information is calculated by Euclidean distance similarity, and then the similarity value is multiplied by 10 as a pre-label of a training sample for training the fitness model. Finally, the pre-label is manually reviewed as a sample label of a training sample until Q training samples are obtained, where Q is a custom parameter;

[0024] The calculation formula of Euclidean distance similarity sim is as follows:

[0025]

[0026] Where A represents the equipment information of a tobacco processing equipment, B represents the farmer information of a tobacco farmer, and Dis represents the Euclidean distance calculation function.

[0027] Furthermore, the calculation formula of the fitness model includes:

[0028] The calculation formula for the first hidden unit is as follows:

[0029] Vec1 = sigmoid(W1A+b1);

[0030] Wherein Vec1 represents the first vector, A represents the equipment information of a tobacco processing equipment, W1 represents the first weight parameter, b1 represents the first bias parameter, and sigmoid represents the sigmoid activation function;

[0031] The calculation formula for the second hidden unit is as follows:

[0032] Vec2 = sigmoid(W2B+b2);

[0033] Wherein Vec2 represents the second vector, B represents the tobacco farmer information of a tobacco farmer, W2 represents the second weight parameter, b2 represents the second bias parameter, and sigmoid represents the sigmoid activation function;

[0034] The calculation formula for the second hidden layer is as follows:

[0035]

[0036] Where Vec3 represents the third vector, represents the number of dimensions of the first vector, represents the number of dimensions of the second vector, W3 represents the third weight parameter, b3 represents the third bias parameter, and softmax represents the softmax activation function.

[0037] Furthermore, the tobacco farmer allocation plan is generated by combining the White Whale optimization algorithm with the fitness model, including the following steps:

[0038] Step S201, randomly generate an initialization population that meets the constraint conditions, each individual in the initialization population is assigned a positive integer number, and the current iteration number t is initialized to 0, and the number of individuals in the initialization population is U, where U is a custom parameter;

[0039] The constraints include: the same tobacco processing equipment can only be assigned to one tobacco farmer at a time; the tobacco allocation of a tobacco processing equipment cannot exceed the upper limit of tobacco allocation, where the upper limit of tobacco allocation is a custom parameter;

[0040] The individual is represented by a vector code, which includes N dimensions. The value of the nth dimension is represented by a real number code. The integer part of the real number code represents the unique identifier of the allocated tobacco farmer, and the decimal part represents the amount of tobacco allocated in kilograms, where 1≤n≤N;

[0041] Step S202, constructing an objective function according to the fitness model, and calculating the fitness value of each individual in the initialized population through the objective function;

[0042] The objective function is calculated as follows:

[0043]

[0044] Where Fit represents the fitness value of an individual, N represents the number of tobacco processing equipment, and match n The nth dimension of an individual corresponds to the equipment information of a tobacco processing equipment and the tobacco farmer's information input into the fitness model output, the fitness value between the tobacco processing equipment and the tobacco farmer, Dis n The Euclidean distance between a tobacco processing equipment and a tobacco farmer corresponding to the nth dimension of an individual;

[0045] Step S203, obtaining a balance factor through a balance factor calculation formula, when the balance factor is greater than a first threshold, proceeding to step S204, otherwise proceeding to step S205, the first threshold is a custom parameter;

[0046] Balance Factor B f The calculation formula is as follows:

[0047]

[0048] Where t represents the current number of iterations, T represents the maximum number of iterations, B0 represents a random number ranging from 0 to 1, and T is a custom parameter;

[0049] Step S204, changing the vector codes of individuals in the initialization population through the calculation formula in the exploration phase;

[0050] Step S205, changing the vector codes of individuals in the initialization population through the calculation formula in the development phase;

[0051] Step S206, obtaining the whale fall probability value through the whale fall probability calculation formula, when the whale fall probability value is greater than or equal to the balance factor, proceed to step S207, otherwise proceed to step S208;

[0052] Whale fall probability value W f The calculation formula is as follows:

[0053]

[0054] Where t represents the current number of iterations, and T represents the maximum number of iterations;

[0055] Step S207, changing the vector codes of individuals in the initialization population through the calculation formula of the whale fall stage;

[0056] Step S208, the current iteration number is increased by 1. When the iteration termination condition is met, the individual with the largest fitness value in the initialized population is output as the tobacco grower allocation plan, otherwise, the process returns to step S202 to continue execution.

