CN114781887A - Warehouse management intelligent early warning method combined with current environment change - Google Patents

Warehouse management intelligent early warning method combined with current environment change Download PDF

Info

Publication number
CN114781887A
CN114781887A CN202210454908.4A CN202210454908A CN114781887A CN 114781887 A CN114781887 A CN 114781887A CN 202210454908 A CN202210454908 A CN 202210454908A CN 114781887 A CN114781887 A CN 114781887A
Authority
CN
China
Prior art keywords
sensor
warehouse
index data
warehouse management
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210454908.4A
Other languages
Chinese (zh)
Inventor
朱浩
童浩
成林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Zhihui Qice Technology Co ltd
Original Assignee
Shenzhen Zhihui Qice Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Zhihui Qice Technology Co ltd filed Critical Shenzhen Zhihui Qice Technology Co ltd
Priority to CN202210454908.4A priority Critical patent/CN114781887A/en
Publication of CN114781887A publication Critical patent/CN114781887A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
    • G01D21/00—Measuring or testing not otherwise provided for
    • G01D21/02—Measuring two or more variables by means not covered by a single other subclass
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B19/00—Alarms responsive to two or more different undesired or abnormal conditions, e.g. burglary and fire, abnormal temperature and abnormal rate of flow

Abstract

The invention relates to the technical field of warehouse storage management, and discloses an intelligent early warning method for warehouse management in combination with current environmental changes, which comprises the following steps: constructing a deployment position optimization objective function of the sensing equipment in the warehouse; optimizing and solving a sensor deployment position optimization objective function by using an improved pigeon group optimization algorithm; deploying sensors according to the sensor positions obtained by solving, and acquiring environmental index data of different sensors; performing data dimension reduction processing on the preprocessed environmental index data to obtain an index data vector; constructing an intelligent warehouse management decision model of a deep neural network; performing parameter optimization on the deep neural network bin management decision model by using an improved grey wolf optimization algorithm; and inputting the index data vector into a deep neural network warehouse management decision model, and outputting an optimal warehouse management decision strategy by the model. The method of the invention realizes intelligent decision early warning of warehouse management by constructing a deep neural network intelligent warehouse management decision model by utilizing environmental information.

