CN101430735B - Protective farming mode selection method - Google Patents
Protective farming mode selection method Download PDFInfo
- Publication number
- CN101430735B CN101430735B CN2008102265088A CN200810226508A CN101430735B CN 101430735 B CN101430735 B CN 101430735B CN 2008102265088 A CN2008102265088 A CN 2008102265088A CN 200810226508 A CN200810226508 A CN 200810226508A CN 101430735 B CN101430735 B CN 101430735B
- Authority
- CN
- China
- Prior art keywords
- mrow
- similarity
- index
- indexes
- 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.)
- Expired - Fee Related
Links
- 230000001681 protective effect Effects 0.000 title claims abstract description 63
- 238000009313 farming Methods 0.000 title claims abstract description 61
- 238000010187 selection method Methods 0.000 title claims description 10
- 238000000034 method Methods 0.000 claims abstract description 37
- 239000002689 soil Substances 0.000 claims abstract description 33
- 238000004519 manufacturing process Methods 0.000 claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000013210 evaluation model Methods 0.000 claims abstract description 9
- 238000002360 preparation method Methods 0.000 claims abstract description 4
- 238000013480 data collection Methods 0.000 claims abstract description 3
- 239000011159 matrix material Substances 0.000 claims description 13
- 239000002131 composite material Substances 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 238000011835 investigation Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000013329 compounding Methods 0.000 claims 1
- 238000013507 mapping Methods 0.000 claims 1
- 238000002474 experimental method Methods 0.000 abstract description 3
- 238000010276 construction Methods 0.000 abstract 2
- 238000004088 simulation Methods 0.000 abstract 1
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 12
- 238000003971 tillage Methods 0.000 description 8
- 230000008901 benefit Effects 0.000 description 6
- 229910052757 nitrogen Inorganic materials 0.000 description 6
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 3
- ZLMJMSJWJFRBEC-UHFFFAOYSA-N Potassium Chemical compound [K] ZLMJMSJWJFRBEC-UHFFFAOYSA-N 0.000 description 3
- 238000005904 alkaline hydrolysis reaction Methods 0.000 description 3
- 238000003973 irrigation Methods 0.000 description 3
- 230000002262 irrigation Effects 0.000 description 3
- 229910052698 phosphorus Inorganic materials 0.000 description 3
- 239000011574 phosphorus Substances 0.000 description 3
- 239000011591 potassium Substances 0.000 description 3
- 229910052700 potassium Inorganic materials 0.000 description 3
- 238000001556 precipitation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 239000010902 straw Substances 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 238000012271 agricultural production Methods 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000000691 measurement method Methods 0.000 description 2
- 238000010899 nucleation Methods 0.000 description 2
- KMUONIBRACKNSN-UHFFFAOYSA-N potassium dichromate Chemical compound [K+].[K+].[O-][Cr](=O)(=O)O[Cr]([O-])(=O)=O KMUONIBRACKNSN-UHFFFAOYSA-N 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 241000894007 species Species 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- DYJGATXRPZAVGS-UHFFFAOYSA-N [Sc].[Sb].[Mo] Chemical compound [Sc].[Sb].[Mo] DYJGATXRPZAVGS-UHFFFAOYSA-N 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004737 colorimetric analysis Methods 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000010411 cooking Methods 0.000 description 1
- 238000012364 cultivation method Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 230000005923 long-lasting effect Effects 0.000 description 1
- 238000000643 oven drying Methods 0.000 description 1
- -1 pH value Substances 0.000 description 1
- 239000004016 soil organic matter Substances 0.000 description 1
- 238000011425 standardization method Methods 0.000 description 1
- 238000004381 surface treatment Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000004454 trace mineral analysis Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for selecting a protective farming mode, which is characterized by comprising the following steps: 1) data collection: soil data, climate data, geographic data and production habit data of a mature area and an area to be implemented in different modes are collected; 2) preparation of mode selection: a decision unit and an index of each protective farming mode are determined according to the data of the mature area; 3) construction of a projection pursuit analysis model; 4) construction of a similarity evaluation model; 5) model selection: the protective farming mode corresponding to the mature area with the highest similarity to the area to be implemented is selected as the protective farming mode of the area to be implemented according to the level of similarity. The invention selects a proper protective farming mode for the area to be implemented by constructing a protective farming mode selection mode, and the whole process is completed by computer analog simulation without field experiment, thereby improving the efficiency of protective farming mode selection, and reducing the cost for popularizing protective farming.
Description
Technical Field
The invention relates to a mode identification technology, in particular to a protective farming mode selection method applied to agricultural production.