[0057] Furthermore, the calculation formula of the exploration phase includes:

[0058]

[0059]

[0060] Where 1≤s≤U, 1≤p≤n, 1≤q≤m, and They represent the element values ​​of the pth row and qth column of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. Represents the element value of the p-th row and q-th column of the matrix encoding of a random individual in the initialized population corresponding to the t-th iteration number, r1 and r2 represent the first random number and the second random number respectively, and the value range of r1 and r2 is 0 to 1, even represents an even number, and odd represents an odd number.

[0061] Furthermore, the calculation formulas for the development phase include:

[0062]

[0063]

[0064]

[0065] C1=2r4(1-t / T);

[0066] Where 1≤s≤U, 1≤p≤n, and They represent the element values ​​of the first column to the qth column of the pth row of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. Indicates the element values ​​of the pth row, the 1st column to the qth column of the matrix encoding of a random individual in the initialized population corresponding to the tth iteration number, Indicates the element value of the first column to the qth column of the pth row corresponding to the individual with the largest fitness value in the initialized population at the tth iteration number, C1 represents the first random coefficient, L F represents the flight factor, r3 and r4 represent the third and fourth random numbers respectively, the value range of r3 and r4 is 0 to 1, μ and τ represent random numbers that conform to the normal distribution, β represents the default constant, which is assigned a value of 1.5, and σ represents the balance parameter.

[0067] Furthermore, the calculation formula for the whale fall stage includes:

[0068]

[0069] X step =(upLimit-downLimit)*exp(-C2t / T);

[0070] C2=2W f *U;

[0071] Where 1≤s≤U, 1≤p≤n, 1≤q≤m, and They represent the element values ​​of the pth row and qth column of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. represents the element value of the pth row and qth column of the matrix encoding of a random individual in the initialized population corresponding to the tth iteration number, r5, r6 and r7 represent the fifth, sixth and seventh random numbers respectively, and the value range of r5, r6 and r7 is 0 to 1, X step represents the step value, upLimit and downLimit represent the upper limit and lower limit respectively, which are assigned values ​​of 1 and k respectively, exp represents the natural exponential function, and C2 represents the second random coefficient.

[0072] The beneficial effect of the present invention is that the present invention calculates the compatibility between tobacco processing equipment and tobacco farmers through a neural network model, and automatically generates a tobacco farmer allocation plan in combination with the White Whale optimization algorithm, thereby improving the utilization rate of equipment resources and human resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a schematic diagram of a digital intelligent master control system for tobacco leaf purchasing of the present invention;

[0074] Figure 2 It is a flow chart of the present invention for generating a tobacco grower allocation plan by combining the White Whale optimization algorithm with a fitness model.

[0075] In the figure: tobacco grower information collection module 101, equipment information collection module 102, fitness model construction module 103, tobacco grower allocation plan generation module 104. DETAILED DESCRIPTION

[0076] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0077] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0078] like Figure 1-2 As shown, a digital intelligent master control system for tobacco leaf purchasing includes:

[0079] The tobacco grower information collection module 101 is used to collect tobacco grower information of M tobacco growers;

[0080] Tobacco farmer information includes: unique identifier, location, health status, years of work, assigned workload, and maximum workload;

[0081] The equipment information collection module 102 is used to collect equipment information of N tobacco processing equipment;

[0082] Equipment information includes: unique identifier, location, current load, maximum processing capacity, energy consumption value and maintenance status;

[0083] A fitness model building module 103, which is used to input equipment information of a tobacco processing equipment and tobacco farmer information of a tobacco farmer into the fitness model, and the output value represents the fitness value between the tobacco processing equipment and the tobacco farmer;

[0084] The fit value between tobacco processing equipment and tobacco farmers ranges from 1 to 10;

[0085] The tobacco grower allocation plan generation module 104 is used to generate a tobacco grower allocation plan by combining the White Whale optimization algorithm with the fitness model;

[0086] According to the tobacco farmer allocation plan, M tobacco farmers are assigned to N tobacco processing equipment.