Description

Warehouse management intelligent early warning method combined with current environment change
Technical Field
The invention relates to the technical field of warehouse storage management, in particular to an intelligent early warning method for warehouse management in combination with current environmental changes.
Background
At present, the traditional warehouse management is mainly manual, for example, when the wind power is high or it rains, the goods stored at the edge of the warehouse need to be manually recovered to the deep position of the warehouse, and when the sun is sunny and the wind power is low, the goods needing to be aired need to be manually transported to the outside for airing. On the one hand, the manual processing mode greatly wastes manpower and material resources, and the efficiency is extremely low. On the other hand, because the temperature and humidity at different positions in the warehouse are also greatly different, if the neural network model is used for modeling control, the dimension of model processing is too high, and a large amount of computing resources are consumed for model training.
Disclosure of Invention
In view of the above, the invention provides an intelligent early warning method for warehouse management in combination with current environmental changes, and aims to (1) deploy multiple types of sensing equipment in a warehouse, obtain environmental information of different positions of the warehouse, and construct an intelligent warehouse management decision model of a deep neural network by using the environmental information, so as to realize intelligent decision early warning of warehouse management and avoid excessive consumption of manpower and material resources; (2) and the training steps of the model are optimized by adopting various heuristic algorithms, so that the training time of the model is shortened, the resource consumption is reduced, and the timeliness of warehouse management is guaranteed.
The invention provides an intelligent early warning method for warehouse management in combination with current environmental changes, which comprises the following steps:
s1: constructing a deployment position optimization objective function of sensing equipment in a warehouse, wherein the sensing equipment comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind sensor;
s2: carrying out rapid optimization solution on the sensor deployment position optimization objective function by using an improved pigeon flock optimization algorithm to determine the sensor deployment position;
s3: deploying sensors according to the sensor positions obtained by solving, acquiring environmental index data of different sensors, and preprocessing the acquired environmental index data;
s4: carrying out data dimension reduction processing on the preprocessed environmental index data to obtain an index data vector;
s5: constructing a deep neural network intelligent warehouse management decision model, wherein the model takes an index data vector as input and takes a warehouse management decision strategy as output;
s6: optimizing the deep neural network warehouse management decision model by using an improved grey wolf optimization algorithm to obtain optimized model parameters;
s7: and inputting the index data vector into a deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy based on environment change self-adaptive adjustment by the model, and performing intelligent management on stored goods in the warehouse according to the optimal warehouse management decision strategy.
As a further improvement of the method:
optionally, in the step S1, an optimization objective function of a deployment location of a sensor device in a warehouse is constructed, where the sensor device includes a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor, and a wind sensor, and includes:
constructing an optimization objective function of the deployment position of the sensing equipment in the warehouse, wherein the constructed objective function is as follows:
Figure BDA0003620123070000011
wherein:
a is a sensing device capable of being arranged in the warehouseIf the prepared position S belongs to A, S represents the number of the positions, then
Figure BDA0003620123070000012
The energy control strategy of the sensing device representing the location s,
Figure BDA0003620123070000013
the sensing device representing location s remains dormant at time t,
Figure BDA0003620123070000014
the sensing device representing location s performs context awareness at time t;
(σs,i)2a measurement variance of the sensor system at the location s, i denotes a type of sensor, i ═ 1,2,3,4,5}, i ═ 1 denotes a temperature sensor, i ═ 2 denotes a humidity sensor, i ═ 3 denotes a carbon dioxide concentration sensor, i ═ 4 denotes an illumination intensity sensor, and i ═ 5 denotes a wind sensor;
t is a time window for the sensor to execute environment sensing;
μs,tfor measuring accurate values at a sensing device of type i at a position s, μiThe average value of the measurements of the sensing devices of the same type in the warehouse is obtained;
the constraint conditions of the objective function are as follows:
Figure BDA0003620123070000021
N≤S
Figure BDA0003620123070000022
wherein:
n represents the total number of the sensing equipment to be deployed;
e represents the number of context aware operations that the sensing device can perform before it is depleted of energy.
Optionally, in the step S2, performing a fast optimization solution on the sensor device deployment position optimization objective function by using an improved pigeon swarm optimization algorithm to obtain the deployment position of the sensor, where the method includes:
carrying out rapid optimization solution on the optimal objective function of the deployment position of the sensing equipment by using an improved pigeon group optimization algorithm, wherein the flow of the improved pigeon group optimization algorithm is as follows:
1) constructing an S-dimensional pigeon flock position solving space, wherein S represents the total number of positions where sensing equipment can be configured in a warehouse;
2) initializing n1Only pigeons and initialize the position and speed of each pigeon, then the position and speed of any jth pigeon is:
xj＝[xj1,xj2,…,xjS]
vj＝[vj1,vj2,…,vjS]
wherein:
xjfor any jth pigeon position vector, each position vector corresponds to a sensing equipment deployment position, xjSIndicates the type of sensor, x, deployed at the S-th warehouse sensing device deployment locationjS∈[0,5]Wherein x isjS0 denotes that no sensor is deployed for this position, xjSX denotes the location at which the temperature sensor is disposedjS2 denotes that the location is equipped with a humidity sensor, xjS3 denotes the location at which the carbon dioxide concentration sensor is deployed, xjS4 denotes that the position is deployed with an illumination intensity sensor, xjS5, representing that the wind sensor is deployed at the position, inputting the position vector into an optimization objective function of the deployment position of the sensing equipment, wherein the value of the optimization objective function is the fitness value of the jth pigeon;
vjis the velocity vector of any jth pigeon;
3) setting the iteration number of the current algorithm as ngAnd n isgInitializing to 0, and setting the number of termination iterations of the first-stage algorithm to ng1The number of times of the termination iteration of the second-stage algorithm is ng2；
4) Judging whether the iteration number of the current algorithm meets ng≥ng1If it is not fullIf yes, the positions and speeds of all pigeons are updated by the following formula:
Figure BDA0003620123070000023
Figure BDA0003620123070000024
Figure BDA0003620123070000025
wherein:
Figure BDA0003620123070000026
is at the n-thgAfter the iteration, the pigeon position vector with the minimum fitness value in the population;
r represents a pigeon position solving space, the flight direction of the pigeons is controlled, and e is a natural constant;
Figure BDA0003620123070000027
representing a differential sequence;
and let n beg＝ng+1, repeating the step;
if n is satisfiedg≥ng1Then let n beg＝ng+1, entering the next step;
5) calculating the order ngAfter iteration, the fitness function of each pigeon is deleted, 1/3 pigeons with the highest fitness function in the pigeon group are deleted, after deletion, the positions and the speeds of the rest pigeons are updated, and n is madeg＝ng+1, repeat this step until ng≥ng2Or only within two pigeons remain, calculating the adaptability value of the remaining pigeons at the moment, and taking the position vector corresponding to the pigeon with the minimum adaptability value as the solving result of the optimal objective function of the deployment position of the sensing equipment.
Optionally, the deploying the sensor according to the sensor deployment position obtained by the solving in the step S3 includes:
obtaining the optimal position vector x according to the solution*＝[x*1,x*2,…,x*S]Respectively traversing the optimal position vector x obtained by solving*And a position set A in the warehouse, where the sensing equipment can be arranged, and different types of sensors are deployed at specified positions of the optimal position vector, wherein x*SIndicates the type of sensor, x, that needs to be deployed at the S-th warehouse sensor equipment deployment location*S∈[0,5]，x*S0 denotes that no sensor is deployed for this position, x*SX denotes the location at which the temperature sensor is disposed*S2 denotes that the location is equipped with a humidity sensor, x*S3 denotes that the carbon dioxide concentration sensor is deployed at the position, x*S4 denotes that the position is deployed with an illumination intensity sensor, x*SThe position at which the wind sensor is deployed is denoted by 5.
Optionally, the step S3 is to collect index data of different sensors, and pre-process the collected index data, including:
all sensors are started, the sensors start sensing of the surrounding environment, environment index data near the sensors are collected, in detail, different types of sensors collect different environment index data, temperature sensors collect temperature time sequence data near the sensors, humidity sensors collect temperature time sequence data near the sensors, carbon dioxide concentration sensors collect carbon dioxide concentration time sequence data near the sensors, illumination intensity sensors collect illumination intensity time sequence data near the sensors, wind power sensors collect wind power intensity time sequence data near the sensors, and the collected environment index data are combined as follows:
{ys,i(q)|s∈A,i∈[0,5],q∈[tl,th]}
wherein:
i represents a sensor type located at the s-th warehouse sensing equipment deployment location;
s represents the s-th warehouse sensing equipment deployment location;
a represents a set of locations within the warehouse where sensing devices may be disposed;
[tl,th]time sequence interval for data acquisition of sensor, tlFor the moment at which the sensor starts data acquisition, thThe moment when the sensor finishes data acquisition;
ys,i(q) is time series data of an environment index i at the deployment position of the s-th warehouse sensing equipment, wherein i is 1 to represent ys,i(q) is temperature index time series data, i-2 represents ys,i(q) is humidity index time series data, i-3 represents ys,i(q) is carbon dioxide concentration index time series data, and i-4 represents ys,i(q) is time series data of the light intensity index, and i-5 represents ys,i(q) is wind strength indicator time series data;
data y of any environment index i in the collected environment index data set at any timei(t') carrying out normalization processing, wherein the formula of the normalization processing is as follows:
Figure BDA0003620123070000031
wherein:
maxithe maximum data value of the environment index i in the environment index data set;