Background
The protective cultivation is a new-type dry land cultivation method, and mainly includes four steps of no-tillage seeding and fertilizing, deep scarification, weed control, straw and ground surface treatment. The core of the method is no-tillage seeding, and the main operation is completed by using machinery. The protective farming technology originates from the United states of the 20 th century and the 30 th century, and at present, the countries such as the United states, Canada and Australia and the like basically adopt the protective farming which is supported by mechanization, and good effect is achieved.
Protective farming has been carried out in china for only a decade of systematic testing and research, and in some typical areas certain results have been achieved, summarizing a series of protective farming patterns suitable for a particular area. Since the soil, climate, geography, crop type, production habits, etc. vary from one region to another, it is not possible to apply the same protective farming pattern to them. In the process of popularizing the protective farming technology, suitable modes are screened mainly by field tests in areas to be implemented in the past, and although the obtained results are visual, the method is long in time consumption and high in input cost, and the types of the modes and the scale of the tests are limited.
Disclosure of Invention
In view of the above problems, the present invention aims to provide a method for selecting a protective farming mode, which is suitable for a region to be cultivated, quickly and reasonably by constructing a protective farming mode selection model and comparing the similarity between a mature region of protective farming and the region to be cultivated under non-experimental conditions.
In order to achieve the purpose, the invention adopts the following technical scheme: a protective farming mode selection method characterized by: it comprises the following steps: 1) data collection: collecting soil data, climate data, geographic data and production habit data of mature areas in different modes; collecting soil data, climate data, geographic data and production habit data of an area to be implemented; 2) mode selection preparation: determining decision units and indexes of each protective farming mode according to the data of the mature region; 3) constructing a projection pursuit analysis model: standardizing the detection and survey data of the mature region, and determining the objective weight of each index by using a projection pursuit technology according to the decision unit and the index <math> <mrow> <msubsup> <mi>β</mi> <mi>j</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <mrow> <mo>{</mo> <msubsup> <mi>β</mi> <mn>1</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>β</mi> <mn>2</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>.</mo> <mo>.</mo> <mo>.</mo> <msubsup> <mi>β</mi> <mi>p</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>}</mo> </mrow> <mo>,</mo> </mrow></math> Wherein p is the number of indices; 4) constructing a similarity evaluation model: calculating the detection and survey data of the region to be performed and the index similarity composite weight beta of each different regionj(ii) a Then calculating the similarity value between the region to be executed and the mature region, and evaluating the similarity and the similarity level of the two regions; 5) mode selection: and selecting the protective farming mode corresponding to the mature region with the highest similarity with the region to be implemented as the protective farming mode of the region to be implemented according to the similarity grade.
In the step 1), the soil data of the mature area is derived from the existing experimental results and field investigation results; the soil data of the region to be implemented is determined by a chemical method; the climate data of the mature region and the region to be executed are obtained according to the 10-year weather data weighted average; and the geographic data and the data of the production habits of the mature region and the region to be implemented are obtained by a field investigation method.
In the step 2), the decision unit corresponds to the category of the mature region, and the index is derived from comparable soil data, climate data, geographic data and production habit index in the mature region.
In the step 3), constructing the projection pursuit analysis model includes the following steps:
a) data normalization: the original data matrix X*Conversion into a standardized matrix X, XijRepresenting the jth index value of the ith decision unit for the element in the standardized matrix X;
b) linear projection: let unit vector a ═ a1，a2，…apIs the one-dimensional linear projection direction, the one-dimensional projection eigenvalue of the normalized matrix X projected onto the projection direction a is zi；
c) Constructing a projection objective function:
Q(a)＝SZ·DZ
wherein E (z) is the sequence { ziI ═ 1, …, n | } average; let rikIs the distance between the projected eigenvalues; r is the density window width, and the value range is as follows: <math> <mrow> <mi>max</mi> <mrow> <mo>(</mo> <msub> <mi>r</mi> <mi>ij</mi> </msub> <mo>)</mo> </mrow> <mo>+</mo> <mfrac> <mi>p</mi> <mn>2</mn> </mfrac> <mo>≤</mo> <mi>R</mi> <mo>≤</mo> <mn>2</mn> <mi>p</mi> <mo>;</mo> </mrow></math> rik＝|zi-zk|，i，k＝1，…，n；
d) optimizing the projection objective function q (a): optimizing the projection direction a by using the projection objective function by utilizing the objective function maximization to ensure that the projection direction in which the projection objective function Q (a) reaches the maximum value is the optimal projection direction
：
MaxQ(a)＝Sz·Dz
e) Calculating objective weight: determining objective weights for the respective indicators
In the step 4), the steps of constructing the similarity evaluation model are as follows: i) classifying indexes; ii) calculating index similarity S (A, B) between different regions, wherein A and B are two different regions; iii) determination of index similarity composite weight:
wherein, the <math> <mrow> <msubsup> <mi>β</mi> <mi>j</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <mrow> <mo>{</mo> <msubsup> <mi>β</mi> <mn>1</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>β</mi> <mn>2</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>.</mo> <mo>.</mo> <mo>.</mo> <msubsup> <mi>β</mi> <mi>p</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>}</mo> </mrow> </mrow></math> Solving by the projection pursuit analysis model; <math> <mrow> <msubsup> <mi>β</mi> <mi>j</mi> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <mrow> <mo>{</mo> <msubsup> <mi>β</mi> <mn>1</mn> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>β</mi> <mn>2</mn> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <msubsup> <mi>β</mi> <mi>p</mi> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>}</mo> </mrow> <mo>,</mo> </mrow></math> parameters for manual adjustment; iv) evaluation of inter-regional similarity: region A contains phi indexes and region B contains
The coincidence indexes of the areas A and B are sigma,
wherein the similarity between the jth similar indexes is represented by SiFrom step ii); beta is ajThe composite weight coefficient of the jth similarity element obtained in the step iii); similarity S between regions A and B*Mathematical model of (a, B):
the p indexes in the step 3) are divided into the following parts according to different measurement methods: a determined value index, a determined value and interval value index, a binary variable index, a nominal variable index and a qualitative index.