[0087] In one embodiment of the present invention, both M and N are custom parameters, M can be set according to the total amount of tobacco that needs to be processed, and N can be set according to the operable tobacco processing equipment.

[0088] In one embodiment of the present invention, the unique identifiers of tobacco processing equipment and tobacco farmers are represented by unique positive integer codes. For example, the unique identifier of tobacco processing equipment ranges from 1 to N, and the unique identifier of tobacco farmers ranges from 1 to M.

[0089] In one embodiment of the present invention, the unique identifiers of the tobacco processing equipment and the tobacco growers may also be generated by a snowflake algorithm to ensure the uniqueness of the unique identifiers of the tobacco processing equipment and the tobacco growers.

[0090] In one embodiment of the present invention, the locations of tobacco processing equipment and tobacco farmers are represented by longitude and latitude, the current load of the tobacco processing equipment is represented by the weight of tobacco that has been allocated, the maximum processing capacity is represented by the weight of tobacco processed per hour, the energy consumption value is represented by the power consumption value processed per hour, and the maintenance status is represented by the number of historical maintenance times.

[0091] In one embodiment of the present invention, the health status of tobacco farmers includes: healthy and unhealthy, which are represented by real number codes. For example, the health status of healthy and unhealthy are represented by real numbers 1.1 and 1.2 respectively. The allocated workload is represented by the weight of tobacco that has been allocated, and the maximum workload is represented by the weight of tobacco processed every day.

[0092] In one embodiment of the present invention, the maintenance status can also be represented by the historical maintenance amount, and the health status of the tobacco farmer can also be represented by a positive integer code. For example, the health status of healthy and unhealthy is represented by positive integers 1 and 2 respectively, and the maximum workload of the tobacco farmer can also be represented by the weight of tobacco processed per hour.

[0093] In one embodiment of the present invention, the equipment information of tobacco processing equipment and the tobacco farmers' information can be preprocessed respectively. For example, missing values ​​can be filled by taking the average value. The maximum and minimum values ​​of the equipment information of the tobacco processing equipment can be extracted and the equipment information of the tobacco processing equipment can be normalized by the Min-Max method. Similarly, the maximum and minimum values ​​of the tobacco farmers' information can also be extracted and the extracted tobacco farmers' information can be normalized to reduce the numerical differences between the data, which can speed up the fitting speed of the fitness model to a certain extent.

[0094] In one implementation of the present invention, the fitness model includes: a first hidden layer, a second hidden layer and a classifier;

[0095] The first hidden layer includes: a first hidden unit and a second hidden unit;

[0096] The first hidden unit inputs equipment information of a tobacco processing equipment and outputs a first vector;

[0097] The second hidden unit inputs tobacco farmer information of a tobacco farmer and outputs a second vector;

[0098] The second hidden layer inputs the first vector and the second vector simultaneously and outputs the third vector;

[0099] The third vector is input into the classifier, and the classification space of the classifier represents the fitness value between the tobacco processing equipment and the tobacco farmers.

[0100] In one embodiment of the present invention, the similarity value between the equipment information of a tobacco processing equipment and the tobacco farmer's ...

[0101] In one embodiment of the present invention, the calculation formula of the Euclidean distance similarity sim is as follows:

[0102]

[0103] Where A represents the equipment information of a tobacco processing equipment, B represents the tobacco farmer information of a tobacco farmer, and Dis represents the Euclidean distance calculation function.