miniis the minimum data value of the environment index i in the environment index data set;
(yi(t '))' is yi(t') normalizing the processed values;
all data in the environment index data set are normalized, the data collected by the sensor are mapped into the range of [ -1,1], and the environment index data set after normalization is as follows:
{y′s,i(q)|s∈A,i∈[0,5],q∈[tl,th]}
wherein:
y′s,i(q) is normalization processing ys,iTime series data after (q).
Optionally, in the step S4, performing data dimension reduction processing on the preprocessed index data to obtain an index data vector, where the method includes:
performing data dimension reduction processing on the preprocessed index data, wherein the data dimension reduction processing flow comprises the following steps:
converting the environment index data set after normalization into a vector form, wherein each dimension vector is time sequence data after normalization, and the length of each dimension vector is the same, so as to obtain an N-dimension vector matrix, wherein N is the total number of deployed sensors; in the present example, y's,i(q) as a one-dimensional vector;
calculating a covariance matrix of the N-dimensional vector matrix, wherein the covariance matrix is in a form of sigma;
and solving the eigenvalue and the eigenvector of the covariance matrix, calculating the contribution rate of each eigenvalue, and selecting the eigenvector of which the cumulative contribution rate reaches 90% as the index data vector after the dimension reduction processing.
Optionally, the constructing a deep neural network intelligent warehouse management decision model in the step S5, where the deep neural network intelligent warehouse management decision model takes an index data vector as input and a warehouse management decision strategy as output, includes:
constructing a deep neural network intelligent warehouse management decision model, wherein the deep neural network intelligent warehouse management decision model comprises an input layer, three convolutional layers and a full connection layer;
in the embodiment of the invention, the deep neural network intelligent warehouse management decision model receives the index data vectors, performs convolution operation on the index data vectors to obtain a convolution characteristic diagram of the index data vectors, inputs the convolution characteristic diagram into the full-connection layer, and outputs the full-connection layer to be matched with the convolution characteristic diagram to a warehouse management decision strategy.
Optionally, the S6 step of performing parameter optimization on the deep neural network bin decision model by using an improved grey wolf optimization algorithm, including:
the method for training the deep neural network warehouse management decision model by using the improved wolf optimization algorithm until the deep neural network warehouse management decision model with the optimized parameters is obtained after the training requirements are met comprises the following steps:
1) collecting warehouse management historical data, carrying out normalization and dimension reduction processing on the collected warehouse management historical data, and taking a processing result as a training data set, wherein the training data set comprises index data vectors and corresponding warehouse management decision strategies;
2) constructing a parameter optimization objective function of a deep neural network bin management decision model:
Figure BDA0003620123070000041
wherein:
m is the total number of index data vectors in the training data set;
Ykthe binary coding result of the bin management decision strategy corresponding to the kth group of index data vectors in the training data set is obtained;
Figure BDA0003620123070000042
a binary coding bin management decision strategy for outputting the k group index data vector for the deep neural network bin management decision model;
theta is a model parameter to be optimized;
in the embodiment of the invention, the model parameters to be optimized in the deep neural network bin management decision model comprise weight vectors and offset of 3 convolutional layers;
3) initializing n2The position dimension of each wolf is 6 dimensions, and the position coordinate of the g-th wolf is as follows:
hg＝[hg1,hg2,…,hg6]
wherein:
[hg1,hg2,hg3]sequentially and respectively weighting vectors of 3 layers of convolution layers;
[hg4,hg5,hg6]sequentially and respectively the offset of 3 convolution layers;
the initialization formula of the gray wolf position coordinate is as follows:
Figure BDA0003620123070000043
the position coordinate of each wolf corresponds to a deep neural network warehouse decision model, and the position coordinate h of each wolf is usedgThe implementation is a deep neural network warehouse decision model, a training data set is input into the implementation model, and the value of a parameter optimization objective function is used as the optimized value of the gray wolf position coordinate;
4) setting the iteration number of the current algorithm as ndAnd n isdInitializing to 0, and setting the maximum algorithm iteration times to Max;
5) calculating the optimized values of all the gray wolf position coordinates in the iterative process of the current round of algorithm;
6) the gray wolf position coordinate with the lowest optimization value at the moment is used as the iterative prey of the current round algorithm
Figure BDA0003620123070000051
7) Judging the iteration number n of the current algorithmdWhether or not n is satisfieddAnd if not, updating the position coordinates of all the gray wolves by using the following formula:
Figure BDA0003620123070000052
and let n bed＝nd+1, repeating the step;
if n is satisfieddIf not less than Max, let nd＝nd+1, entering the next step;
8) prey at that time
Figure BDA0003620123070000053
And the corresponding parameter vector is used as the parameter vector of the deep neural network bin management decision model to obtain the deep neural bin management decision model after parameter optimization.
Optionally, in the step S7, the index data vector is input into the deep neural network warehouse decision model after parameter optimization, and the model outputs an optimal warehouse management decision policy adaptively adjusted based on environmental change, where the optimal warehouse management decision policy includes:
and (4) inputting the index data vector obtained in the step (S4) into a deep neural network warehouse decision model after parameter optimization, outputting an optimal warehouse management decision strategy based on the current environmental index by the model, intelligently managing stored goods in the warehouse according to the optimal warehouse management decision strategy, and performing risk early warning on a corresponding warehouse position area in the index data vector.
In order to solve the above problem, the present invention further provides an intelligent early warning device for warehouse management in combination with current environmental changes, wherein the device comprises:
the device comprises a sensing equipment deployment optimizing device and a monitoring system, wherein the sensing equipment deployment optimizing device is used for constructing a sensing equipment deployment position optimizing objective function in a warehouse, the sensing equipment comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind sensor, the sensing equipment deployment position optimizing objective function is rapidly optimized and solved by using an improved pigeon swarm optimization algorithm, the sensor deployment position is determined, and sensor deployment is carried out according to the sensor position obtained by solving;
the data acquisition processing module is used for acquiring environmental index data of different sensors, preprocessing the acquired environmental index data, reducing the dimension of the data, identifying a warehouse area with large environmental change and acquiring an index data vector;
the warehouse management decision module is used for constructing a deep neural network intelligent warehouse management decision model, optimizing the deep neural network warehouse management decision model by using an improved grey wolf optimization algorithm, inputting index data vectors into the deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy which is adaptively adjusted based on environmental change by the model, and carrying out intelligent risk early warning and management on stored goods in a warehouse area with environmental change according to the optimal warehouse management decision strategy.
In order to solve the above problem, the present invention also provides an electronic device, including:
a memory storing at least one instruction; and
and the processor executes the instructions stored in the memory to realize the intelligent early warning method for warehouse management in combination with the current environmental change.
In order to solve the above problem, the present invention further provides a computer-readable storage medium, where at least one instruction is stored in the computer-readable storage medium, and the at least one instruction is executed by a processor in an electronic device to implement the above intelligent early warning method for warehouse management in combination with current environmental changes.
Compared with the prior art, the invention provides an intelligent early warning method for warehouse management in combination with current environmental changes, and the technology has the following advantages:
firstly, the scheme provides a method for optimizing the deployment position of the sensing equipment in the warehouse, and an objective function is optimized by constructing the deployment position of the sensing equipment in the warehouse, wherein the constructed objective function is as follows:
Figure BDA0003620123070000054
wherein: a is the position where the sensing equipment can be arranged in the warehouse, S belongs to A, then S represents the number of positions, then
Figure BDA0003620123070000061
The energy control strategy of the sensing device representing the location s,
Figure BDA0003620123070000062
the sensing device representing location s remains dormant at time t,
Figure BDA0003620123070000063
the sensing device representing location s performs context awareness at time t; (sigma.)s,i)2The measurement variance of the sensor system at position s, i denotes the type of sensor, i ═ 1,2,3,4,5}, i ═ 1 denotes the temperature sensor, i ═ 2 denotes the humidity sensor, i ═ 3 denotes the two positionsA carbon oxide concentration sensor, i-4 is denoted as an illumination intensity sensor, and i-5 is denoted as a wind sensor; t is a time window for the sensor to execute environment sensing; mu.ss,tFor measuring accurate values at a sensing device of type i at a location s, muiThe average value of the measurements of the sensing devices of the same type in the warehouse is obtained; the constraint conditions of the objective function are as follows:
Figure BDA0003620123070000064
N≤S
Figure BDA0003620123070000065
wherein: n represents the total number of the sensing equipment to be deployed; e represents the number of environment sensing operations which can be performed by the sensing device before the energy is exhausted; according to the set objective function, different types of sensors are arranged at positions where sensing equipment can be arranged in the warehouse, the measurement variances and measurement errors of the same type of sensors are used as the deployment position optimization target of the sensing equipment, the deployment positions and energy of the sensors are constrained, and the position deployment strategy of the different types of sensors, which is small in measurement error and measurement fluctuation and low in energy consumption, can be obtained by solving the objective function. Carrying out rapid optimization solution on the deployment position optimization objective function of the sensing equipment by using an improved pigeon flock optimization algorithm, wherein the improved pigeon flock optimization algorithm comprises the following processes: 1) constructing an S-dimensional pigeon flock position solving space, wherein S represents the total number of positions in a warehouse where sensing equipment can be configured; 2) initializing n1Doves are selected, the position and the speed of each dove are initialized, and the position and the speed of any jth dove are as follows:
