CN110598937A - CO poisoning prediction method based on meteorological data - Google Patents
CO poisoning prediction method based on meteorological data Download PDFInfo
- Publication number
- CN110598937A CN110598937A CN201910878879.2A CN201910878879A CN110598937A CN 110598937 A CN110598937 A CN 110598937A CN 201910878879 A CN201910878879 A CN 201910878879A CN 110598937 A CN110598937 A CN 110598937A
- Authority
- CN
- China
- Prior art keywords
- poisoning
- cases
- meteorological data
- predicted
- grade
- 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
Links
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention provides a meteorological data-based CO poisoning prediction method, which comprises the following steps: acquiring historical meteorological data and the number of CO poisoning cases of an area to be predicted; dividing the number of cases of CO poisoning into k grades; selecting significant correlated effective characteristics by pearson correlation analysis and partial correlation analysis; according to the selected effective characteristic set X, constructing an ordered multi-classification Logistic regression model of the CO poisoning risk level y; acquiring meteorological data of the area to be predicted on the day needing prediction, substituting the meteorological data into the regression model to calculate the probability P of each CO poisoning risk levelj，PjThe maximum value max (p) of (a) corresponds to the value of j, which is the predicted CO poisoning risk level. The CO poisoning prediction method based on meteorological data can predict the CO poisoning risk level, and the prediction of CO poisoning is simple, convenient and quick.
Description
Technical Field
The invention relates to a CO poisoning prediction method, in particular to a CO poisoning prediction method based on meteorological data.
Background
A non-occupational CO (carbon monoxide) poisoning prediction model of the Beijing area is constructed by a professional weather station in the Beijing city by utilizing weather forecast factors and derived factors in the Beijing area, 1418 samples of CO poisoning cases from 2, 13 days in 2002 to 12, 31 days in 2005 by a Beijing emergency center are collected, and a CO poisoning index (4-level) forecasting and corresponding risk level evaluation mode is established by adopting a quasi-multiple regression index probability classification technology according to local weather conditions. This mode is designed as follows: y isi＝YWi(B0+∑BjXj) In the formula, YiThe regression fitting value of the CO poisoning meteorological index is defined as the average number of CO poisoning people per ten million of population; y isWi(i 1, 2.. n.) is the background harmonic value of the secondary CO poisoning person, XjFor the predictor of the entry equation, Bj(i-0, 1, 2.. said., m) are the corresponding regression coefficients, m and n identify the number of factors entering the equation and the maximum number of samples in the forecast year, respectively, but the prediction model has its limitations: 1. the poisoning data only come from 16 first-aid centers in the urban area of Beijing, actually, a patient suffering from poisoning does not pass through an first-aid station but directly arrives at a hospital, and the researched data may be obviously less than the actual data; 2. the weather forecast factors and derivative factors of the area are used for prediction, the derivative factors are many and are difficult to obtain at the same time in actual prediction, and the prediction difficulty is increased; 3. the prediction model is profound, not intuitive enough and not easy to understand. The climate and life habit of each place are different, and the places are influenced by economy and various conditions, the weather bureaus and the weather bureausThe elements of the meteorological factors and environmental factors that can be obtained by the environmental protection agency are also different, so it is necessary to establish a prediction method suitable for the risk of CO poisoning in local areas according to local conditions.