[0104] In one embodiment of the present invention, the calculation formula of the fitness model includes:

[0105] The calculation formula for the first hidden unit is as follows:

[0106] Vec1 = sigmoid(W1A+b1);

[0107] Wherein Vec1 represents the first vector, A represents the equipment information of a tobacco processing equipment, W1 represents the first weight parameter, b1 represents the first bias parameter, and sigmoid represents the sigmoid activation function;

[0108] The calculation formula for the second hidden unit is as follows:

[0109] Vec2 = sigmoid(W2B+b2);

[0110] Wherein Vec2 represents the second vector, B represents the tobacco farmer information of a tobacco farmer, W2 represents the second weight parameter, b2 represents the second bias parameter, and sigmoid represents the sigmoid activation function;

[0111] The calculation formula for the second hidden layer is as follows:

[0112]

[0113] Where Vec3 represents the third vector, represents the number of dimensions of the first vector, represents the number of dimensions of the second vector, W3 represents the third weight parameter, b3 represents the third bias parameter, and softmax represents the softmax activation function.

[0114] In one embodiment of the present invention, Figure 2 As shown, the tobacco farmer allocation plan is generated by combining the White Whale optimization algorithm with the fitness model, including the following steps:

[0115] Step S201, randomly generate an initialization population that meets the constraint conditions, each individual in the initialization population is assigned a positive integer number, and the current iteration number t is initialized to 0, the number of individuals in the initialization population is U, where U is a custom parameter, preferably, U is set to 50, and the positive integer number ranges from 1 to 50;

[0116] The constraints include: the same tobacco processing equipment can only be assigned to one tobacco farmer at a time; the tobacco allocation amount of a tobacco processing equipment cannot exceed the upper limit of tobacco allocation amount, where the upper limit of tobacco allocation amount is a custom parameter and can be manually set based on the equipment information of the tobacco processing equipment;

[0117] The individual is represented by a vector code, which includes N dimensions. The value of the nth dimension is represented by a real number code. The integer part of the real number code represents the unique identifier of the allocated tobacco farmer, and the decimal part represents the amount of tobacco allocated in kilograms, where 1≤n≤N;

[0118] For example, the value of the second dimension of the vector encoding of an individual is 3.22, which means that the tobacco processing equipment with a unique identifier of 2 is assigned a tobacco farmer with a unique identifier of 3, and the amount of tobacco allocated is 22 kg;

[0119] Step S202, constructing an objective function according to the fitness model, and calculating the fitness value of each individual in the initialized population through the objective function;

[0120] The objective function is calculated as follows:

[0121]

[0122] Where Fit represents the fitness value of an individual, N represents the number of tobacco processing equipment, and match n The nth dimension of an individual corresponds to the equipment information of a tobacco processing equipment and the tobacco farmer's information input into the fitness model output, the fitness value between the tobacco processing equipment and the tobacco farmer, Dis n The Euclidean distance between a tobacco processing equipment and a tobacco farmer corresponding to the nth dimension of an individual;

[0123] Step S203, obtaining a balance factor through a balance factor calculation formula, when the balance factor is greater than a first threshold, proceeding to step S204, otherwise proceeding to step S205, the first threshold is a custom parameter, preferably, the first threshold is set to 0.5;

[0124] Balance Factor B f The calculation formula is as follows:

[0125]

[0126] Where t represents the current number of iterations, T represents the maximum number of iterations, B0 represents a random number ranging from 0 to 1, and T is a custom parameter. Preferably, T is set to 100;

[0127] Step S204, changing the vector codes of individuals in the initialization population through the calculation formula in the exploration phase;

[0128] Step S205, changing the vector codes of individuals in the initialization population through the calculation formula in the development phase;

[0129] Step S206, obtaining the whale fall probability value through the whale fall probability calculation formula, when the whale fall probability value is greater than or equal to the balance factor, proceed to step S207, otherwise proceed to step S208;

[0130] Whale fall probability value W f The calculation formula is as follows:

[0131]

[0132] Where t represents the current number of iterations, and T represents the maximum number of iterations;

[0133] Step S207, changing the vector codes of individuals in the initialization population through the calculation formula of the whale fall stage;

[0134] Step S208, the current iteration number is increased by 1. When the iteration termination condition is met, the individual with the largest fitness value in the initialized population is output as the tobacco grower allocation plan, otherwise, the process returns to step S202 to continue execution.