xj＝[xj1,xj2,…,xjS]
vj＝[vj1,vj2,…,vjS]
wherein: x is a radical of a fluorine atomjFor the position vector of any jth pigeon, each position vector corresponds to one type of sensor equipment deploymentPosition, xjSIndicates the type of sensor, x, deployed at the S-th warehouse sensing device deployment locationjS∈[0,5]Wherein x isjS0 denotes that no sensor is deployed for this position, xjS1 denotes the location at which the temperature sensor is deployed, xjS2 denotes the location at which the humidity sensor is disposed, xjS3 denotes the location at which the carbon dioxide concentration sensor is deployed, xjS4 denotes that the position is deployed with an illumination intensity sensor, xjS5, representing that the wind sensor is deployed at the position, inputting the position vector into an optimization objective function of the deployment position of the sensing equipment, wherein the value of the optimization objective function is the fitness value of the jth pigeon; v. ofjIs the velocity vector of any jth pigeon; 3) setting the iteration number of the current algorithm as ngAnd n isgInitializing to 0, and setting the number of termination iterations of the first-stage algorithm to ng1The number of times of the termination iteration of the second-stage algorithm is ng2(ii) a 4) Judging whether the iteration number of the current algorithm meets ng≥ng1If not, updating the positions and speeds of all pigeons by the following formula:
Figure BDA0003620123070000066
Figure BDA0003620123070000067
Figure BDA0003620123070000068
wherein:
Figure BDA0003620123070000069
is at the n-thgAfter the iteration, the pigeon position vector with the minimum fitness value in the population; r represents a pigeon position solving space, the flight direction of the pigeons is controlled, and e is a natural constant;
Figure BDA00036201230700000610
representing a differential sequence; and let ng＝ng+1, repeating the step; if n is satisfiedg≥ng1Then let ng＝ng+1, entering the next step; 5) calculating the order ngAfter iteration, the fitness function of each pigeon is deleted, 1/3 pigeons with the highest fitness function in the pigeon group are deleted, after deletion, the positions and the speeds of the rest pigeons are updated, and n is madeg＝ng+1, repeat this step until ng≥ng2Or only within two pigeons remain, calculating the adaptability value of the remaining pigeons at the moment, and taking the position vector corresponding to the pigeon with the minimum adaptability value as the solving result of the optimal objective function of the deployment position of the sensing equipment. Obtaining the optimal position vector x according to the solution*＝[x*1,x*2,…,x*S]Respectively traversing the optimal position vector x obtained by solving*And a position set A in which the sensing equipment can be arranged in the warehouse, and different types of sensors are deployed at specified positions of an optimal position vector, wherein x*SIndicates the type of sensor, x, that needs to be deployed at the S-th warehouse sensor equipment deployment location*S∈[0,5]，x*S0 denotes that no sensor is deployed for this position, x*S1 denotes the location at which the temperature sensor is deployed, x*S2 denotes the location at which the humidity sensor is disposed, x*S3 denotes the location at which the carbon dioxide concentration sensor is deployed, x*S4 denotes that the position is deployed with an illumination intensity sensor, x*SAnd 5 denotes that the wind sensor is deployed at this position. Compared with the traditional scheme, the scheme has the advantages that the sensor equipment deployment optimization target function is constructed through position coordinate constraint in the warehouse and energy constraint of the sensor, the target function is solved through a heuristic algorithm, sensor types deployed at different positions of the warehouse are obtained, meanwhile, in the target function solving process, the pigeon group optimization algorithm is improved, such as a pigeon group position difference updating strategy is added, so that the population diversity is kept in the position updating process, the local optimization is prevented, and the global searching capability is improved.
Meanwhile, the scheme provides an intelligent early warning method for warehouse management, and the method comprises the following steps of performing data dimension reduction processing on preprocessed index data, wherein the data dimension reduction processing flow comprises the following steps: converting the environment index data set after normalization into a vector form, wherein each dimension vector is time sequence data after normalization, and the length of each dimension vector is the same, so as to obtain an N-dimension vector matrix, wherein N is the total number of deployed sensors; calculating a covariance matrix of the N-dimensional vector matrix, wherein the covariance matrix is in a form of sigma; solving the eigenvalue and the eigenvector of the covariance matrix, calculating the contribution rate of each eigenvalue, and selecting the eigenvector with the accumulated contribution rate reaching 90% as the index data vector after the dimension reduction processing. The index data vectors after the data dimension reduction processing only reserve index vectors with high variance contribution rate, the environmental change variance of the reserved index vectors is large, the problem of improper cargo storage caused by environmental change may exist, and the acquisition of the position area corresponding to the index vector needs to be managed and decided with high probability. The method comprises the steps of constructing a deep neural network intelligent warehouse management decision model, wherein the deep neural network intelligent warehouse management decision model takes index data vectors as input and takes a warehouse management decision strategy as output, specifically, the deep neural network intelligent warehouse management decision model receives the index data vectors, carries out convolution operation on the index data vectors to obtain a convolution characteristic diagram of the index data vectors, inputs the convolution characteristic diagram into a full connection layer, matches the warehouse management decision strategy with the full connection layer output and the convolution characteristic diagram, and optimizes the deep neural network warehouse management decision model by utilizing a wolf optimization algorithm to obtain optimized model parameters. The index data vector is input into the deep neural network warehouse management decision model, the management decision strategy of the warehouse location area corresponding to the index data vector is output by the model, the stored goods in the warehouse location area are intelligently managed according to the warehouse management decision strategy, and risk early warning is carried out on the warehouse location area corresponding to the index data vector.
Drawings
Fig. 1 is a schematic flow chart of an intelligent early warning method for warehouse management in combination with current environmental changes according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of one step of the embodiment of FIG. 1;
FIG. 3 is a schematic flow chart of another step of the embodiment of FIG. 1;
fig. 4 is a functional block diagram of an intelligent early warning apparatus for warehouse management in combination with current environmental changes according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device for implementing a warehouse management intelligent early warning method for current environmental changes according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
The embodiment of the application provides a warehouse management intelligent early warning method combined with current environment change. The execution subject of the intelligent early warning method for warehouse management in combination with the current environmental change includes, but is not limited to, at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiments of the present application. In other words, the intelligent early warning method for warehouse management in combination with the current environmental change can be executed by software or hardware installed in the terminal device or the server device, and the software can be a block chain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like.
Example 1:
s1: and constructing a deployment position optimization objective function of sensing equipment in the warehouse, wherein the sensing equipment comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind power sensor.
In the step S1, an optimization objective function of deployment positions of sensing devices in the warehouse is constructed, where the sensing devices include a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor, and a wind sensor, and the optimization objective function includes:
constructing an optimization objective function of the deployment position of the sensing equipment in the warehouse, wherein the constructed objective function is as follows:
Figure BDA0003620123070000081
wherein:
a is the position where the sensing equipment can be arranged in the warehouse, S belongs to A, S represents the number of the positions, and
Figure BDA0003620123070000082
the energy control strategy of the sensing device representing the location s,
Figure BDA0003620123070000083
the sensing device representing location s remains dormant at time t,
Figure BDA0003620123070000084
the sensing device representing the location s performs context awareness at time t;
(σs,i)2a measurement variance of the sensor device at the location s, i denotes a type of sensor, i ═ 1,2,3,4,5, i ═ 1 denotes a temperature sensor, i ═ 2 denotes a humidity sensor, i ═ 3 denotes a carbon dioxide concentration sensor, i ═ 4 denotes an illumination intensity sensor, and i ═ 5 denotes a wind sensor;
t is a time window for the sensor to execute environment sensing;
μs,tfor measuring accurate values at a sensing device of type i at a position s, μiThe average value of the measurements of the sensing devices of the same type in the warehouse is obtained;
the constraint conditions of the objective function are as follows:
Figure BDA0003620123070000085
N≤S
Figure BDA0003620123070000086
wherein:
n represents the total number of the sensing equipment to be deployed;
e represents the number of environment sensing operations that the sensing device can perform before the energy is exhausted;
it should be explained that different types of sensors are arranged at positions where sensing equipment can be arranged in a warehouse, measurement variances and measurement errors of the same type of sensors are used as optimization targets of the deployment positions of the sensing equipment, the deployment positions and energy of the sensors are constrained, and a position deployment strategy of the different types of sensors, which enables the measurement errors and the measurement fluctuations to be small and energy consumption to be small, can be obtained by solving an objective function.
In the embodiment of the invention, the sensing devices are connected together through a ZigBee network, the sensing devices deployed in a warehouse area are slave nodes in the ZigBee network, the sensing devices corresponding to the slave nodes sense and acquire surrounding environment information in real time and send the acquired environment information data to a ZigBee network master node positioned at a server end, and the master node is responsible for summarizing and processing the data to realize intelligent early warning of warehouse management.
S2: and carrying out fast optimization solution on the optimized objective function of the deployment position of the sensing equipment by using an improved pigeon group optimization algorithm to determine the deployment position of the sensor.
In the step S2, performing fast optimization solution on the optimized objective function of the deployment position of the sensing device by using an improved pigeon swarm optimization algorithm to obtain the deployment position of the sensor, including:
an improved pigeon flock optimization algorithm is used for carrying out rapid optimization solution on the optimized objective function of the deployment position of the sensing equipment, and in detail, referring to fig. 2, the improved pigeon flock optimization algorithm comprises the following steps:
s21, constructing a pigeon group position solving space, initializing a pigeon group, and initializing the position and speed of each pigeon;
s22, setting the iteration times of the first stage and the second stage of the algorithm, and updating each pigeon by using the updating formula of the position and the speed of the pigeon in the first stage of the algorithm;
s23, in the second stage of the algorithm, removing the 1/3 pigeons with the highest fitness function in the pigeon group, continuously updating the positions and the speeds of the pigeons, repeating the steps until the preset second stage iteration times are reached or only pigeons within two remain, calculating the fitness values of the remaining pigeons at the moment, and taking the position vector corresponding to the pigeons with the minimum fitness values as the solving result of the deployment position optimization target function of the sensing equipment.
S3: and deploying the sensors according to the sensor positions obtained by solving, acquiring the environmental index data of different sensors, and preprocessing the acquired environmental index data.
In the step S3, deploying the sensor according to the sensor deployment position obtained by the solving, includes:
obtaining the optimal position vector x according to the solution*＝[x*1,x*2,…,x*S]Respectively traversing the optimal position vector x obtained by solving*And a position set A in which the sensing equipment can be arranged in the warehouse, and different types of sensors are deployed at specified positions of an optimal position vector, wherein x*SIndicates the type of sensor, x, that needs to be deployed at the S-th warehouse sensor equipment deployment location*S∈[0,5]，x*S0 denotes that no sensor is deployed for this position, x*SX denotes the location at which the temperature sensor is disposed*S2 denotes that the location is equipped with a humidity sensor, x*S3 denotes that the carbon dioxide concentration sensor is deployed at the position, x*SThe position is represented by 4, x*SThe position at which the wind sensor is deployed is denoted by 5.
In the step S3, index data of different sensors are collected, and the collected index data is preprocessed, including:
it should be explained that after the sensor devices are deployed, all of the sensor devices within the warehouse may be turned on simultaneously by the sensor device controller.
All sensors are started, the sensors start sensing of the surrounding environment, environment index data near the sensors are collected, in detail, different types of sensors collect different environment index data, a temperature sensor collects temperature time sequence data near the sensors, a humidity sensor collects temperature time sequence data near the sensors, a carbon dioxide concentration sensor collects carbon dioxide concentration time sequence data near the sensors, an illumination intensity sensor collects illumination intensity time sequence data near the sensors, a wind power sensor collects wind power intensity time sequence data near the sensors, and the collected environment index data are combined as follows:
{ys,i(q)|s∈A,i∈[0,5],q∈[tl,th]}
wherein:
i represents a sensor type located at the s-th warehouse sensing equipment deployment location;
s represents the s-th warehouse sensing equipment deployment location;
a represents a set of locations within the warehouse where sensing devices may be disposed;
[tl,th]time sequence interval for data acquisition of sensor, tlFor the moment when the sensor starts data acquisition, thThe moment when the sensor finishes data acquisition;
ys,i(q) is time series data of an environment index i at the deployment position of the s-th warehouse sensing equipment, wherein i is 1 and represents ys,i(q) is temperature index time series data, and i-2 represents ys,i(q) is humidity index time series data, and i-3 represents ys,i(q) is carbon dioxide concentration index time series data, and i-4 represents ys,i(q) is time series data of the light intensity index, and i-5 represents ys,i(q) is wind strength indicator time series data;
data y of any environmental index i in the collected environmental index data set at any timei(t') performing normalization, wherein the formula of the normalization is as follows:
Figure BDA0003620123070000091
wherein:
maxiring for environment index data setThe maximum data value of the environmental index i;
miniis the minimum data value of the environment index i in the environment index data set;
(yi(t '))' is yi(t') normalizing the processed values;
all data in the environment index data set are normalized, the data collected by the sensor are mapped into the range of [ -1,1], and the environment index data set after normalization is as follows:
{y′s,i(q)|s∈A,i∈[0,5],q∈[tl,th]}
wherein:
y′s,i(q) is normalization processing ys,iTime series data after (q).
S4: and performing data dimension reduction on the preprocessed environmental index data to obtain an index data vector, and reducing data correlation.
In the step S4, performing data dimensionality reduction on the preprocessed index data to obtain an index data vector, including:
the data dimensionality reduction processing is performed on the preprocessed index data, and in detail, referring to fig. 3, the flow of the data dimensionality reduction processing is as follows:
s41, converting the environment index data set after normalization into a vector form, wherein each dimension vector is time sequence data after normalization, the length of each dimension vector is the same, and an N-dimension vector matrix is obtained, wherein N is the total number of deployed sensors;
s42, calculating a covariance matrix of the N-dimensional vector matrix, wherein the covariance matrix is in a form of sigma;
s43, solving the eigenvalue and the eigenvector of the covariance matrix, calculating the contribution rate of each eigenvalue, and selecting the eigenvector with the cumulative contribution rate reaching 90% as the index data vector after the dimension reduction processing.
It should be understood that, only the index vector with a high variance contribution rate is retained in the index data vectors after the data dimension reduction processing, and the environmental change variance of the retained index vector is large, which may cause a problem of improper cargo storage due to environmental change, and there is a high probability that a management decision needs to be performed on the acquisition of the location area corresponding to the index vector.
S5: and constructing an intelligent warehouse management decision model of the deep neural network, wherein the model takes the index data vector as input and takes a warehouse management decision strategy as output.
The step S5 is to construct a deep neural network intelligent warehouse management decision model, where the deep neural network intelligent warehouse management decision model takes an index data vector as input and a warehouse management decision strategy as output, and includes:
constructing a deep neural network intelligent warehouse management decision model, wherein the deep neural network intelligent warehouse management decision model comprises an input layer, three convolutional layers and a full connection layer;
in the embodiment of the invention, the deep neural network intelligent warehouse management decision model receives the index data vectors, performs convolution operation on the index data vectors to obtain a convolution characteristic diagram of the index data vectors, inputs the convolution characteristic diagram into the full-link layer, and matches the output of the full-link layer with the convolution characteristic diagram to a warehouse management decision strategy.
S6: and optimizing the deep neural network bin management decision model by using an improved grey wolf optimization algorithm to obtain optimized model parameters.
In the step S6, performing parameter optimization on the deep neural network bin management decision model by using an improved grey wolf optimization algorithm, including:
it should be explained that, because the deep neural network model has numerous internal parameters, the internal parameters of the model need to be adaptively adjusted according to training data, and thus the deep neural network intelligent warehouse management decision model needs to perform training operations.
In detail, the training of the deep neural network bin management decision model by using the improved grey wolf optimization algorithm until the training requirement is met to obtain the parameter-optimized deep neural network bin management decision model comprises the following steps:
1) collecting warehouse management historical data, carrying out normalization and dimension reduction processing on the collected warehouse management historical data, and taking a processing result as a training data set, wherein the training data set comprises index data vectors and corresponding warehouse management decision strategies;
2) constructing a parameter optimization objective function of a deep neural network bin management decision model:
Figure BDA0003620123070000101
wherein:
m is the total number of index data vectors in the training data set;
Ykthe binary coding result of the bin management decision strategy corresponding to the kth group of index data vectors in the training data set is obtained;
Figure BDA0003620123070000102
a binary coding bin management decision strategy for outputting the k group index data vector for the deep neural network bin management decision model;
theta is a model parameter to be optimized;
in the embodiment of the invention, the model parameters to be optimized in the deep neural network bin management decision model comprise weight vectors and offset of 3 convolutional layers;
3) initializing n2The position dimension of each gray wolf is 6-dimensional, and the position coordinate of the g-th gray wolf is as follows:
hg＝[hg1,hg2,…,hg6]
wherein:
[hg1,hg2,hg3]sequentially and respectively weighting vectors of 3 layers of convolution layers;
[hg4,hg5,hg6]sequentially and respectively calculating the offset of 3 convolution layers;
the initialization formula of the gray wolf position coordinate is as follows:
Figure BDA0003620123070000111
each wolf of Grey wolfThe position coordinate of the system corresponds to a deep neural network warehouse decision model, and the position coordinate h of each wolf is usedgThe method comprises the steps of implementing the model into a deep neural network warehouse decision model, inputting a training data set into the implemented model, and taking a value of a parameter optimization objective function as an optimization value of a gray wolf position coordinate;
4) setting the iteration number of the current algorithm as ndAnd n isdInitializing to 0, and setting the maximum algorithm iteration number to Max;
5) calculating the optimized values of all the gray wolf position coordinates in the iterative process of the current round of algorithm;