Disclosure of Invention
In view of one of the above technical problems, a method for predicting CO poisoning based on meteorological data is provided, which can predict a CO poisoning risk level, and the prediction of CO poisoning is simpler and more convenient.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a CO poisoning prediction method based on meteorological data comprises the following steps:
acquiring historical meteorological data of an area to be predicted and the number y' of CO poisoning cases, wherein the meteorological data comprises a plurality of characteristics;
the number of cases of CO poisoning y' is divided into k grades from small to large: 1,2 …, k, obtaining a CO poisoning risk grade set y;
the acquired number y 'of cases of CO poisoning in the region and historical meteorological data form a data set, correlation among all features in the historical meteorological data is detected through pearson correlation analysis, a net correlation coefficient and a corresponding P value between each feature and the number y' of cases of CO poisoning after influences of other features are eliminated are obtained through partial correlation analysis, and the net correlation coefficient and the corresponding P value are obtained through the P value<Significant relevant significant features are selected from the range of 0.05 to form a significant feature set X (X ═ X)1,X2,…,Xm) Wherein X isiI is 1,2, … m represents the ith valid feature in the valid feature set X;
according to the selected effective characteristic set X, constructing an ordered multi-classification Logistic regression model of the CO poisoning risk level y:
logit[P(y≤j)]＝βj0+β1X1+...+βmXm
wherein, Pj＝P(y＝jX)，j＝1,2,…,k,PjRepresenting the probability that y takes the order j, P1+P2+...+Pk1, the model has k-1 intercepts betaj0M regression coefficients beta1，β2，...βmThe regression model corresponds to a probability model of the form:
wherein j is 1,2, …, k-1
Acquiring meteorological data of the area to be predicted at the required prediction time, substituting the meteorological data into the probability model to calculate the probability P of each CO poisoning risk levelj，PjThe maximum value max (p) of (a) corresponds to the value of j, which is the predicted CO poisoning risk level.
Further, the characteristics in the meteorological data comprise average air temperature, 24h variable temperature, average air pressure, 24h variable pressure, average humidity, highest wind direction and average wind speed.
Further, the CO poisoning risk level is divided into 5 levels, wherein the level 1 indicates that meteorological conditions do not cause CO poisoning; grade 2 indicates that meteorological conditions are unlikely to cause CO poisoning; level 3 indicates that meteorological conditions may induce CO poisoning; grade 4 indicates that meteorological conditions are more likely to cause CO poisoning; a rating of 5 indicates that meteorological conditions are highly likely to cause CO poisoning.
Further, the dividing the number y' of the cases of CO poisoning into 5 grades from small to large specifically comprises:
when the number y' of cases of CO poisoning is 0, grade 1;
grade 2 when the number y' of cases of CO poisoning is 1;
when the number y' of the CO poisoning cases is more than 1 and less than or equal to 3, the grade is 3;
when the number y' of the CO poisoning cases is more than 3 and less than or equal to 7, the grade is 4;
when the number of cases of CO poisoning y' is more than 7, grade 5.
Further, the historical meteorological data and the number y' of cases of CO poisoning are data of the first two years of the current year to be predicted.
Further, the meteorological data come from a city weather station of the region to be predicted; the number of cases of CO poisoning y' comes from the hospital in the area to be predicted.
Further, the number y' of CO poisoning cases excludes CO poisoning cases due to suicide, suicidal miss, and accident.
Further, the number of cases of CO poisoning originated from a designated hospital treated by the local health care committee for CO poisoning patients.
Further, the ordered multi-classification Logistic regression model of the CO poisoning risk level y is as follows:
logitPj＝βj0-0.1946*X1+0.2236*X2-0.3565*X3
wherein j is 1,2,3, 4; x1Represents the average air temperature X2Denotes the 24h temperature change, X3Represents the average wind speed; beta is a10＝-5.3212；β20＝-4.1478；β30＝-3.2183；β40-2.1961; when j is 5, P5By the formula P5＝1-P1-P2-P3-P4To calculate.
Due to the adoption of the technical scheme, the invention has the following beneficial effects:
1. according to the CO poisoning prediction method based on the meteorological data, the number of the CO poisoning cases is converted into the risk grade of CO poisoning according to historical meteorological data and the number of the CO poisoning cases of the area to be predicted, the correlation relation among all features in the meteorological data is obtained through pearson correlation analysis and partial correlation analysis, effective features are selected from the features according to local actual conditions to construct an ordered multi-classification Logistic regression model to predict the risk grade of CO poisoning, the meteorological science and technology service level for preventing CO poisoning is improved, and the incidence rate of CO poisoning and the fatality rate of CO causing are reduced. Meanwhile, the CO poisoning prediction method based on the meteorological data only needs to obtain local meteorological data during prediction, the data are easy to obtain, effective characteristics can be selected from the meteorological data for prediction according to local actual conditions, the accurate prediction effect can be achieved, the meteorological data are not difficult to obtain, cross-department coordination is not needed, and the CO poisoning prediction is simpler, more convenient and more rapid.