[0135] In one implementation of the present invention, the calculation formula of the exploration phase includes:

[0136]

[0137]

[0138] Where 1≤s≤U, 1≤p≤n, 1≤q≤m, and They represent the element values ​​of the pth row and qth column of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. Represents the element value of the p-th row and q-th column of the matrix encoding of a random individual in the initialized population corresponding to the t-th iteration number, r1 and r2 represent the first random number and the second random number respectively, and the value range of r1 and r2 is 0 to 1, even represents an even number, and odd represents an odd number.

[0139] In one implementation of the present invention, the calculation formula in the development phase includes:

[0140]

[0141]

[0142]

[0143] C1=2r4(1-t / T);

[0144] Where 1≤s≤U, 1≤p≤n, and They represent the element values ​​of the first column to the qth column of the pth row of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. Indicates the element values ​​of the pth row, the 1st column to the qth column of the matrix encoding of a random individual in the initialized population corresponding to the tth iteration number, Indicates the element value of the first column to the qth column of the pth row corresponding to the individual with the largest fitness value in the initialized population at the tth iteration number, C1 represents the first random coefficient, L F represents the flight factor, r3 and r4 represent the third and fourth random numbers respectively, the value range of r3 and r4 is 0 to 1, μ and τ represent random numbers that conform to the normal distribution, β represents the default constant, which is assigned a value of 1.5, and σ represents the balance parameter.

[0145] In one embodiment of the present invention, the calculation formula of the whale fall stage includes:

[0146]

[0147] X step =(upLimit-downLimit)*exp(-C2t / T);

[0148] C2=2W f *U;

[0149] Where 1≤s≤U, 1≤p≤n, 1≤q≤m, and They represent the element values ​​of the pth row and qth column of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. represents the element value of the pth row and qth column of the matrix encoding of a random individual in the initialized population corresponding to the tth iteration number, r5, r6 and r7 represent the fifth, sixth and seventh random numbers respectively, and the value range of r5, r6 and r7 is 0 to 1, X step represents the step value, upLimit and downLimit represent the upper limit and lower limit respectively, which are assigned values ​​of 1 and k respectively, exp represents the natural exponential function, and C2 represents the second random coefficient.

[0150] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A digital intelligent master control system for tobacco leaf purchasing, characterized in that: include: A tobacco grower information collection module, which is used to collect tobacco grower information of M tobacco growers; Tobacco farmer information includes: unique identifier, location, health status, years of work, assigned workload, and maximum workload; An equipment information collection module, which is used to collect equipment information of N tobacco processing equipment; Equipment information includes: unique identifier, location, current load, maximum processing capacity, energy consumption value and maintenance status; A fitness model building module, which is used to input equipment information of a tobacco processing equipment and tobacco farmer information of a tobacco farmer into the fitness model, and the output value represents the fitness value between the tobacco processing equipment and the tobacco farmer; The fit value between tobacco processing equipment and tobacco farmers ranges from 1 to 10; A tobacco grower allocation plan generation module, which is used to generate a tobacco grower allocation plan by combining the White Whale optimization algorithm with the fitness model; According to the tobacco farmer allocation plan, M tobacco farmers are assigned to N tobacco processing equipment.

2. A digital intelligent master control system for tobacco leaf purchasing according to claim 1, characterized in that: M and N are custom parameters.

3. The digital intelligent master control system for tobacco leaf purchasing according to claim 1 is characterized in that: The unique identifiers of tobacco processing equipment and tobacco farmers are represented by unique positive integer codes.

4. A digital intelligent master control system for tobacco leaf purchasing according to claim 1, characterized in that: The locations of tobacco processing equipment and tobacco farmers are represented by longitude and latitude, the current load of tobacco processing equipment is represented by the weight of tobacco that has been allocated, the maximum processing capacity is represented by the weight of tobacco processed per hour, the energy consumption value is represented by the power consumption value per hour, the maintenance status is represented by the number of historical maintenance times, the health status of tobacco farmers includes: healthy and unhealthy, represented by real number coding, the allocated workload is represented by the weight of tobacco that has been allocated, and the maximum workload is represented by the weight of tobacco processed per day.

5. The digital intelligent master control system for tobacco leaf purchasing according to claim 1 is characterized in that: The fitness model includes: a first hidden layer, a second hidden layer and a classifier; The first hidden layer includes: a first hidden unit and a second hidden unit; The first hidden unit inputs equipment information of a tobacco processing equipment and outputs a first vector; The second hidden unit inputs tobacco farmer information of a tobacco farmer and outputs a second vector; The second hidden layer inputs the first vector and the second vector simultaneously and outputs the third vector; The third vector is input into the classifier, and the classification space of the classifier represents the fitness value between the tobacco processing equipment and the tobacco farmers; First, the similarity between the equipment information of a tobacco processing equipment and the tobacco farmer's information is calculated by Euclidean distance similarity, and then the similarity value is multiplied by 10 as a pre-label of a training sample for training the fitness model. Finally, the pre-label is manually reviewed as a sample label of a training sample until Q training samples are obtained, where Q is a custom parameter; The calculation formula of Euclidean distance similarity sim is as follows: Where A represents the equipment information of a tobacco processing equipment, B represents the tobacco farmer information of a tobacco farmer, and Dis represents the Euclidean distance calculation function.

6. A digital intelligent master control system for tobacco leaf purchasing according to claim 5, characterized in that: The calculation formula of the fitness model includes: The calculation formula for the first hidden unit is as follows: Vec1 = sigmoid(W1A+b1); Wherein Vec1 represents the first vector, A represents the equipment information of a tobacco processing equipment, W1 represents the first weight parameter, b1 represents the first bias parameter, and sigmoid represents the sigmoid activation function; The calculation formula for the second hidden unit is as follows: Vec2 = sigmoid(W2B+b2); Wherein Vec2 represents the second vector, B represents the tobacco farmer information of a tobacco farmer, W2 represents the second weight parameter, b2 represents the second bias parameter, and sigmoid represents the sigmoid activation function; The calculation formula for the second hidden layer is as follows: Where Vec3 represents the third vector, represents the number of dimensions of the first vector, represents the number of dimensions of the second vector, W3 represents the third weight parameter, b3 represents the third bias parameter, and softmax represents the softmax activation function.

7. The digital intelligent master control system for tobacco leaf purchasing according to claim 1 is characterized in that: The tobacco farmer allocation plan is generated by combining the White Whale optimization algorithm with the fitness model, including the following steps: Step S201, randomly generate an initialization population that meets the constraint conditions, each individual in the initialization population is assigned a positive integer number, and the current iteration number t is initialized to 0, and the number of individuals in the initialization population is U, where U is a custom parameter; The constraints include: the same tobacco processing equipment can only be assigned to one tobacco farmer at a time; the tobacco allocation of a tobacco processing equipment cannot exceed the upper limit of tobacco allocation, where the upper limit of tobacco allocation is a custom parameter; The individual is represented by a vector code, which includes N dimensions. The value of the nth dimension is represented by a real number code. The integer part of the real number code represents the unique identifier of the allocated tobacco farmer, and the decimal part represents the amount of tobacco allocated in kilograms, where 1≤n≤N; Step S202, constructing an objective function according to the fitness model, and calculating the fitness value of each individual in the initialized population through the objective function; The objective function is calculated as follows: Where Fit represents the fitness value of an individual, N represents the number of tobacco processing equipment, and match n The nth dimension of an individual corresponds to the equipment information of a tobacco processing equipment and the tobacco farmer's information input into the fitness model output, the fitness value between the tobacco processing equipment and the tobacco farmer, Dis n The Euclidean distance between a tobacco processing equipment and a tobacco farmer corresponding to the nth dimension of an individual; Step S203, obtaining a balance factor through a balance factor calculation formula, when the balance factor is greater than a first threshold, proceeding to step S204, otherwise proceeding to step S205, the first threshold is a custom parameter; Balance Factor B f The calculation formula is as follows: Where t represents the current number of iterations, T represents the maximum number of iterations, B0 represents a random number ranging from 0 to 1, and T is a custom parameter; Step S204, changing the vector codes of individuals in the initialization population through the calculation formula in the exploration phase; Step S205, changing the vector codes of individuals in the initialization population through the calculation formula in the development phase; Step S206, obtaining the whale fall probability value through the whale fall probability calculation formula, when the whale fall probability value is greater than or equal to the balance factor, proceed to step S207, otherwise proceed to step S208; Whale fall probability value W f The calculation formula is as follows: Where t represents the current number of iterations, and T represents the maximum number of iterations; Step S207, changing the vector codes of individuals in the initialization population through the calculation formula of the whale fall stage; Step S208, the current iteration number is increased by 1. When the iteration termination condition is met, the individual with the largest fitness value in the initialized population is output as the tobacco grower allocation plan, otherwise, the process returns to step S202 to continue execution.

8. A digital intelligent master control system for tobacco leaf purchasing according to claim 7, characterized in that: The calculation formula for the exploration phase includes: Where 1≤s≤U, 1≤p≤n, 1≤q≤m, and They represent the element values ​​of the pth row and qth column of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. Represents the element value of the p-th row and q-th column of the matrix encoding of a random individual in the initialized population corresponding to the t-th iteration number, r1 and r2 represent the first random number and the second random number respectively, and the value range of r1 and r2 is 0 to 1, even represents an even number, and odd represents an odd number.

9. A digital intelligent master control system for tobacco leaf purchasing according to claim 7, characterized in that: The calculation formulas for the development phase include: C1=2r4(1-t / T); Where 1≤s≤U, 1≤p≤n, and They represent the element values ​​of the first column to the qth column of the pth row of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. Indicates the element values ​​of the pth row, the 1st column to the qth column of the matrix encoding of a random individual in the initialized population corresponding to the tth iteration number, Indicates the element value of the first column to the qth column of the pth row corresponding to the individual with the largest fitness value in the initialized population at the tth iteration number, C1 represents the first random coefficient, L F represents the flight factor, r3 and r4 represent the third and fourth random numbers respectively, the value range of r3 and r4 is 0 to 1, μ and τ represent random numbers that conform to the normal distribution, β represents the default constant, which is assigned a value of 1.5, and σ represents the balance parameter.

10. A digital intelligent master control system for tobacco leaf purchasing according to claim 7, characterized in that: The calculation formula for the whale fall stage includes: X step =(upLimit-downLimit)*exp(-C2t / T); C2=2W f *U; Where 1≤s≤U, 1≤p≤n, 1≤q≤m, and They represent the element values ​​of the pth row and qth column of the matrix encoding of the individual with the positive integer number s in the initialized population corresponding to the t+1th iteration and the tth iteration, respectively. represents the element value of the pth row and qth column of the matrix encoding of a random individual in the initialized population corresponding to the tth iteration number, r5, r6 and r7 represent the fifth, sixth and seventh random numbers respectively, and the value range of r5, r6 and r7 is 0 to 1, X step represents the step value, upLimit and downLimit represent the upper limit and lower limit respectively, which are assigned values ​​of 1 and k respectively, exp represents the natural exponential function, and C2 represents the second random coefficient.

Images