Due to the adoption of the technical scheme, the invention has the following advantages: 1. the protective farming method and the protective farming system have the advantages that the protective farming mature area and the success mode thereof are used as standard templates, the protective farming mode selection model is constructed, the proper protective farming mode is selected for the area to be implemented, the whole process is completed through computer simulation, and field experiments are not needed, so that the protective farming mode selection efficiency is improved, and the protective farming popularization cost is reduced. 2. According to the collected experimental data of the mature area of the protective farming, a decision unit and indexes are determined, a protective farming mode selection model based on similarity is constructed, and similarity evaluation is performed on the area to be implemented, so that a protective farming mode suitable for the area is determined; the protective farming mode selection model of the invention adopts different methods to determine the similarity of each index according to the difference of different index properties, and has the advantages of advanced analysis method and accurate result. The protective farming mode selection method can select proper protective farming modes for regions to be implemented with different climates, geographies and production habits.
Drawings
FIG. 1 is a schematic diagram of a protective farming mode selection method of the present invention
FIG. 2 is a flow chart of the present invention for constructing a projection pursuit analysis model
FIG. 3 is a flow chart for constructing a similarity evaluation model according to the present invention
Detailed Description
The invention is described in detail below with reference to the figures and examples.
As shown in fig. 1, the protective farming mode selection method provided by the present invention includes the steps of:
1) and (6) collecting data. Aiming at the characteristics of mature areas of protective farming and the difference of used machines, technologies and the like, the mature areas of protective farming are divided into different protective farming modes, and soil data, climate data, geographic data, production habits and other important data of the mature areas are collected. The areas where protective farming is to be performed are detected and investigated, and soil data, climate data, geographical data, production habits and other important data of the areas are collected.
2) Mode selection preparation. And determining decision units and indexes of each protective farming mode according to detection and investigation data of the mature region.
3) And constructing a projection pursuit analysis model. Firstly, standardizing detection and survey data of a mature region; secondly, determining the objective weight of each index by using a projection pursuit technology according to the decision unit and the index determined in the step 2) <math> <mrow> <msubsup> <mi>β</mi> <mi>j</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <mrow> <mo>{</mo> <msubsup> <mi>β</mi> <mn>1</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>β</mi> <mn>2</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>.</mo> <mo>.</mo> <mo>.</mo> <msubsup> <mi>β</mi> <mi>p</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>}</mo> </mrow> <mo>,</mo> </mrow></math> Where p is the number of indices.
4) And constructing a similarity evaluation model. Calculating the detection and survey data of the region to be performed and the index similarity composite weight beta of each different regionj(ii) a And then calculating a similarity value between the region to be implemented and the mature region, and evaluating the similarity and the similarity grade of the two regions.
5) And (4) selecting a mode. And selecting a protective farming mode corresponding to a mature region with high similarity to the region to be implemented as the protective farming mode of the region to be implemented according to the similarity grade.
In the step 1) of the method, the protective cultivation mature areas are firstly distinguished according to the characteristics of the areas and the difference of used machines, technologies and the like. Such as: in this embodiment, the mature area is divided into seven different areas, and each area corresponds to one or more protective farming modes (for example, but not limited thereto):
1. no-tillage whole stalk covering, shallow loosening whole stalk covering;
2. no-tillage straw cover, shallow loose straw cover;
3. no-tillage low stubble covering and shallow loosening low stubble covering;
4. no-tillage high stubble covering, shallow loosening high stubble covering;
5. covering with deep loose crushed stalks;
6. no-tillage fixed road;
7. no-tillage ridge culture.
The invention respectively calculates the similarity between the region to be executed and the seven mature regions to select a reasonable protective farming mode. Experimental data for mature and to-be-administered regions, including soil data, climate data, geographic data, production habits and other important data, are collected and collated prior to modeling. The soil data of the mature region are derived from the existing experimental results and field investigation results, and the soil data of the region to be implemented is measured by the following chemical method: the method comprises the steps of measuring the volume weight of soil by a circular knife method, measuring organic matters of the soil by a potassium dichromate volumetric method, measuring the water content of the soil by a constant-temperature oven drying method, measuring the total nitrogen of the soil by a Kelvin method, measuring the alkaline hydrolysis nitrogen of the soil by a diffusion method, measuring the available phosphorus of the soil by a molybdenum-antimony-scandium colorimetric method, measuring the available potassium of the soil by a flame photometer, and measuring the pH value of the soil by a pH meter. And the climate data such as annual average precipitation, annual average temperature and the like of the mature region and the region to be executed are obtained according to the 10-year meteorological data in a weighted average manner. Geographic data, production habits and other important data of the mature regions and regions to be implemented are obtained by a field investigation method.
In step 2) of the method, a decision unit and indexes of the protective farming mode are determined by data mining according to survey data of the mature region and researching data information such as a measuring method, a relation mode between a survey range and a research area, wherein the decision unit corresponds to the type of the mature region, and the indexes are derived from comparable soil data, climate data, geographic data, production habits and other important data in the mature region. In this embodiment, the number n of decision units is 7, and each decision unit includes 1 mature region, so that there are 7 mature regions in total. The index number p is 20 (this is merely an example, but not limited thereto), and the 20 indexes are: soil volume weight, water content, organic matters, total nitrogen, alkaline hydrolysis nitrogen, available phosphorus, quick-acting potassium, pH value, soil texture, annual average precipitation, annual average air temperature, accumulated temperature, sunshine hours, gradient, slope length, longitude, latitude, curing type, crop species and irrigation conditions.
As shown in fig. 2, in step 3), the method of the present invention includes the following steps:
a) and (6) standardizing data. Firstly, determining 7 decision units and 20 indexes to form an original data matrix X according to detection and survey data of mature regions*The original data matrix X*And converting the data into a standardized matrix X to eliminate the influence of different dimensions and magnitude differences of the indexes. Wherein xijThe element in the normalized matrix X represents the j index value of the i decision unit after normalization. Embodiments of the present invention differentiate their standardization strategies based on the actual significance of the indicators. In agricultural production, the higher the yield of a crop, the more acceptable it is to farmers, and this means crop yieldThe index belongs to a benefit type index; the fuel consumption is less and more easily accepted by farmers, and the indexes of the fuel consumption belong to cost indexes; the closer the pH value of the soil is to the intermediate value of 7, the better the soil property is, and the indexes of the pH value of the soil belong to fixed indexes, which are also called intermediate value optimal indexes. The standardization method for indexes with different meanings is as follows:
for the benefit type index with larger numerical value and better numerical value, the following processing mode is adopted:
for the cost type index with smaller numerical value and better numerical value, the following processing mode is adopted:
for the index with the optimal intermediate value, the following method is adopted:
in the formula: i is a decision unit; j is an index.
max(xj) Maximum value of jth index
min(xj) Minimum value of jth index
mid(xj) Intermediate values of the jth index
When the number of the samples of the j index is odd, the sample value with the middle value is taken as mid (x)j) (ii) a When the number of the samples of the jth index is even, the average value of two sample values with the middle value is taken as mid (x)j)。
b) And (4) linear projection. I.e. the original high dimensional data xijAnd projecting to a one-dimensional linear space for processing. Because the number of the original indexes is p, p-dimensional data is difficult to observe in space, the real characteristics of the data cannot be reflected, and dimension disaster is easy to generate. When the original high dimensional data xijAfter the data are projected to a one-dimensional linear space, the observed projection value which is substantially the p-dimensional index data reflects the characteristics of the original index data, and the data are visual and convenient. Let unit vector a ═ a1，a2，…aPIs the one-dimensional linear projection direction, the one-dimensional projection eigenvalue of the normalized matrix X projected onto the projection direction a is zi；
c) A projection objective function is constructed. Let the original high dimensional data xijInter-class distances S spread in one dimensionzAnd class inner density DzWhile taking the maximum value. The projection objective function is then expressed as:
Q(a)＝SZ·DZ
wherein E (z) is the sequence { ziI is the average of 1, …, n | }, n is the number of decision units, and p is the number of indices. Let rikThe distance between projection characteristic values is shown, R is the density window width, and the value range is as follows: <math> <mrow> <mi>max</mi> <mrow> <mo>(</mo> <msub> <mi>r</mi> <mi>ij</mi> </msub> <mo>)</mo> </mrow> <mo>+</mo> <mfrac> <mi>p</mi> <mn>2</mn> </mfrac> <mo>≤</mo> <mi>R</mi> <mo>≤</mo> <mn>2</mn> <mi>p</mi> <mo>.</mo> </mrow></math> rik＝|zi-zk|，i＝1，…，n；k＝1，...，n。 <math> <mrow> <mi>f</mi> <mrow> <mo>(</mo> <mi>t</mi> <mo>)</mo> </mrow> <mo>=</mo> <mrow> <mfenced open='{' close=''> <mtable> <mtr> <mtd> <mn>0</mn> </mtd> <mtd> <mi>t</mi> <mo>≥</mo> <mn>0</mn> </mtd> </mtr> <mtr> <mtd> <mn>1</mn> </mtd> <mtd> <mi>t</mi> <mo><</mo> <mn>0</mn> </mtd> </mtr> </mtable> </mfenced> </mrow> <mo>.</mo> </mrow></math>
d) optimizing the projection objective function q (a). For a given sample set index value, the projection objective function Q (a) changes along with the change of the projection direction a, and the projection direction which can enable the projection objective function Q (a) to reach the maximum value is the optimal projection direction
. Therefore, the projection objective function is optimized for the projection direction a using the objective function maximization:
MaxQ(a)＝Sz·Dz
e) objective weight
And (4) calculating. Optimal projection direction optimized according to step d)
The indexes are arranged according to the value size, and the objective weight of each index can be determined by using the following formulaIt reflects the contribution of the amount of information contained in the index itself to the evaluation result.
As shown in fig. 3, in step 4) of the method of the present invention, the steps of constructing the similarity evaluation model are as follows:
I) and (5) index classification. According to different index measurement methods, the indexes are classified into different categories so as to calculate the index similarity between different regions according to the characteristics of the categories. In this embodiment, the index values of the mature areas of protective farming are used as standard values, and the 20 indexes are classified as follows (as shown in table 1):
table 1: classification of indices
| Numbering | Categories | Index (I) |
| 1 | Determining a value index | Soil organic matter, total nitrogen, alkaline hydrolysis nitrogen, available phosphorus, quick-acting potassium, longitude and latitude |
| 2 | Deterministic value and range value indicators | Soil pH, water content, volume weight |
| 3 | Section value index | Annual average precipitation, annual average temperature, accumulated temperature, sunshine duration, slope and length of slope |
| 4 | Binary variable index | Type of cooking |
| 5 | Nominal variable index | Species of crop |
| 6 | Qualitative index | Texture of soil and irrigation conditions |
II) calculating index similarity of different regions. According to the index similarity characteristics in the table 1, the similarity of indexes with different properties is calculated by adopting different methods. Let two different areas A and B, their certain common index L (A) and L (B), and the similarity S (A, B) of the two indexes, and S (A, B) is the [0, 1 ].
i) And (5) determining value index similarity calculation. When indexes of the areas A and B are subjected to data standardization in the step a) in the step 3) to obtain determined values, the indexes L (A), L (B) are belonged to [0, 1 ]. The similarity between the determination values can be expressed by a distance between two points, and the closer the distance, the higher the similarity. Then there are:
ii) determining the similarity of the value and the interval value index. When the indices l (a) and l (b) in both regions are normalized, l (a) ═ a1，a2]，L(B)＝[b1，b2]，a1<a2，b1<b2And a is a1，a2，b1，b2∈[0，1]. The similarity between the determined value and the interval is related to the distance between the determined value and the midpoint of the interval and the length of the interval respectively, and the smaller the distance between the determined value and the midpoint of the interval is, and the shorter the length of the interval is, the higher the similarity is.
iii) interval value index similarity calculation: the similarity between two intervals is expressed by the coverage degree of the intervals, and the larger the coverage degree is, the higher the similarity degree is.
Wherein, Tx(AB) is the length of the overlap between section indices L (A) and L (B), and T (A) ═ a2-a1|，T(B)＝|b2-b1|；Tmax(AB)＝max(a1，a2，b1，b2)-mim(a1，a2，b1，b2)。β∈[0，1]For weighting, weighting is performed manually.
iv) binary variable index similarity calculation: binary variables represent only 0 or 1 state variables in the index. Let h denote the number of variables for which both the indices L (A) and L (B) are 1, f denote the number of variables for which the index L (A) is 1 and the index L (B) is 0, s denote the number of variables for which the index L (A) is 0 and the index L (B) is 1, t denote the number of variables for which both the indices L (A) and L (B) are 0, and d (A, B) denote the degree of dissimilarity of the indices L (A) and L (B), i.e., their Euclidean distances. The similarity of the binary variables is:
v) nominal variable index similarity calculation: the nominal variable is a generalization of a binary variable, i.e., more than two variables in the index represent states. Assuming that m represents the number of coincidences of variables in the indexes L (A) and L (B), q represents the number of all variables in the two indexes, and d (A, B) represents the dissimilarity degree of the indexes L (A) and L (B), the nominal variable similarity is as follows:
vi) qualitative index similarity calculation: and (3) the similarity of qualitative concepts is generally mapped into numerical description, and the qualitative to quantitative conversion of the similarity is realized by comparing numerical values and utilizing a similarity formula of two determined value indexes in the step i). For example, the irrigation condition status { good, general, poor } and the value {1, 0.8, 0.6, 0.4, 0.2} are mapped.
III) index similarity composite weight betajAnd (4) determining. The index similarity composite weight is used for determining the similarity of the same index in two different regions.
Wherein, <math> <mrow> <msubsup> <mi>β</mi> <mi>j</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <mrow> <mo>{</mo> <msubsup> <mi>β</mi> <mn>1</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>β</mi> <mn>2</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>.</mo> <mo>.</mo> <mo>.</mo> <msubsup> <mi>β</mi> <mi>p</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>}</mo> </mrow> <mo>,</mo> </mrow></math> the evaluation result is obtained by the projection trace analysis model described above while reflecting the contribution of the information amount contained in the index itself. <math> <mrow> <msubsup> <mi>β</mi> <mi>j</mi> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <mrow> <mo>{</mo> <msubsup> <mi>β</mi> <mn>1</mn> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <msubsup> <mi>β</mi> <mn>2</mn> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <msubsup> <mi>β</mi> <mi>p</mi> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>}</mo> </mrow> <mo>,</mo> </mrow></math> The preference of each element or attribute of a decision maker or the importance degree of the attribute is reflected, and the parameter is adjusted manually.
IV) evaluation of inter-regional similarity. The method is used for determining the comprehensive similarity and the similarity grade of the mature region and the region to be executed. Suppose that region A contains phi indicators and region B contains phi indicators
The coincidence indexes of the areas A and B are sigma,
wherein the similarity between the jth similar indexes is represented by SjFrom step II). Degree of similarity determined by number of indices is represented by SnIs represented by a similarity SjDetermined similarity by SuIs represented by betajThe composite weight of the jth similarity element obtained in the step III) is expressed, and the composite weight not only considers the preference of a decision maker, but also considers the self information of the index. Similarity S between regions A and B*The mathematical model of (A, B) is represented as follows:
in step 5) of the method of the present invention, the mode selection is used to select an appropriate protective farming mode for the area to be performed when the similarity level is I, II level, based on the similarity level of the mature area for protective farming and the area to be performed. The correspondence between the similarity level and the similarity value is as follows (as shown in table 2):
table 2: correspondence between similarity level and similarity value
Through the mode matching of the region to be implemented and the mature region, a field experiment which is long-lasting in the open field is not needed, and the protective farming mode which is most suitable for the actual conditions of the region to be implemented is selected for the region to be implemented.
The protective farming mode selection model has the advantages of advanced analysis method and accurate result. Wherein the determination of the index is completed by a field survey and expert interview method; determination of objective weights by projection pursuit techniques
The indexes are treated differently in the process of (1), different methods are adopted for data standardization according to the characteristics of different indexes, and an accelerated genetic algorithm based on real numbers and MATLAB programming are adopted when an objective function is optimized. In the process of evaluating the system similarity, different methods are adopted to determine the similarity of each index according to the difference of different index properties. Composite weight betajDerived from objective weights
And subjective weight
Determined using the aforementioned projection tracking technique,and (4) synthesizing expert opinions by using an expert survey method.
The protective farming mode selection method can select proper protective farming modes for regions to be implemented with different climates, geographies and production habits.
Claims (2)
1. A protective farming mode selection method characterized by: it comprises the following steps:
1) data collection: collecting soil data, climate data, geographic data and production habit data of mature areas in different modes; collecting soil data, climate data, geographic data and production habit data of an area to be implemented;
2) mode selection preparation: determining decision units and indexes of each protective farming mode according to the data of the mature regions, wherein each type of the mature regions corresponds to one decision unit, and the indexes are derived from comparable soil data, climate data, geographic data and production habit indexes in the mature regions;
3) constructing a projection pursuit analysis model: standardizing the detection and survey data of the mature region, and determining the objective weight of each index by using a projection pursuit technology according to the decision unit and the indexWherein u is 1, …, n, p is the number of index, construct the projection pursuit analysis model including the following steps:
a) data normalization: the original data matrix X*Conversion into a standardized matrix X, XijRepresenting the jth index value of the ith decision unit for the element in the standardized matrix X;
b) linear projection: let unit vector a ═ a1，a2，…apIs the one-dimensional linear projection direction, the one-dimensional projection eigenvalue of the normalized matrix X projected onto the projection direction a is zi；
c) Constructing a projection objective function:
Q(a)＝SZ·DZ
wherein E (z) is the sequence { ziI ═ 1, …, n | } average; let rikIs the distance between the projected eigenvalues; n is the number of categories of mature regions; j ═ 1, 2, …, p; r is the density window width, and the value range is as follows:
rik＝|zi-zk|，i，k＝1，…，n； 
d) optimizing the projection objective function q (a): optimizing the projection direction a by using the projection objective function by utilizing the objective function maximization to ensure that the projection direction in which the projection objective function Q (a) reaches the maximum value is the optimal projection direction
MaxQ(a)＝Sz·Dz
e) Calculating objective weight: determining objective weights for the respective indicators
4) Constructing a similarity evaluation model: according to the data of each mature region, calculating index similarity composite weight betaj(ii) a Then calculating the similarity value between the region to be executed and each mature region, evaluating the similarity and the similarity condition between the region to be executed and each mature region, and constructing a similarity evaluation model, wherein the similarity evaluation model comprises the following steps:
i) classifying indexes;
ii) calculating the index similarity S (A, B) of the same index between the region A to be administered and the mature region B, comprising the steps of:
when the values of the two indexes are determined values, the step of calculating the similarity of the indexes is as follows: when indexes of the areas A and B are subjected to data standardization in the step a) in the step 3) to obtain determined values, the indexes L (A), L (B) belong to [0, 1 ]; the similarity between the determination values can be represented by the distance between two points, and the closer the distance, the higher the similarity, the following are:
secondly, when the index value of the area A is a determined value and the same index value of the area B is an interval value, the step of calculating the index similarity is as follows: l (A) e [ a ]1，a2]The index l (B) of the area B is a normalized interval, and l (B) ([ B ])1，b2]，a1＜a2，b1＜b2And a is a1，a2，b1，b2∈[0，1](ii) a Determining similarity between the values and the intervals, wherein the similarity is respectively related to the distance between the values and the middle points of the intervals and the length of the intervals, and the similarity is higher when the distance between the values and the middle points of the intervals is smaller and the length of the intervals is shorter;
and thirdly, if the values of the two indexes are both interval values, the calculation steps of the index similarity are as follows: the similarity between the two intervals is expressed by adopting the coverage degree of the intervals, and the larger the coverage degree is, the higher the similarity is;
fourthly, calculating the similarity of the binary variable indexes: the binary variable index shows that the value of the index is only 0 or 1, if the area A and the area B have a plurality of same binary variable indexes, the similarity calculation steps of all the same binary variable indexes are as follows: let h denote the number of indexes that both take the value of the index of the area a and the same index of the area B as 1, f denote the number of indexes that both take the value of the index of the area a as 1 and the value of the same index of the area B as 0, s denote the number of indexes that both take the value of the index of the area a as 0 and the value of the same index of the area B as 1, t denote the number of indexes that both take the value of the index of the area a and the same index of the area B as 0, and d (a, B) denote the dissimilarity of all the same binary variable indexes of the area a and the area B, i.e., their euclidean distances, then the similarity of all the same binary variables is:
fifthly; if the area A and the area B have a plurality of nominal variable indexes, the similarity calculation steps of all the nominal variable indexes are as follows: the nominal variable index is the popularization of a binary variable index, namely the number of the states of values in the index is more than two; assuming that g represents the number of indexes with the same value of the same nominal variable index in the area a and the area B, q represents the number of all nominal variable indexes in the area a and the area B, and d (a, B) represents the dissimilarity of the nominal variable indexes of the area a and the area B, the nominal variable similarity is:
sixthly, if the two indexes are qualitative indexes, the similarity calculation of the qualitative indexes comprises the following steps: the similarity of qualitative indexes is realized by mapping fuzzy qualitative indexes into numerical description, and comparing numerical values and utilizing a similarity formula of the two definite value indexes in the step I; (ii) a
iii) determination of index similarity composite weight:
wherein, the
Solving by the projection pursuit analysis model;
iv) evaluation of inter-regional similarity: region A contains phi indexes and region B contains
The index phi is less than or equal to p,
the coincidence of areas a and B indicates a,
wherein the degree of similarity S between the m-th similar indexesmIndex similarity S (a, B) equal to the mth index obtained in step ii), m being 1 … σ; beta is amCompounding a weight coefficient for the similarity of the mth index obtained in the ii i) step; similarity S between regions A and B*Mathematical model of (a, B):
5) mode selection: and selecting the protective farming mode corresponding to the mature region with the highest similarity with the region to be implemented as the protective farming mode of the region to be implemented according to the similarity grade.
2. A protective farming mode selection method according to claim 1, wherein: in the step 1), the soil data of the mature area is derived from the existing experimental results and field investigation results; the soil data of the region to be implemented is determined by a chemical method; the climate data of the mature region and the region to be executed are obtained according to the 10-year weather data weighted average; and the geographic data and the data of the production habits of the mature region and the region to be implemented are obtained by a field investigation method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102265088A CN101430735B (en) | 2008-11-13 | 2008-11-13 | Protective farming mode selection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2008102265088A CN101430735B (en) | 2008-11-13 | 2008-11-13 | Protective farming mode selection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101430735A CN101430735A (en) | 2009-05-13 |
| CN101430735B true CN101430735B (en) | 2011-09-21 |
Family
ID=40646128
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008102265088A Expired - Fee Related CN101430735B (en) | 2008-11-13 | 2008-11-13 | Protective farming mode selection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101430735B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103593534A (en) * | 2013-11-21 | 2014-02-19 | 中国能源建设集团广东省电力设计研究院 | Shield tunneling machine intelligent model selection method and device based on engineering geology factor relevance |
| US20150170067A1 (en) * | 2013-12-17 | 2015-06-18 | International Business Machines Corporation | Determining analysis recommendations based on data analysis context |
| CN104770275A (en) * | 2015-04-14 | 2015-07-15 | 天津大学 | Agricultural irrigation water optimal configuration method on basis of planting structure regulation |
| CN104778328B (en) * | 2015-04-23 | 2018-02-16 | 中国农业大学 | No-tillage seeding machine earth pack mode selecting method |
| CN105550938B (en) * | 2015-12-03 | 2020-08-11 | 中国农业大学 | Method for testing abnormal value of county-area cultivated land quality evaluation result |
| CN110489716A (en) * | 2019-05-20 | 2019-11-22 | 浙江省农业科学院 | Radix Rehmanniae cures evaluation method and system |
-
2008
- 2008-11-13 CN CN2008102265088A patent/CN101430735B/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN101430735A (en) | 2009-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jin et al. | Estimation of maize yield by assimilating biomass and canopy cover derived from hyperspectral data into the AquaCrop model | |
| Kucharik | Evaluation of a process-based agro-ecosystem model (Agro-IBIS) across the US Corn Belt: Simulations of the interannual variability in maize yield | |
| Legler et al. | Impact of ENSO-related climate anomalies on crop yields in the US | |
| CN113268923B (en) | Summer corn yield estimation method based on simulated multispectral | |
| Kitchen et al. | Development of a conservation-oriented precision agriculture system: Crop production assessment and plan implementation | |
| CN107256451A (en) | Agriculture suitability evaluation analysis method of the agriculture big data based on GIS | |
| CN101430735B (en) | Protective farming mode selection method | |
| Zhao et al. | Simple assessment of nitrogen nutrition index in summer maize by using chlorophyll meter readings | |
| CN107505271B (en) | Plant nitrogen estimation method and system based on nitrogen component radiation transmission model | |
| CN111044516B (en) | Remote sensing estimation method for chlorophyll content of rice | |
| CN114863289B (en) | Dynamic remote sensing monitoring method and system based on land utilization | |
| AU2016273991A1 (en) | Detection of environmental conditions | |
| CN114254964B (en) | Rice regional climate quality assessment method and system | |
| Beyene et al. | Coupling remote sensing and crop growth model to estimate national wheat yield in Ethiopia | |
| Ali et al. | Modeling wheat yield with data-intelligent algorithms: artificial neural network versus genetic programming and minimax probability machine regression | |
| CN117077868A (en) | Model-based grain yield prediction system and method | |
| Eeswaran et al. | Evaluating the applicability of soil moisture-based metrics for gauging the resiliency of rainfed agricultural systems in the midwestern United States | |
| CN113298859A (en) | Crop nitrogen fertilizer variable management method based on unmanned aerial vehicle image | |
| CN116681169A (en) | Method for evaluating influence of extreme climate on crop yield | |
| Cui et al. | Estimating maize yield in the black soil region of Northeast China using land surface data assimilation: integrating a crop model and remote sensing | |
| CN113567358B (en) | Cultivated land pollution monitoring and early warning system based on remote sensing and Internet of things | |
| Chu et al. | Mapping and forecasting of rice cropping systems in central China using multiple data sources and phenology-based time-series similarity measurement | |
| Jia et al. | Multi-stage fertilizer recommendation for spring maize at the field scale based on narrowband vegetation indices | |
| Fadl et al. | Assessing soil productivity potential in arid region using remote sensing vegetation indices | |
| CN115481366A (en) | Method for measuring and calculating farmland resource production potential based on space downscaling regression model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20110921 Termination date: 20131113 |