6) the gray wolf position coordinate with the lowest optimization value at the moment is used as the prey of the current round of algorithm iteration
Figure BDA0003620123070000112
7) Judging the iteration number n of the current algorithmdWhether or not n is satisfieddAnd if not, updating the position coordinates of all the gray wolves by using the following formula:
Figure BDA0003620123070000113
and let n bed＝nd+1, repeating the step;
if n is satisfieddIf not less than Max, let nd＝nd+1, entering the next step;
8) prey at that time
Figure BDA0003620123070000114
And the corresponding parameter vector is used as the parameter vector of the deep neural network bin management decision model to obtain the parameter-optimized deep neural bin management decision model.
S7: and inputting the index data vector into the deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy based on environment change self-adaptive adjustment by the model, and intelligently managing the stored goods in the warehouse according to the optimal warehouse management decision strategy.
In the step S7, the index data vector is input into the deep neural network warehouse management decision model after parameter optimization, and the model outputs an optimal warehouse management decision strategy adaptively adjusted based on environmental change, including:
and (4) inputting the index data vector obtained in the step (S4) into the deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy based on the current environmental index by the model, intelligently managing the stored goods in the warehouse according to the optimal warehouse management decision strategy, and performing risk early warning on the corresponding warehouse position area in the index data vector.
In detail, the index data vector is input into the deep neural network warehouse management decision model, the model outputs a management decision strategy of a warehouse location area corresponding to the index data vector, the stored goods in the warehouse location area are intelligently managed according to the warehouse management decision strategy, and risk early warning is carried out on the warehouse location area corresponding to the index data vector.
Example 2:
fig. 4 is a functional block diagram of a warehouse management intelligent warning device according to an embodiment of the present invention, which is combined with a current environmental change, and is capable of implementing the intelligent warning method in embodiment 1.
The intelligent early warning device 100 for warehouse management combined with the current environmental change can be installed in electronic equipment. According to the realized function, the intelligent early warning device for warehouse management in combination with the current environmental change may include a sensing device deployment optimization device 101, a data acquisition and processing module 102, and a warehouse management decision module 103. The module of the present invention, which may also be referred to as a unit, refers to a series of computer program segments that can be executed by a processor of an electronic device and that can perform a fixed function, and that are stored in a memory of the electronic device.
The device comprises a sensing equipment deployment optimizing device 101, a sensor deployment optimization device and a control device, wherein the sensing equipment deployment optimizing device 101 is used for constructing a sensing equipment deployment position optimizing objective function in a warehouse, the sensing equipment comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind sensor, the sensing equipment deployment position optimizing objective function is rapidly optimized and solved by using an improved pigeon group optimization algorithm, the sensor deployment position is determined, and sensor deployment is carried out according to the sensor position obtained by solving;
the data acquisition processing module 102 is configured to acquire environmental index data of different sensors, perform preprocessing and data dimension reduction processing on the acquired environmental index data, identify a warehouse area with a large environmental change, and acquire an index data vector;
the warehouse management decision module 103 is used for constructing a deep neural network intelligent warehouse management decision model, optimizing the deep neural network warehouse management decision model by using an improved grey wolf optimization algorithm, inputting the index data vector into the deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy which is adaptively adjusted based on environmental change by the model, and performing intelligent risk early warning and management on goods stored in a warehouse area with environmental change according to the optimal warehouse management decision strategy.
In detail, in the embodiment of the present invention, when the modules in the intelligent early warning device 100 for warehouse management in combination with current environmental changes are used, the same technical means as the above-mentioned intelligent early warning method for warehouse management in combination with current environmental changes in fig. 1 are used, and the same technical effects can be produced, which are not described herein again.
Example 3:
fig. 5 is a schematic structural diagram of an electronic device for implementing the intelligent early warning method for warehouse management in combination with current environmental changes according to an embodiment of the present invention.
The electronic device 1 may include a processor 10, a memory 11, and a bus, and may further include a computer program, such as a warehouse management smart warning program 12, stored in the memory 11 and executable on the processor 10.
The memory 11 includes at least one type of readable storage medium, which includes flash memory, removable hard disk, multimedia card, card type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, e.g. a removable hard disk of the electronic device 1. The memory 11 may also be an external storage device of the electronic device 1 in other embodiments, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the electronic device 1. Further, the memory 11 may also include both an internal storage unit and an external storage device of the electronic device 1. The memory 11 may be used to store not only application software installed in the electronic device 1 and various data, such as codes of the intelligent early warning program 12 for warehouse management, but also temporarily store data that has been output or will be output.
The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same or different functions, including one or more Central Processing Units (CPUs), microprocessors, digital Processing chips, graphics processors, and combinations of various control chips. The processor 10 is a Control Unit of the electronic device, connects various components of the whole electronic device by using various interfaces and lines, and executes various functions and processes data of the electronic device 1 by running or executing programs or modules (warehouse management intelligent warning programs, etc.) stored in the memory 11 and calling data stored in the memory 11.
The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is arranged to enable connection communication between the memory 11 and at least one processor 10 or the like.
Fig. 5 only shows an electronic device with components, and it will be understood by a person skilled in the art that the structure shown in fig. 5 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or a combination of certain components, or a different arrangement of components.
For example, although not shown, the electronic device 1 may further include a power supply (such as a battery) for supplying power to each component, and preferably, the power supply may be logically connected to the at least one processor 10 through a power management device, so as to implement functions of charge management, discharge management, power consumption management, and the like through the power management device. The power supply may also include any component of one or more dc or ac power sources, recharging devices, power failure detection circuitry, power converters or inverters, power status indicators, and the like. The electronic device 1 may further include various sensors, a bluetooth module, a Wi-Fi module, and the like, which are not described herein again.
Further, the electronic device 1 may further include a network interface, and optionally, the network interface may include a wired interface and/or a wireless interface (such as a WI-FI interface, a bluetooth interface, etc.), which are generally used for establishing a communication connection between the electronic device 1 and other electronic devices.
Optionally, the electronic device 1 may further comprise a user interface, which may be a Display (Display), an input unit (such as a Keyboard), and optionally a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, or the like. The display, which may also be referred to as a display screen or display unit, is suitable for displaying information processed in the electronic device 1 and for displaying a visualized user interface, among other things.
It is to be understood that the described embodiments are for purposes of illustration only and that the scope of the appended claims is not limited to such structures.
The intelligent early warning program 12 stored in the memory 11 of the electronic device 1 is a combination of instructions, which when executed in the processor 10, can implement:
constructing a deployment position optimization objective function of sensing equipment in a warehouse, wherein the sensing equipment comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind sensor;
carrying out rapid optimization solution on the optimized objective function of the deployment position of the sensing equipment by using an improved pigeon group optimization algorithm to determine the deployment position of the sensor;
deploying sensors according to the sensor positions obtained by solving, acquiring environment index data of different sensors, and preprocessing the acquired environment index data;
carrying out data dimension reduction on the preprocessed environmental index data to obtain an index data vector and reduce data correlation;
constructing a deep neural network intelligent warehouse management decision model, wherein the model takes an index data vector as input and takes a warehouse management decision strategy as output;
optimizing the deep neural network bin management decision model by using an improved grey wolf optimization algorithm to obtain optimized model parameters;
and inputting the index data vector into a deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy based on environment change self-adaptive adjustment by the model, and performing intelligent management on stored goods in the warehouse according to the optimal warehouse management decision strategy.
Specifically, the specific implementation method of the processor 10 for the instruction may refer to the description of the relevant steps in the embodiments corresponding to fig. 1 to fig. 5, which is not repeated herein.
It should be noted that the above-mentioned numbers of the embodiments of the present invention are merely for description, and do not represent the merits of the embodiments. And the terms "comprises," "comprising," or any other variation thereof, herein are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention or portions thereof contributing to the prior art may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above and includes several instructions for enabling a terminal device (which may be a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all equivalent structures or equivalent processes performed by the present invention or directly or indirectly applied to other related technical fields are also included in the scope of the present invention.

Claims (10)

1. A warehouse management intelligent early warning method combined with current environment change is characterized by comprising the following steps:
s1: constructing a deployment position optimization objective function of sensing equipment in a warehouse, wherein the sensing equipment comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind sensor;
s2: carrying out rapid optimization solution on the optimized objective function of the deployment position of the sensing equipment by using an improved pigeon group optimization algorithm to determine the deployment position of the sensor;
s3: deploying sensors according to the sensor positions obtained by solving, acquiring environmental index data of different sensors, and preprocessing the acquired environmental index data;
s4: performing data dimension reduction processing on the preprocessed environmental index data to obtain an index data vector;
s5: constructing a deep neural network intelligent warehouse management decision model, wherein the model takes an index data vector as input and takes a warehouse management decision strategy as output;
s6: optimizing the deep neural network bin management decision model by using an improved grey wolf optimization algorithm to obtain optimized model parameters;
s7: and inputting the index data vector into a deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy based on environment change self-adaptive adjustment by the model, and performing intelligent management on stored goods in the warehouse according to the optimal warehouse management decision strategy.
2. The intelligent early warning method for warehouse management in combination with the current environmental change as claimed in claim 1, wherein the step S1 is implemented to construct an objective function for optimizing the deployment location of the sensing devices in the warehouse, wherein the sensing devices include a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind sensor, and the method comprises:
constructing an optimization objective function of the deployment position of the sensing equipment in the warehouse, wherein the constructed objective function is as follows:
Figure FDA0003620123060000011
wherein:
a is the position where the sensing equipment can be arranged in the warehouse, S belongs to A, S represents the number of the positions, and
Figure FDA0003620123060000012
the energy control strategy of the sensing device representing the location s,
Figure FDA0003620123060000013
the sensing device representing location s remains dormant at time t,
Figure FDA0003620123060000014
the sensing device representing the location s performs context awareness at time t;
(σs，i)2a measurement variance of the sensor device at the location s, i denotes a type of sensor, i ═ 1,2,3,4,5, i ═ 1 denotes a temperature sensor, i ═ 2 denotes a humidity sensor, i ═ 3 denotes a carbon dioxide concentration sensor, i ═ 4 denotes an illumination intensity sensor, and i ═ 5 denotes a wind sensor;
t is a time window for the sensor to execute environment sensing;
μs,tfor measuring accurate values at a sensing device of type i at a location s, muiThe average value of the measurements of the sensing devices of the same type in the warehouse is obtained;
the constraint conditions of the objective function are as follows:
Figure FDA0003620123060000015
N≤S
Figure FDA0003620123060000016
wherein:
n represents the total number of the sensing equipment to be deployed;
e represents the number of context aware operations that the sensing device is capable of performing before energy is depleted.
3. The intelligent early warning method for warehouse management combined with the current environmental change as claimed in claim 2, wherein the step S2 is implemented by using an improved pigeon flock optimization algorithm to perform fast optimization solution on the optimized objective function of the deployment position of the sensing device, so as to obtain the deployment position of the sensor, and includes:
carrying out rapid optimization solution on the optimal objective function of the deployment position of the sensing equipment by using an improved pigeon group optimization algorithm, wherein the flow of the improved pigeon group optimization algorithm is as follows:
1) constructing an S-dimensional pigeon flock position solving space, wherein S represents the total number of positions where sensing equipment can be configured in a warehouse;
2) initializing n1Only pigeons and initialize the position and speed of each pigeon, then the position and speed of any jth pigeon is:
xj＝[xj1，xj2，...，xjS]
vj＝[vj1，vj2，...，vjS]
wherein:
xjfor any jth pigeon position vector, each position vector corresponds to a sensing equipment deployment position, xjSIndicates the type of sensor, x, deployed at the S-th warehouse sensing device deployment locationjS∈[0，5]Wherein x isjS0 denotes that no sensor is deployed for this position, xjS1 denotes the location at which the temperature sensor is deployed, xjS2 denotes that the location is equipped with a humidity sensor, xjS3 denotes the location at which the carbon dioxide concentration sensor is deployed, xjS4 denotes that the position is deployed with an illumination intensity sensor, xjS5, representing that the wind sensor is deployed at the position, inputting the position vector into an optimization objective function of the deployment position of the sensing equipment, wherein the value of the optimization objective function is the fitness value of the jth pigeon;
vjis the velocity vector of any jth pigeon;
3) setting the iteration number of the current algorithm as ngAnd n isgInitializing to 0, and setting the number of termination iterations of the first-stage algorithm to ng1The number of times of the termination iteration of the second-stage algorithm is ng2；
4) Judging whether the iteration number of the current algorithm meets ng≥ng1If not, updating the positions and speeds of all pigeons by the following formula:
Figure FDA0003620123060000021
Figure FDA0003620123060000022
Figure FDA0003620123060000023
wherein:
Figure FDA0003620123060000024
is at the n-thgAfter the iteration, the pigeon position vector with the minimum fitness value in the population is obtained;
r represents a pigeon group position solving space, the flight direction of the pigeons is controlled, and e is a natural constant;
Figure FDA0003620123060000025
representing a differential sequence;
and let n beg＝ng+1, repeating the step;
if n is satisfiedg≥ng1Then let n beg＝ng+1, entering the next step;
5) calculating the order ngDeleting the fitness function of each pigeon after iteration, deleting the 1/3 pigeons with the highest fitness function in the pigeon group, updating the positions and the speeds of the rest pigeons after deletion, and enabling n to beg＝ng+1, repeat this step until ng≥ng2Or only within two pigeons remain, calculating the adaptability value of the remaining pigeons at the moment, and taking the position vector corresponding to the pigeon with the minimum adaptability value as the solving result of the optimal objective function of the deployment position of the sensing equipment.
4. The intelligent warning method for warehouse management in combination with the current environmental change as claimed in claim 3, wherein the deploying of the sensor according to the solved sensor deployment location in step S3 includes:
obtaining the optimal position vector x according to the solution*＝[x*1，x*2，...，x*S]Respectively traversing the obtained optimal position vector x*And a position set A in which the sensing equipment can be arranged in the warehouse, and different types of sensors are deployed at specified positions of an optimal position vector, wherein x*SIndicates the type of sensor, x, that needs to be deployed at the S-th warehouse sensing equipment deployment location*S∈[0，5]，x*s0 denotes that no sensor is deployed for this position, x*SX denotes the location at which the temperature sensor is disposed*S2 denotes that the location is equipped with a humidity sensor, x*S3 denotes that the carbon dioxide concentration sensor is deployed at the position, x*SThe position is represented by 4, x*SThe position at which the wind sensor is deployed is denoted by 5.
5. The intelligent early warning method for warehouse management in combination with the current environmental change as claimed in claim 1, wherein the step S3 is to collect index data of different sensors and to preprocess the collected index data, including:
all sensors are started, the sensors start sensing of the surrounding environment, environment index data near the sensors are collected, the sensors of different types collect different environment index data, the temperature sensors collect temperature time sequence data near the sensors, the humidity sensors collect temperature time sequence data near the sensors, the carbon dioxide concentration sensors collect carbon dioxide concentration time sequence data near the sensors, the illumination intensity sensors collect illumination intensity time sequence data near the sensors, the wind power sensors collect wind power intensity time sequence data near the sensors, and the collected environment index data are combined as follows:
{ys，i(q)|s∈A，i∈[0，5]，q∈[tl，th]}
wherein:
i represents a sensor type at the s-th warehouse sensing equipment deployment location;
s represents the s-th warehouse sensing equipment deployment location;
a represents a set of locations within the warehouse where sensing devices may be disposed;
[tl，th]time sequence interval for data acquisition of sensor, tlFor the moment at which the sensor starts data acquisition, thThe moment when the sensor finishes data acquisition;
ys，i(q) is time series data of an environment index i at the deployment position of the s-th warehouse sensing equipment, wherein i is 1 and represents ys，i(q) is temperature index time series data, and i-2 represents ys，i(q) is humidity index time series data, and i-3 represents ys，i(q) is carbon dioxide concentration index time series data, and i-4 represents ys，i(q) is time series data of the light intensity index, and i-5 represents ys，i(q) is wind strength indicator time series data;
data y of any environment index i in the collected environment index data set at any timei(t') performing normalization, wherein the formula of the normalization is as follows:
Figure FDA0003620123060000031
wherein:
maxithe maximum data value of the environment index i in the environment index data set;
miniis the minimum data value of the environment index i in the environment index data set;
(yi(t '))' is yi(t') normalizing the processed values;
all data in the environment index data set are normalized, the data collected by the sensor are mapped into the range of [ -1,1], and the environment index data set after normalization is as follows:
{y′s，i(q)|s∈A，i∈[0，5]，q∈[tl，th]}
wherein:
y′s，i(q) is normalization processing ys，iAfter (q) timeAnd (4) sequence data.
6. The intelligent early warning method for warehouse management in combination with the current environmental change as recited in claim 5, wherein the step S4 of performing data dimension reduction processing on the preprocessed index data to obtain an index data vector comprises:
carrying out data dimension reduction on the preprocessed index data, wherein the data dimension reduction process comprises the following steps:
converting the environment index data set after normalization into a vector form, wherein each dimension vector is time sequence data after normalization, and the length of each dimension vector is the same, so as to obtain an N-dimension vector matrix, wherein N is the total number of deployed sensors;
calculating a covariance matrix of the N-dimensional vector matrix, wherein the covariance matrix is in the form of sigma;
solving the eigenvalue and the eigenvector of the covariance matrix, calculating the contribution rate of each eigenvalue, and selecting the eigenvector with the accumulated contribution rate reaching 90% as the index data vector after the dimension reduction processing.
7. The intelligent early warning method for warehouse management in combination with current environmental changes as claimed in claim 1, wherein the step S5 is implemented as an intelligent deep neural network warehouse decision model, wherein the intelligent deep neural network warehouse decision model takes an index data vector as input and a warehouse management decision strategy as output, and comprises:
constructing a deep neural network intelligent warehouse management decision model, wherein the deep neural network intelligent warehouse management decision model comprises an input layer, three convolutional layers and a full connection layer; the deep neural network intelligent warehouse management decision model receives the index data vectors, performs convolution operation on the index data vectors to obtain a convolution characteristic diagram of the index data vectors, inputs the convolution characteristic diagram into the full-connection layer, and matches the output of the full-connection layer with the convolution characteristic diagram to a warehouse management decision strategy.
8. The intelligent early warning method for warehouse management combined with the current environmental change as claimed in claim 7, wherein the step of S6 using the improved grey wolf optimization algorithm to perform parameter optimization on the deep neural network warehouse decision model includes:
the method for training the deep neural network warehouse management decision model by using the improved wolf optimization algorithm until the deep neural network warehouse management decision model with the optimized parameters is obtained after the training requirements are met comprises the following steps:
1) collecting warehouse management historical data, carrying out normalization and dimension reduction processing on the collected warehouse management historical data, and taking a processing result as a training data set, wherein the training data set comprises index data vectors and corresponding warehouse management decision strategies;
2) constructing a parameter optimization objective function of a deep neural network bin management decision model:
Figure FDA0003620123060000041
wherein:
m is the total number of index data vectors in the training data set;
Yka binary coding result of the warehouse management decision strategy corresponding to the kth group of index data vectors in the training data set;
Figure FDA0003620123060000042
a binary coding bin management decision strategy for outputting the k group index data vector for the deep neural network bin management decision model;
theta is a model parameter to be optimized;
3) initializing n2The position dimension of each gray wolf is 6-dimensional, and the position coordinate of the g-th gray wolf is as follows:
hg＝[hg1，hg2，...，hg6]
wherein:
[hg1，hg2，hg3]sequentially and respectively taking the weight vectors of the 3 convolutional layers;
[hg4，hg5，hg6]sequentially and respectively calculating the offset of 3 convolution layers;
the initialization formula of the gray wolf position coordinate is as follows:
Figure FDA0003620123060000043
the position coordinate of each wolf corresponds to a deep neural network warehouse decision model, and the position coordinate h of each wolf is usedgThe implementation is a deep neural network warehouse decision model, a training data set is input into the implementation model, and the value of a parameter optimization objective function is used as the optimized value of the gray wolf position coordinate;
4) setting the iteration number of the current algorithm as ndAnd n isdThe initialization is 0, and the iteration number of the maximum algorithm is set to Max
5) Calculating the optimized values of all the gray wolf position coordinates in the iterative process of the current round of algorithm;
6) the gray wolf position coordinate with the lowest optimization value at the moment is used as the prey of the current round of algorithm iteration
Figure FDA0003620123060000044
7) Judging the iteration number n of the current algorithmdWhether or not n is satisfieddAnd if the position coordinates of all the gray wolves are not satisfied, updating the position coordinates of all the gray wolves by using the following formula:
Figure FDA0003620123060000045
and let nd＝nd+1, repeating the step;
if n is satisfieddIf not less than Max, let nd＝nd+1, entering the next step;
8) prey at that time
Figure FDA0003620123060000046
And the corresponding parameter vector is used as the parameter vector of the deep neural network bin management decision model to obtain the parameter-optimized deep neural bin management decision model.
9. The intelligent early warning method for warehouse management in combination with current environmental changes as claimed in claims 6-8, wherein the step S7 is to input the index data vector into a deep neural network warehouse decision model after parameter optimization, and the model outputs an optimal warehouse management decision strategy adaptively adjusted based on environmental changes, including:
and (4) inputting the index data vector obtained in the step (S4) into a deep neural network warehouse decision model after parameter optimization, outputting an optimal warehouse management decision strategy based on the current environmental index by the model, intelligently managing stored goods in the warehouse according to the optimal warehouse management decision strategy, and performing risk early warning on a corresponding warehouse position area in the index data vector.
10. The utility model provides a warehouse management intelligence early warning device who combines current environmental change which characterized in that, the device includes:
the sensor equipment deployment optimization device is used for constructing a sensor equipment deployment position optimization objective function in a warehouse, wherein the sensor equipment comprises a temperature sensor, a humidity sensor, a carbon dioxide concentration sensor, an illumination intensity sensor and a wind sensor, the sensor equipment deployment position optimization objective function is rapidly optimized and solved by using an improved pigeon swarm optimization algorithm, the sensor deployment position is determined, and sensor deployment is carried out according to the sensor position obtained by solving;
the data acquisition processing module is used for acquiring environment index data of different sensors, preprocessing the acquired environment index data, reducing the dimension of the data, identifying a warehouse area with large environmental change and acquiring an index data vector;
the warehouse management decision module is used for constructing a deep neural network intelligent warehouse management decision model, optimizing the deep neural network warehouse management decision model by using an improved grey wolf optimization algorithm, inputting index data vectors into the deep neural network warehouse management decision model after parameter optimization, outputting an optimal warehouse management decision strategy adaptively adjusted based on environmental change by the model, and performing intelligent risk early warning and management on goods stored in a warehouse area with environmental change according to the optimal warehouse management decision strategy so as to realize the warehouse management intelligent early warning method combined with the current environmental change as claimed in claim 1.
CN202210454908.4A 2022-04-24 2022-04-24 Warehouse management intelligent early warning method combined with current environment change Pending CN114781887A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210454908.4A CN114781887A (en) 2022-04-24 2022-04-24 Warehouse management intelligent early warning method combined with current environment change

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210454908.4A CN114781887A (en) 2022-04-24 2022-04-24 Warehouse management intelligent early warning method combined with current environment change

Publications (1)

Publication Number Publication Date
CN114781887A true CN114781887A (en) 2022-07-22

Family

ID=82433876

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210454908.4A Pending CN114781887A (en) 2022-04-24 2022-04-24 Warehouse management intelligent early warning method combined with current environment change

Country Status (1)

Country Link
CN (1) CN114781887A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116227363A (en) * 2023-04-25 2023-06-06 湖南省水务规划设计院有限公司 Flood early warning method based on sensor distribution optimization

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116227363A (en) * 2023-04-25 2023-06-06 湖南省水务规划设计院有限公司 Flood early warning method based on sensor distribution optimization
CN116227363B (en) * 2023-04-25 2023-08-15 湖南省水务规划设计院有限公司 Flood early warning method based on sensor distribution optimization

Similar Documents

Publication Publication Date Title
CN109858569A (en) Multi-tag object detecting method, system, device based on target detection network
CN110674323B (en) Unsupervised cross-modal Hash retrieval method and system based on virtual label regression
CN110647920A (en) Transfer learning method and device in machine learning, equipment and readable medium
CN110660478A (en) Cancer image prediction and discrimination method and system based on transfer learning
CN111881958B (en) License plate classification recognition method, device, equipment and storage medium
CN109685765A (en) A kind of X-ray pneumonia prediction of result device based on convolutional neural networks
CN114781887A (en) Warehouse management intelligent early warning method combined with current environment change
CN112272074B (en) Information transmission rate control method and system based on neural network
CN113312957A (en) off-Shift identification method, device, equipment and storage medium based on video image
CN112016617B (en) Fine granularity classification method, apparatus and computer readable storage medium
CN115272894A (en) Unmanned aerial vehicle-oriented image target detection method and device, electronic equipment and storage medium
CN115393690A (en) Light neural network air-to-ground observation multi-target identification method
CN117195059B (en) Quick detection method for water pollution of water supply pipe network
CN116109945B (en) Remote sensing image interpretation method based on ordered continuous learning
CN117372877A (en) Star map identification method and device based on neural network and related medium
CN116524296A (en) Training method and device of equipment defect detection model and equipment defect detection method
CN116151323A (en) Model generation method, device, electronic equipment and storage medium
CN116029440A (en) Ultra-short-term power prediction method and device for photovoltaic power station
CN115359337A (en) Searching method, system and application of pulse neural network for image recognition
CN114611719A (en) XGboost training method based on cuckoo search algorithm
CN112187894A (en) Container dynamic scheduling method based on load correlation prediction
CN110399812A (en) Face characteristic intelligent extract method, device and computer readable storage medium
Trush et al. Mask Mode Monitoring Systems Using IT Technologies
CN110415110A (en) Progress monitoring method, progress monitoring device and electronic equipment
CN116522105B (en) Method, device, equipment and medium for integrally constructing data based on cloud computing

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination