CN113537459A - Method for predicting humiture of drug storage room - Google Patents
Method for predicting humiture of drug storage room Download PDFInfo
- Publication number
- CN113537459A CN113537459A CN202110719129.8A CN202110719129A CN113537459A CN 113537459 A CN113537459 A CN 113537459A CN 202110719129 A CN202110719129 A CN 202110719129A CN 113537459 A CN113537459 A CN 113537459A
- Authority
- CN
- China
- Prior art keywords
- temperature
- humidity
- layer
- data
- temporal
- 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.)
- Granted
Links
- 239000003814 drug Substances 0.000 title claims abstract description 60
- 229940079593 drug Drugs 0.000 title claims abstract description 52
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000002123 temporal effect Effects 0.000 claims abstract description 42
- 238000012549 training Methods 0.000 claims abstract description 20
- 238000012360 testing method Methods 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 11
- 230000000694 effects Effects 0.000 claims abstract description 6
- 238000010606 normalization Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 239000000284 extract Substances 0.000 claims description 4
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000005728 strengthening Methods 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 230000037211 monthly cycles Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000003442 weekly effect Effects 0.000 description 1
Images
Classifications
-
- 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
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Economics (AREA)
- Quality & Reliability (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Pure & Applied Mathematics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Optimization (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Operations Research (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Fuzzy Systems (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marketing (AREA)
- Evolutionary Biology (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
Abstract
The invention discloses a medicine storehouse temperature and humidity prediction method based on Temporal _ ATT-Former, which comprises the following steps: acquiring historical data of temperature and humidity of a drug warehouse environment, and preprocessing the historical data; establishing a Temporal _ ATT-Former temperature and humidity prediction model; performing temperature and humidity prediction model training on the temperature _ ATT-Former according to the acquired temperature and humidity data characteristics of the drug storeroom; testing the effect of the Temporal _ ATT-Former model; and establishing a temperature and humidity early warning system of the drug storage room. The method accurately predicts the temperature and the humidity of the drug storage room in a certain time interval in the future, meets the requirement of storing drugs in the safe environment of the storage room, and has wide application prospect and practical value.
Description
Technical Field
The invention relates to a supervision algorithm and a time sequence prediction technology, in particular to a medicine storehouse temperature and humidity prediction method based on Temporal _ ATT-Former.
Background
At present, researchers can select mathematical modeling of collected temperature and humidity data and predict the temperature and humidity by combining statistics and machine learning methods, and the method has obvious defects, large calculation amount and low precision.
The existing paper mainly extracts collected temperature and humidity features based on the traditional recurrent neural network method to realize the prediction of the temperature and the humidity. For example, the method comprises the steps of Sezuku, Zhengduch, Bajun and Suzhongtang, prediction of temperature and humidity in the closed pigsty based on deep learning [ J/OL ], agricultural machinery science and newspaper, 2021, and establishment of a long-term and short-term memory network prediction model by combining historical data in the closed pigsty monitored by an actual temperature and humidity sensor. But the prediction speed of the model is reduced because the training speed of the long-short term memory network is slow and the model cannot be parallel. There is also a paper combining CNN and GRU to predict humiture, such as zhao quanming, songzao, liqifeng, zheng wenggang, liuyu, zhang bellelili. However, the CNN is not combined with the time series for losing part of the humiture characteristics when extracting the characteristics.
Traditional humiture monitoring of drug storage room environment is based on manual monitoring, and this kind of monitoring can not guarantee that the medicine is saved under specific humiture environment, consumes the cost of labor moreover.
Disclosure of Invention
The purpose of the invention is as follows: the invention provides a temperature and humidity prediction method for a drug storage room based on Temporal _ ATT-Former, which can realize accurate prediction of temperature and humidity data of the drug storage room and meet the requirement of safe storage of drugs in the storage room.
The technical scheme is as follows: the invention provides a medicine storehouse temperature and humidity prediction method based on Temporal _ ATT-Former, which comprises the following steps:
step 1: acquiring historical data of temperature and humidity of a drug warehouse environment, and preprocessing the historical data;
step 2: establishing a Temporal _ ATT-Former temperature and humidity prediction model;
and step 3: performing temperature and humidity prediction model training on the temperature _ ATT-Former according to the acquired temperature and humidity data characteristics of the drug storeroom;
and 4, step 4: testing the effect of the Temporal _ ATT-Former model;
and 5: and establishing a temperature and humidity early warning system of the drug storage room.
Further, historical data of the temperature and the humidity of the drug warehouse environment is acquired in the step 1, and the steps of preprocessing the historical data are as follows:
step 1.1: for each collection point of the drug storage room, a temperature and humidity sensor is used for collecting a first characteristic parameter TH of the drug storage room1The method comprises the steps of numbering a temperature and humidity sensor, the position of a collection point, temperature data, humidity data, a collected time sequence and alarm information;
step 1.2: preprocessing the first characteristic data, removing repeated data, filling up lost data to obtain second characteristic data TH2The method comprises the steps of collecting time sequence, temperature and humidity data and alarm information;
step 1.3: respectively acquiring daily, weekly and monthly cycle data of the temperature and humidity data of the time series obtained in the step 1.2, and collectingNormalizing and standardizing the data by a zero-mean method to obtain third characteristic data TH3。
Further, the specific steps of establishing the temperature and humidity prediction model of the Temporal _ ATT-Former in the step 2 are as follows:
step 2.1: the Temporal _ ATT-Former comprises an Encoder module and a Decoder module;
step 2.2: establishing an Encoder module which comprises an input layer, a convolutional layer, a time attention layer and a normalization and characteristic output layer;
step 2.3: and establishing a Decoder module which comprises an input layer, a multi-head time attention covering layer, a multi-head time attention layer, a full connection layer and an output layer.
Further, the Encoder module in step 2.2 is specifically:
a first layer: an input layer comprising temperature and humidity characteristics, local time series characteristics, and global time series characteristics, namely temperature and humidity input at the time tWherein the content of the first and second substances,the characteristics of the temperature and the humidity are adopted,is a local time series characteristic of the time series,is a global time series feature;
a second layer: the convolution layer adopts one-dimensional convolution operation to respectively extract the characteristics of the humiture of the drug storage, namely
And a third layer: a multi-head time attention layer for strengthening the temperature and humidity characteristics of the time sequence of the input layer;
a fourth layer: and a normalization and feature output layer.
Further, the Decoder module in step 2.3 specifically is:
a first layer: input layer, input of temperature and humidity data shifted forward by T time stamps based on time series, i.e.
A second layer: an attention layer including a multi-temporal attention layer and a multi-temporal attention layer;
and a third layer: a fully-connected layer;
a fourth layer: and an output layer for outputting the predicted temperature and humidity data of the time series.
Further, the step 3 of performing temperature and humidity prediction model training on the Temporal _ ATT-Former according to the acquired temperature and humidity data characteristics of the drug warehouse comprises the following specific steps:
step 3.1: dividing a data set, namely dividing the temperature and humidity data set TH of the drug storage room preprocessed in the step 13Dividing a training set and a test set according to a certain proportion, and respectively expressing as train and test;
step 3.2: setting a training round epoch and a batch round batch;
step 3.3: training a humiture prediction model of a drug warehouse, inputting train into a Temporal _ ATT-Former network for training, and storing the trained model.
Further, the effect of the Temporal _ ATT-Former model is tested in the step 4, and the specific steps are as follows:
step 4.1: verifying the performance of the Temporal _ ATT-Former model, loading the stored model, and inputting the test into the model to obtain a prediction result r of the test set;
step 4.2: and comparing the temperature and humidity r predicted by the model with the real temperature and humidity value, calculating the average absolute error and the root mean square error, adjusting the parameters of the model, and optimizing the network structure.
Further, a temperature and humidity early warning system of the drug storage room is established in the step 5, and the method specifically comprises the following steps:
step 5.1: the temperature and humidity early warning system for the drug storage room comprises a temperature and humidity acquisition module, a storage module, a temperature and humidity prediction module and an early warning module;
step 5.2: the temperature and humidity acquisition module comprises temperature and humidity data of each acquisition point of the drug warehouse and transmits the temperature and humidity data to the server end;
step 5.3: the storage module is arranged at the server end and used for storing temperature and humidity data sent by the temperature and humidity sensor;
step 5.4: the temperature and humidity prediction module is used for cleaning, preprocessing, normalizing and standardizing temperature and humidity data, building a Temporal _ ATT-Former neural network, predicting the temperature and humidity value of a given time interval and transmitting the result to the cloud server end;
step 5.5: the early warning module judges a temperature and humidity prediction result in the cloud server, and if the temperature and humidity prediction result is greater than a certain set temperature and humidity threshold value, an early warning signal is sent to the client.