2. According to the method for predicting CO poisoning based on meteorological data, the number of CO poisoning cases is derived from four comprehensive hospitals which are configured with hyperbaric oxygen therapy in Liuzhou city, and the four hospitals are specified hospitals for treating patients with carbon monoxide poisoning by the Ministry of defense in Liuzhou city, so that the data source is more comprehensive and reliable.
Drawings
Fig. 1 is a flowchart of a method for predicting CO poisoning based on meteorological data according to the present invention.
Fig. 2 is a statistical chart of the number of cases of CO poisoning in liuzhou city in 2015-2017.
Fig. 3 is a utility box plot of the risk of CO poisoning in 2017 predicted by the CO poisoning prediction method of the present invention, in which the abscissa is the predicted risk level of CO poisoning, and the ordinate is the actual number of cases of CO poisoning corresponding to the predicted level.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used in the description of the invention herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
Referring to fig. 1, a preferred embodiment of the present invention provides a method for predicting CO poisoning based on meteorological data, comprising the following steps:
step S1: acquiring historical meteorological data of an area to be predicted and the number y' of cases of CO poisoning, wherein the meteorological data comprises a plurality of characteristics. In the present embodiment, the characteristics in the meteorological data include an average air temperature (c), a 24-hour temperature change (c), an average air pressure (hPa), a 24-hour pressure change (hPa), an average humidity (%), a maximum wind direction (c), and an average wind speed (m/s). The historical meteorological data and the number of cases of CO poisoning y' are preferably data of the first two years of the year to be predicted. Wherein the meteorological data is from a central office of a region to be forecasted; the number of cases of CO poisoning y' comes from a hospital with hyperbaric oxygen in the area to be predicted. Preferably, the number y' of cases of CO poisoning excludes cases of CO poisoning due to suicide, suicidal miss and collusion to further improve the accuracy of prediction.
Step S2: the number of cases of CO poisoning y' is divided into k grades from small to large: 1,2 …, k, obtaining a set y of CO poisoning risk levels. Preferably, the number of cases of CO poisoning y' can be graded from small to large in 5, wherein a grade of 1 indicates that meteorological conditions do not induce CO poisoning; grade 2 indicates that meteorological conditions are unlikely to cause CO poisoning; level 3 indicates that meteorological conditions may induce CO poisoning; grade 4 indicates that meteorological conditions are more likely to cause CO poisoning; a rating of 5 indicates that meteorological conditions are highly likely to cause CO poisoning.
Step S3: the acquired CO poisoning case number y 'and meteorological data in the region form a data set, correlation among all features in the meteorological data is detected through pearson correlation analysis, a net correlation coefficient and a corresponding P value (probability value) between each feature and the CO poisoning case number y' after influences of other features are eliminated are obtained through partial correlation analysis, and the net correlation coefficient and the corresponding P value (probability value) are obtained through the P value<Significant relevant significant features are selected from the range of 0.05 to form a significant feature set X (X ═ X)1,X2,…,Xm) Wherein X isiAnd i is 1,2, … m, which represents the ith feature in the valid feature set X.
Step S4: according to the selected effective characteristic set X, constructing an ordered multi-classification Logistic regression model of the CO poisoning risk level y:
logit[P(y≤j)]＝βj0+β1X1+...+βmXm(formula 2)
Wherein, Pj＝P(y＝jX)，j＝1,2,…,k,PjRepresenting the probability that y takes the order j, P1+P2+...+Pk1, the model has k-1 intercepts betaj0M regression coefficients beta1，β2，...βmThe regression model corresponds to a probability model of the form:
formula 1 and formula 2 are different expressions of the regression model, wherein formula 2 is the core of the regression model.
Step S5: acquiring meteorological data of the area to be predicted on the day needing prediction, substituting the meteorological data into the probability model to calculate the probability P of each CO poisoning risk levelj，PjThe maximum value max (p) of (a) corresponds to the value of j, which is the predicted CO poisoning risk level.
The following describes a method for predicting CO poisoning based on meteorological data according to a specific example.
Step S1: acquiring historical meteorological data of an area to be predicted and the number y' of cases of CO poisoning, wherein the meteorological data comprises a plurality of characteristics.
In this embodiment, the area to be predicted is Guangxi Liuzhou city, a prediction model is constructed by using meteorological data and the number of cases of CO poisoning in 2015-2016 of Guangxi Liuzhou city as historical data, and the constructed prediction model is used for predicting the CO poisoning condition in 2017 Guangxi Liuzhou city and inspecting the prediction result. The number of cases of CO poisoning was derived from the poisoning data collected during 2015-2017 by the emergency medical department of the suzhou city with hyperbaric oxygen chamber hospital, and the cases of CO poisoning caused by suicide, suicidal attempted treatments and accidents were excluded. The four hospitals are appointed hospitals for treating patients with carbon monoxide poisoning by the ministry of health care in Liuzhou city, so that the data source is more comprehensive and reliable. Meteorological data during 2015-2017 was obtained from the Liuzhou central office. Table 1 shows the meteorological data and CO poisoning case data acquired daily from 2015 to 2017 years in Liuzhou city (due to the fact thatMore data, table 1 shows only partial data), in table 1, x1-x7Respectively represent 7 characteristics of average air temperature (DEG C), 24h temperature change (DEG C), average air pressure (hPa), 24h pressure change (hPa), average humidity (%), maximum wind direction (DEG C) and average wind speed (m/s). y' represents the number of cases of CO poisoning; y represents a CO poisoning risk level.
TABLE 1
Step S2: the number of cases of CO poisoning y' is divided into k grades from small to large: 1,2 …, k, obtaining a set y of CO poisoning risk levels.
In this example, statistics of the number of cases of CO poisoning in the year 2015-2016 in Liuzhou city are ranked according to the following 5 grades:
the number of CO poisoning people is 0, grade 1;
the number of CO poisoning people is 1, grade 2;
the number of CO poisoning people is more than 1 and less than or equal to 3, and the grade is 3;
the number of CO poisoning people is more than 3 and less than or equal to 7, and the grade is 4;
the number of CO poisoning people is more than 7, and the grade is 5;
grade 1: meteorological conditions do not cause CO poisoning. Grade 2: the meteorological conditions are not easy to cause CO poisoning. Grade 3: meteorological conditions may trigger CO poisoning. Grade 4: the meteorological conditions are easy to cause CO poisoning. Grade 5: the meteorological conditions are very likely to cause CO poisoning.
Step S3: detecting the correlation among the characteristics in the meteorological data of 2015-2016 (see table 2), wherein the data in table 2 are correlation coefficient P 'values, the data outside the included number are correlation coefficients, and the data inside the parentheses are corresponding P' values, by pearson correlation analysis, obtaining the net correlation coefficient between each characteristic and the number y 'of cases with CO poisoning after eliminating the influence of other characteristics and the corresponding net correlation coefficient and the number y' of cases with CO poisoning by partial correlation analysisP-value (see Table 3, data in Table 3 are net correlation coefficient and corresponding P-value), in Table 3, data outside the parenthesis are net correlation coefficient, data in the parenthesis are P-value, and P-value<Significant relevant significant features are selected from the range of 0.05 to form a significant feature set X (X ═ X)1,X2,…,Xm) Wherein X isiAnd i is 1,2, … m, which represents the ith feature in the valid feature set X.
In this embodiment, according to the person correlation analysis, a conclusion that there is a strong correlation between the feature values is obtained, and 3 features having a P value less than 0.05 in the partial correlation analysis are selected, that is: average air temperature (x)1) 24h variable temperature (x)2) Average wind speed (x)7) Which constitutes the set of valid features X, X ═ X (X)1,X2,X3)。
TABLE 2
TABLE 3
| y’ | |
| x1 | -0.19170885612707(1.75577486478423e-10) |
| x2 | 0.267896166897144(2.26785587179656e-19) |
| x3 | 0.00666154315887346(0.82611888463062) |
| x4 | -0.0143377938428245(0.636323370654626) |
| x5 | 0.0415903301622056(0.170023247446175) |
| x6 | 0.0871180326434513(0.00399712464342717) |
| x7 | -0.144618039265648(1.63217908812047e-06) |
Step S4: according to the selected effective characteristic set X, constructing an ordered multi-classification Logistic regression model of the CO poisoning risk level y:
in this embodiment, the above-mentioned 3 characteristics are selected from the data in 2015-2016: the average air temperature, the 24h variable temperature and the average wind speed are taken as effective characteristics and are sequentially recorded as X1,X2,X3，X1,X2,X3Forming an effective characteristic set; adopting R software for auxiliary calculation to construct an ordered multi-classification Logistic regression model of the CO poisoning risk level y, wherein the result is as follows:
logit[P(y≤j)]＝βj0+β1X1+...+β3X3，j＝1，2，...4
the four calculated intercept values are: beta is a10＝-5.3212；β20＝-4.1478；β30＝-3.2183；β40-2.1961; the calculated 3 regression coefficient values are: beta is a1＝0.1946；β2＝-0.2236；β30.3565, the prediction model was obtained as follows:
logitPj＝βj0-0.1946*X1+0.2236*X2-0.3565*X3(formula 4)
In formula 4, j is 1,2,3, 4; x1Means mean gasTemperature, X2Denotes the 24h temperature change, X3Represents the average wind speed; beta is a10＝-5.3212；β20＝-4.1478；β30＝-3.2183；β40-2.1961. When j is 5, P5By the formula P5＝1-P1-P2-P3-P4To calculate. Equation 4 can be used for CO poisoning risk prediction in southern regions in other cities with similar meteorological environments as in the city of suzhou.
Note that, since the model is created by using the polr () function of the R language MASS package, where the model defined in this package is logic (p) ═ a1- (b1 × 1+ b2 × 2+ b3 × 3+ b4 × 4+ b5 × 5+ b6 × 6, the regression coefficients obtained are opposite to the commonly defined signs.
The embodiment also performs fitting test on the model, the test statistic is 1762.582, the statistic obeys chi-square distribution with the degree of freedom of 3, p is less than 0.01, and the model has statistical significance.
Step S5: acquiring meteorological data of the area to be predicted on the day needing prediction, substituting the meteorological data into the regression model to calculate the probability P of each CO poisoning risk levelj，PjThe maximum value max (p) of (a) corresponds to the value of j, which is the predicted CO poisoning risk level.
In this embodiment, the above model is used to predict CO poisoning in 2017 of liuzhou city, and the prediction results are shown in fig. 2-3 and table 4:
TABLE 4
As can be seen from fig. 2 to 3 and table 4, for all 101 days in which the number of poisoned people is greater than 3, the risk level is predicted to be 76 days or more, the effective rate is 75%, for 55 typical CO poisoning days in which the number of poisoned people is greater than 7, the risk level is predicted to be 3 days or more, the effective rate is 85%, and the effect on the prediction of CO poisoning is good.
The meteorological data-based CO poisoning prediction method has large sample amount, 3 characteristics, intercept and regression coefficients are obtained, and the prediction method can be applied to CO poisoning prediction of various cities and regions. The prediction model of the invention only uses basic elements which can be collected by various meteorological offices such as air temperature, air pressure, humidity, wind speed, wind direction and the like, and has commonality.
It is understood that, in other embodiments, step S3 may be performed first and then step S2 is performed, or step S3 and step S2 may be performed simultaneously.
It is understood that in other embodiments, the CO poisoning risk level may be divided into other numbers according to needs.
It is understood that in other embodiments, data to be predicted 1 year or 3 years before the current year may be collected for model construction.
The above description is intended to describe in detail the preferred embodiments of the present invention, but the embodiments are not intended to limit the scope of the claims of the present invention, and all equivalent changes and modifications made within the technical spirit of the present invention should fall within the scope of the claims of the present invention.
Claims (9)
1. A CO poisoning prediction method based on meteorological data is characterized by comprising the following steps:
acquiring historical meteorological data of an area to be predicted and the number y' of CO poisoning cases, wherein the meteorological data comprises a plurality of characteristics;
the number of cases of CO poisoning y' is divided into k grades from small to large: 1,2 …, k, obtaining a CO poisoning risk grade set y;
the acquired number y 'of cases of CO poisoning in the region and historical meteorological data form a data set, correlation among all features in the historical meteorological data is detected through pearson correlation analysis, a net correlation coefficient and a corresponding P value between each feature and the number y' of cases of CO poisoning after influences of other features are eliminated are obtained through partial correlation analysis, and the net correlation coefficient and the corresponding P value are obtained through the P value<Significant relevant significant features are selected from the range of 0.05 to form a significant feature set X (X ═ X)1,X2,…,Xm) Wherein X isiI is 1,2, … m represents the ith valid feature in the valid feature set X;
according to the selected effective characteristic set X, constructing an ordered multi-classification Logistic regression model of the CO poisoning risk level y:
logit[P(y≤j)]＝βj0+β1X1+...+βmXm
wherein, Pj＝P(y＝jX)，j＝1,2,…,k,PjRepresenting the probability that y takes the order j, P1+P2+...+Pk1, the model has k-1 intercepts betaj0M regression coefficients beta1，β2，...βmThe regression model corresponds to a probability model of the form:
wherein j is 1,2, …, k-1
Acquiring meteorological data of the area to be predicted at the required prediction time, substituting the meteorological data into the probability model to calculate the probability P of each CO poisoning risk levelj，PjThe maximum value max (p) of (a) corresponds to the value of j, which is the predicted CO poisoning risk level.
2. The CO poisoning prediction method of claim 1, wherein the features in the meteorological data comprise an average air temperature, a 24h temperature change, an average air pressure, a 24h pressure change, an average humidity, a highest wind direction, and an average wind speed.
3. The CO poisoning prediction method of claim 1, wherein the CO poisoning risk level is divided into 5 levels, wherein a level 1 indicates that meteorological conditions do not induce CO poisoning; grade 2 indicates that meteorological conditions are unlikely to cause CO poisoning; level 3 indicates that meteorological conditions may induce CO poisoning; grade 4 indicates that meteorological conditions are more likely to cause CO poisoning; a rating of 5 indicates that meteorological conditions are highly likely to cause CO poisoning.
4. The method for predicting CO poisoning of claim 3, wherein the dividing the number of cases of CO poisoning y' into 5 grades from small to large specifically comprises:
when the number y' of cases of CO poisoning is 0, grade 1;
grade 2 when the number y' of cases of CO poisoning is 1;
when the number y' of the CO poisoning cases is more than 1 and less than or equal to 3, the grade is 3;
when the number y' of the CO poisoning cases is more than 3 and less than or equal to 7, the grade is 4;
when the number of cases of CO poisoning y' is more than 7, grade 5.
5. The method of claim 1, wherein the historical meteorological data and the number of cases of CO poisoning y' are data of the first two years of the current year to be predicted.
6. The CO poisoning prediction method of claim 1, wherein the meteorological data is from a city station of a region to be predicted; the number of cases of CO poisoning y' comes from the hospital in the area to be predicted.
7. The method of predicting CO poisoning of claim 6, wherein the number of cases of CO poisoning y' excludes cases of CO poisoning due to suicide, suicidal misfires, and accidents.
8. The method of predicting CO poisoning of claim 1, wherein the number of cases of CO poisoning is from a designated hospital treated by a local health care committee for carbon monoxide poisoning patients.
9. The CO poisoning prediction method of claim 1, wherein the ordered multi-class Logistic regression model for the CO poisoning risk level y is:
logitPj＝βj0-0.1946*X1+0.2236*X2-0.3565*X3
wherein j is 1,2,3, 4; x1Represents the average air temperature X2Is shown in (2)Changing the temperature and X within 4h3Represents the average wind speed; beta is a10＝-5.3212；β20＝-4.1478；β30＝-3.2183；β40-2.1961; when j is 5, P5By the formula P5＝1-P1-P2-P3-P4To calculate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910878879.2A CN110598937A (en) | 2019-09-18 | 2019-09-18 | CO poisoning prediction method based on meteorological data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910878879.2A CN110598937A (en) | 2019-09-18 | 2019-09-18 | CO poisoning prediction method based on meteorological data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110598937A true CN110598937A (en) | 2019-12-20 |
Family
ID=68860544
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910878879.2A Pending CN110598937A (en) | 2019-09-18 | 2019-09-18 | CO poisoning prediction method based on meteorological data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110598937A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111967655A (en) * | 2020-07-28 | 2020-11-20 | 中国南方电网有限责任公司 | Short-term load prediction method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567808A (en) * | 2010-12-31 | 2012-07-11 | 北京工业大学 | Method for forecasting and warning accident consequence of major hazard installation by combining with real-time meteorological information |
| CN106126484A (en) * | 2016-07-06 | 2016-11-16 | 中交第公路勘察设计研究院有限公司 | The multi-factor comprehensive ever-frozen ground ground temperature zoning methods of multiple linear regression analysis |
| CN108052528A (en) * | 2017-11-09 | 2018-05-18 | 华中科技大学 | A kind of storage device sequential classification method for early warning |
| CN110047594A (en) * | 2019-05-27 | 2019-07-23 | 北京气象在线科技有限公司 | Respiratory disease Prediction of Incidence Trend method based on weather environment monitoring data |
-
2019
- 2019-09-18 CN CN201910878879.2A patent/CN110598937A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102567808A (en) * | 2010-12-31 | 2012-07-11 | 北京工业大学 | Method for forecasting and warning accident consequence of major hazard installation by combining with real-time meteorological information |
| CN106126484A (en) * | 2016-07-06 | 2016-11-16 | 中交第公路勘察设计研究院有限公司 | The multi-factor comprehensive ever-frozen ground ground temperature zoning methods of multiple linear regression analysis |
| CN108052528A (en) * | 2017-11-09 | 2018-05-18 | 华中科技大学 | A kind of storage device sequential classification method for early warning |
| CN110047594A (en) * | 2019-05-27 | 2019-07-23 | 北京气象在线科技有限公司 | Respiratory disease Prediction of Incidence Trend method based on weather environment monitoring data |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111967655A (en) * | 2020-07-28 | 2020-11-20 | 中国南方电网有限责任公司 | Short-term load prediction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111860979B (en) | Short-term load prediction method based on TCN and IPSO-LSSVM combined model | |
| CN105069525B (en) | Round-the-clock 96 Day Load Curve Forecastings and optimization update the system | |
| Heydarpour et al. | Potential impact of air pollution on multiple sclerosis in Tehran, Iran | |
| Guest et al. | Climate and mortality in Australia: retrospective study, 1979-1990, and predicted impacts in five major cities in 2030 | |
| Witt et al. | The effects of climate change on patients with chronic lung disease: a systematic literature review | |
| AU2020104020A4 (en) | A Method for Predicting CO Poisoning Based on Data from Meteorological Conditions and Atmospheric Pollutants | |
| CN110047594A (en) | Respiratory disease Prediction of Incidence Trend method based on weather environment monitoring data | |
| CN106779129A (en) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor | |
| Agarwal et al. | Impact of temperature on morbidity: New evidence from China | |
| CN113935608A (en) | Evaluation method and device for community endowment service facilities | |
| CN110598937A (en) | CO poisoning prediction method based on meteorological data | |
| Wang et al. | The health impacts of two policies regulating SO2 air pollution: evidence from China | |
| Sato et al. | Quantitative assessment of the contribution of meteorological variables to the prediction of the number of heat stroke patients for Tokyo | |
| Charlier 1 et al. | Fuel poverty and health: a shared agenda for policy | |
| CN104143117B (en) | A kind of related coefficient extracting method between power grid specific load and daily load | |
| Filomena et al. | Unsafe temperatures, unsafe jobs: The impact of weather conditions on work-related injuries | |
| Li et al. | An efficient performance assessing index for building climatic zoning methods based on probability density functions | |
| Zeng et al. | Individualized evaluation of health cost and health risks | |
| Matyasovszky et al. | Estimating the daily Poaceae pollen concentration in Hungary by linear regression conditioning on weather types | |
| CN116862283A (en) | Method and system for evaluating and controlling environment of aged people | |
| CN110633856A (en) | CO poisoning prediction method based on meteorological and atmospheric pollutant data | |
| AU2020104025A4 (en) | A Method for Predicting CO Poisoning Based on Meteorological Data | |
| Nakazawa et al. | Change in strategic interaction after introducing policy | |
| Gong et al. | Journal of Environmental Economics and Management | |
| CN117895478A (en) | Novel human comfort evaluation method based on order relation method-entropy weight method |
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 |