Has the advantages that: compared with the prior art, the invention has the following remarkable advantages: according to the invention, firstly, the temperature and humidity data of the drug storage room are preprocessed, normalization and standardization processing are carried out on the data by adopting a zero-mean method, and supervised training is carried out on the collected temperature and humidity historical data of the drug storage room by utilizing a time attention mechanism and combining with an improved Transformer algorithm, so that the temperature and humidity prediction of the drug storage room with higher accuracy is realized, the prediction accuracy and speed are optimized, the manual monitoring cost can be reduced, and the drug can be stored for a long time in a safe environment.
Drawings
FIG. 1 is a general flow diagram of the present invention;
FIG. 2 is a model diagram of temperature and humidity prediction for constructing a Temporal _ ATT-Former pharmacy room in the invention;
FIG. 3 is a flowchart of the Temporal _ ATT-Former algorithm training and testing of the present invention;
fig. 4 is a flow chart of the temperature and humidity early warning system for the pharmaceutical warehouse in the invention.
Detailed Description
The invention is further elucidated with reference to the drawings and the embodiments.
As shown in fig. 1-4, a method for predicting temperature and humidity of a pharmacy based on Temporal _ ATT-Former includes the following steps:
step 1: acquiring historical data of temperature and humidity of a drug storage environment, and preprocessing the historical data, wherein the specific method comprises the following steps:
step 1.1: for each collection point of the drug storage room, a temperature and humidity sensor is used for collecting a first characteristic parameter TH of the drug storage room1The method comprises the steps of numbering a temperature and humidity sensor, the position of a collection point, temperature data, humidity data, a collected time sequence and alarm information;
step 1.2: preprocessing the first characteristic data, removing repeated data, filling up lost data to obtain second characteristic data TH2The method comprises the steps of collecting time sequence, temperature and humidity data and alarm information;
step 1.3: respectively acquiring cycle data of each day, each week and each month from the time series temperature and humidity data acquired in the step 1.2, and normalizing and standardizing the data by adopting a zero-mean method to obtain third characteristic data TH3。
Step 2: the method for establishing the temperature and humidity prediction model of the Temporal _ ATT-Former comprises the following specific steps:
step 2.1: the Temporal _ ATT-Former comprises an Encoder module and a Decoder module;
step 2.2: establishing an Encoder module which comprises an input layer, a convolutional layer, a time attention layer and a normalization and characteristic output layer;
a first layer: an input layer comprising temperature and humidity characteristics, local time series characteristics, and global time series characteristics, namely temperature and humidity input at the time tWhereinThe characteristics of the temperature and the humidity are adopted,is a local time series characteristic of the time series,is a global time series feature.
A second layer: the convolution layer adopts one-dimensional convolution operation to respectively extract the characteristics of the humiture of the drug storage room, namely
And a third layer: and the multi-head time attention layer strengthens the temperature and humidity characteristics of the time sequence of the input layer.
A fourth layer: a normalization sum feature output layer;
step 2.3: establishing a Decoder module which comprises an input layer, a multi-head time attention layer, a full connection layer and an output layer;
a first layer: an input layer for inputting the temperature and humidity data based on time series by shifting the data forward by T time stamps
A second layer: an attention layer including a multi-temporal attention layer and a multi-temporal attention layer;
and a third layer: a fully-connected layer;
a fourth layer: and an output layer for outputting the predicted temperature and humidity data of the time series.
And step 3: performing temperature and humidity prediction model training on the acquired temperature and humidity data characteristics of the drug warehouse by using a Temporal _ ATT-Former, wherein the specific method comprises the following steps:
step 3.1: dividing a data set, namely dividing the temperature and humidity data set TH of the drug storage room preprocessed in the step 13Dividing a training set and a test set according to a certain proportion, and respectively expressing as train and test;
step 3.2: setting a training round epoch and a batch round batch;
step 3.3: training a humiture prediction model of a drug warehouse, inputting train into a Temporal _ ATT-Former network for training, and storing the trained model.
And 4, step 4: the method for testing the effect of the Temporal _ ATT-Former model comprises the following steps:
step 4.1: verifying the performance of the Temporal _ ATT-Former model, loading the stored model, and inputting the test into the model to obtain a prediction result r of the test set;
step 4.2: and comparing the temperature and humidity r predicted by the model with the real temperature and humidity value, calculating the average absolute error and the root mean square error, adjusting the parameters of the model, and optimizing the network structure.
And 5: the method for establishing the temperature and humidity early warning system of the drug storage room comprises the following steps:
step 5.1: the temperature and humidity early warning system for the drug storage room comprises a temperature and humidity acquisition module, a storage module, a temperature and humidity prediction module and an early warning module;
step 5.2: the temperature and humidity acquisition module comprises temperature and humidity data of each acquisition point of the drug storage room and transmits the temperature and humidity data to the server.
Step 5.3: the storage module is arranged at the server end and used for storing temperature and humidity data sent by the temperature and humidity sensor.
Step 5.4: the temperature and humidity prediction module is used for cleaning, preprocessing, normalizing and standardizing temperature and humidity data, building a Temporal _ ATT-Former neural network, predicting the temperature and humidity value of a given time interval and transmitting the result to the cloud server end;
step 5.5: the early warning module judges a temperature and humidity prediction result in the cloud server, and if the temperature and humidity prediction result is greater than a certain set temperature and humidity threshold value, an early warning signal is sent to the client.
In order to better illustrate the effectiveness of the method, 46088 pieces of humiture data of a drug depot for four months are subjected to normalization and standardization processing by adopting a zero-mean method, a time attention mechanism and an improved Transformer model are introduced to predict the humiture of the drug depot, the evaluation standard mean absolute error mae of the prediction model based on the Temporal _ ATT-Former is 0.172, and the root mean square error rmse is 0.231, so that the effect of the model is remarkably improved compared with that of the traditional machine learning and deep learning prediction algorithm.
The invention can be combined with a computer system so as to complete the automatic prediction of the humiture of the drug storage room.
The invention creatively provides a medicine storehouse temperature and humidity prediction method based on Temporal _ ATT-Former, and a medicine storehouse temperature and humidity prediction model is obtained through multiple experiments.
The medicine warehouse temperature and humidity prediction method based on Temporal _ ATT-Former can be used for medicine warehouse temperature and humidity prediction and can also be used for prediction based on other time prediction sequences.
Details not described herein are well within the skill of those in the art.
Claims (8)
1. A drug warehouse temperature and humidity prediction method based on Temporal _ ATT-Former is characterized by comprising the following steps:
step 1: acquiring historical data of temperature and humidity of a drug warehouse environment, and preprocessing the historical data;
step 2: establishing a Temporal _ ATT-Former temperature and humidity prediction model;
and step 3: performing temperature and humidity prediction model training on the temperature _ ATT-Former according to the acquired temperature and humidity data characteristics of the drug storeroom;
and 4, step 4: testing the effect of the Temporal _ ATT-Former model;
and 5: and establishing a temperature and humidity early warning system of the drug storage room.
2. The method for predicting the humiture of the pharmacy based on Temporal _ ATT-Former as claimed in claim 1, wherein the specific method in step 1 is as follows:
step 1.1: for each collection point of drug storeroomAcquiring a first characteristic parameter TH of the drug storage room by using a temperature and humidity sensor1The method comprises the steps of numbering a temperature and humidity sensor, the position of a collection point, temperature data, humidity data, a collected time sequence and alarm information;
step 1.2: preprocessing the first characteristic data, removing repeated data, filling up lost data to obtain second characteristic data TH2The method comprises the steps of collecting time sequence, temperature and humidity data and alarm information;
step 1.3: respectively acquiring cycle data of each day, each week and each month from the time series temperature and humidity data acquired in the step 1.2, and normalizing and standardizing the data by adopting a zero-mean method to obtain third characteristic data TH3。
3. The method for predicting the humiture of the pharmacy based on Temporal _ ATT-Former as claimed in claim 1, wherein the specific method in the step 2 is as follows:
step 2.1: the Temporal _ ATT-Former comprises an Encoder module and a Decoder module;
step 2.2: establishing an Encoder module which comprises an input layer, a convolutional layer, a time attention layer and a normalization and characteristic output layer;
step 2.3: and establishing a Decoder module which comprises an input layer, a multi-head time attention covering layer, a multi-head time attention layer, a full connection layer and an output layer.
4. The method for predicting the temperature and the humidity of the pharmacy based on the Temporal _ ATT-Former as claimed in claim 3, wherein the Encoder module is specifically:
a first layer: an input layer comprising temperature and humidity characteristics, local time series characteristics, and global time series characteristics, namely temperature and humidity input at the time tWherein the content of the first and second substances,the characteristics of the temperature and the humidity are adopted,is a local time series characteristic of the time series,is a global time series feature;
a second layer: the convolution layer adopts one-dimensional convolution operation to respectively extract the characteristics of the humiture of the drug storage, namely
And a third layer: a multi-head time attention layer for strengthening the temperature and humidity characteristics of the time sequence of the input layer;
a fourth layer: and a normalization and feature output layer.
5. The method for predicting the temperature and the humidity of the pharmacy based on the Temporal _ ATT-Former as claimed in claim 3, wherein the Decoder module is specifically:
a first layer: input layer, input of temperature and humidity data shifted forward by T time stamps based on time series, i.e.
A second layer: an attention layer including a multi-temporal attention layer and a multi-temporal attention layer;
and a third layer: a fully-connected layer;
a fourth layer: and an output layer for outputting the predicted temperature and humidity data of the time series.
6. The method for predicting the humiture of the pharmacy based on Temporal _ ATT-Former as claimed in claim 1, wherein the specific method in step 3 is as follows:
step 3.1: dividing a data set, namely dividing the temperature and humidity data set TH of the drug storage room preprocessed in the step 13According to a certain proportionDividing a training set and a test set in proportion, and respectively representing as train and test;
step 3.2: setting a training round epoch and a batch round batch;
step 3.3: training a humiture prediction model of a drug warehouse, inputting train into a Temporal _ ATT-Former network for training, and storing the trained model.
7. The method for predicting the humiture of the pharmacy based on Temporal _ ATT-Former as claimed in claim 1, wherein the specific method in the step 4 is
Step 4.1: verifying the performance of the Temporal _ ATT-Former model, loading the stored model, and inputting the test into the model to obtain a prediction result r of the test set;
step 4.2: and comparing the temperature and humidity r predicted by the model with the real temperature and humidity value, calculating the average absolute error and the root mean square error, adjusting the parameters of the model, and optimizing the network structure.
8. The method for predicting the humiture of the pharmacy based on Temporal _ ATT-Former as claimed in claim 1, wherein the specific method in step 5 is as follows:
step 5.1: the temperature and humidity early warning system for the drug storage room comprises a temperature and humidity acquisition module, a storage module, a temperature and humidity prediction module and an early warning module;
step 5.2: the temperature and humidity acquisition module comprises temperature and humidity data of each acquisition point of the drug warehouse and transmits the temperature and humidity data to the server end;
step 5.3: the storage module is arranged at the server end and used for storing temperature and humidity data sent by the temperature and humidity sensor;
step 5.4: the temperature and humidity prediction module is used for cleaning, preprocessing, normalizing and standardizing temperature and humidity data, building a Temporal _ ATT-Former neural network, predicting the temperature and humidity value of a given time interval and transmitting the result to the cloud server end;
step 5.5: the early warning module judges a temperature and humidity prediction result in the cloud server, and if the temperature and humidity prediction result is greater than a certain set temperature and humidity threshold value, an early warning signal is sent to the client.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110719129.8A CN113537459B (en) | 2021-06-28 | 2021-06-28 | Drug warehouse temperature and humidity prediction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110719129.8A CN113537459B (en) | 2021-06-28 | 2021-06-28 | Drug warehouse temperature and humidity prediction method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113537459A true CN113537459A (en) | 2021-10-22 |
CN113537459B CN113537459B (en) | 2024-04-26 |
Family
ID=78096979
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110719129.8A Active CN113537459B (en) | 2021-06-28 | 2021-06-28 | Drug warehouse temperature and humidity prediction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113537459B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024016586A1 (en) * | 2022-07-18 | 2024-01-25 | 中国电信股份有限公司 | Machine room temperature control method and apparatus, and electronic device and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016116958A1 (en) * | 2015-01-19 | 2016-07-28 | 株式会社東芝 | Sequential data analysis device and program |
CN109708689A (en) * | 2018-11-14 | 2019-05-03 | 沈阳理工大学 | Based on the chain pharmacy temperature and humidity early warning system for improving long Memory Neural Networks in short-term |
CN110188167A (en) * | 2019-05-17 | 2019-08-30 | 北京邮电大学 | A kind of end-to-end session method and system incorporating external knowledge |
CN111061862A (en) * | 2019-12-16 | 2020-04-24 | 湖南大学 | Method for generating abstract based on attention mechanism |
CN111488984A (en) * | 2020-04-03 | 2020-08-04 | 中国科学院计算技术研究所 | Method for training trajectory prediction model and trajectory prediction method |
CN112330215A (en) * | 2020-11-26 | 2021-02-05 | 长沙理工大学 | Urban vehicle demand prediction method, equipment and storage medium |
CN112329760A (en) * | 2020-11-17 | 2021-02-05 | 内蒙古工业大学 | Method for recognizing and translating Mongolian in printed form from end to end based on space transformation network |
CN112529283A (en) * | 2020-12-04 | 2021-03-19 | 天津天大求实电力新技术股份有限公司 | Comprehensive energy system short-term load prediction method based on attention mechanism |
-
2021
- 2021-06-28 CN CN202110719129.8A patent/CN113537459B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016116958A1 (en) * | 2015-01-19 | 2016-07-28 | 株式会社東芝 | Sequential data analysis device and program |
CN109708689A (en) * | 2018-11-14 | 2019-05-03 | 沈阳理工大学 | Based on the chain pharmacy temperature and humidity early warning system for improving long Memory Neural Networks in short-term |
CN110188167A (en) * | 2019-05-17 | 2019-08-30 | 北京邮电大学 | A kind of end-to-end session method and system incorporating external knowledge |
CN111061862A (en) * | 2019-12-16 | 2020-04-24 | 湖南大学 | Method for generating abstract based on attention mechanism |
CN111488984A (en) * | 2020-04-03 | 2020-08-04 | 中国科学院计算技术研究所 | Method for training trajectory prediction model and trajectory prediction method |
CN112329760A (en) * | 2020-11-17 | 2021-02-05 | 内蒙古工业大学 | Method for recognizing and translating Mongolian in printed form from end to end based on space transformation network |
CN112330215A (en) * | 2020-11-26 | 2021-02-05 | 长沙理工大学 | Urban vehicle demand prediction method, equipment and storage medium |
CN112529283A (en) * | 2020-12-04 | 2021-03-19 | 天津天大求实电力新技术股份有限公司 | Comprehensive energy system short-term load prediction method based on attention mechanism |
Non-Patent Citations (1)
Title |
---|
汤雪: "基于深度学习的文本情感分类研究", 《中国优秀硕士学位论文全文数据库》, no. 12, pages 138 - 2120 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2024016586A1 (en) * | 2022-07-18 | 2024-01-25 | 中国电信股份有限公司 | Machine room temperature control method and apparatus, and electronic device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN113537459B (en) | 2024-04-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ma et al. | Real-time monitoring of water quality using temporal trajectory of live fish | |
CN113919231B (en) | PM2.5 concentration space-time change prediction method and system based on space-time diagram neural network | |
Trebing et al. | Wind speed prediction using multidimensional convolutional neural networks | |
Zhang et al. | Evaluation and comparison of gross primary production estimates for the Northern Great Plains grasslands | |
CN110737732A (en) | electromechanical equipment fault early warning method | |
CN112735094A (en) | Geological disaster prediction method and device based on machine learning and electronic equipment | |
CN110456026B (en) | Soil moisture content monitoring method and device | |
CN111126662A (en) | Irrigation decision making method, device, server and medium based on big data | |
CN115660262B (en) | Engineering intelligent quality inspection method, system and medium based on database application | |
Guillén-Navarro et al. | A deep learning model to predict lower temperatures in agriculture | |
Aburasain et al. | Drone-based cattle detection using deep neural networks | |
CN110849821B (en) | Black and odorous water body remote sensing identification method based on Bayesian theorem | |
CN113537459A (en) | Method for predicting humiture of drug storage room | |
CN115545334A (en) | Land use type prediction method, land use type prediction device, electronic device, and storage medium | |
US20210056410A1 (en) | Sensor data forecasting system for urban environment | |
Nasir et al. | Deep learning detection of types of water-bodies using optical variables and ensembling | |
Belotti et al. | Long‐term analysis of persistence and size of swallow and martin roosts in the US Great Lakes | |
CN112016744B (en) | Forest fire prediction method and device based on soil moisture and storage medium | |
CN115062686A (en) | Multi-KPI (Key performance indicator) time sequence abnormity detection method and system based on multi-angle features | |
CN113408787A (en) | Scenic spot tourist flow prediction method and device | |
Janbain et al. | Use of long short-term memory network (LSTM) in the reconstruction of missing water level data in the River Seine | |
CN113256444A (en) | Low-voltage transformer area household transformation relation identification method and device | |
Zhang et al. | A three-dimensional convolutional-recurrent network for convective storm nowcasting | |
CN107316105B (en) | Large-area agricultural prediction system | |
CN116842351B (en) | Coastal wetland carbon sink assessment model construction method, assessment method and electronic equipment |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |