CN-119302101-B - Knowledge-graph-based water and fertilizer integrated intelligent irrigation method for fruit trees

CN119302101BCN 119302101 BCN119302101 BCN 119302101BCN-119302101-B

Abstract

The application relates to the technical field of intelligent agriculture, and provides a fruit tree water and fertilizer integrated intelligent irrigation method based on a knowledge graph, which comprises the steps of obtaining the knowledge graph according to the professional knowledge in the fruit tree water and fertilizer field; the method comprises the steps of establishing a strategy library containing each fertilization element and the professional knowledge of the fruit tree water and fertilizer field based on a knowledge graph and a fruit tree fertilization example, constructing a fertilization strategy similarity calculation model of the fruit tree water and fertilizer field based on the knowledge graph and the strategy library by integrating semantic structure similarity of the body of the fruit garden field, similarity of the fruit garden water and fertilizer irrigation attribute elements and similarity of the fertilization irrigation example, and matching an optimal irrigation strategy by combining the similarity calculation model with an inference mechanism when receiving the fertilization elements. The application can accurately and quickly match and call the fertilization strategy, and effectively avoid the unreasonable conditions of water shortage, fertilizer shortage and the like of the fruit trees.

Inventors

CUI DONGDONG
WANG XIAOFANG
ZHANG QIAN
WANG RANRAN
WANG DONG
KONG XIANGLU
TAO JIHAN
NIE PEIXIAN
LI YINGFANG
LIU WEIYUN
ZHOU FEI

Assignees

山东省果树研究所

Dates

Publication Date: 20260505
Application Date: 20241120

Claims (9)

1. A fruit tree water and fertilizer integrated intelligent irrigation method based on a knowledge graph is characterized by comprising the following steps: According to the water and fertilizer field profession of fruit trees knowledge is obtained to obtain a knowledge graph; based on the knowledge graph and the fruit tree fertilization example, establishing a strategy library containing fertilization elements and professional knowledge in the fruit tree water and fertilizer field; Based on the knowledge graph and the strategy library, the semantic structure similarity of the body of the orchard field, the similarity of the water and fertilizer irrigation attribute elements of the orchard and the similarity of the fertilization and irrigation examples are synthesized, and a fertilization strategy similarity calculation model of the water and fertilizer field of the orchard is constructed; When receiving the fertilization elements, matching an optimal irrigation strategy by combining a similarity calculation model with an inference mechanism; The strategy library comprising fertilization elements and the professional knowledge in the water and fertilizer field of the fruit tree is established based on the knowledge graph and the fruit tree fertilization example, specifically, Classifying and storing water and fertilizer structured, semi-structured and unstructured knowledge of an orchard, wherein the structured knowledge is stored by MySQL and Oracle relational data, the structured part of the semi-structured knowledge is stored by a relational database, the unstructured part of the semi-structured knowledge is stored by a hard disk and a server, the unstructured knowledge is stored by the hard disk, knowledge representation is completed by using ontology languages based on XML and OWL, the knowledge is expressed after being analyzed by an analysis tool in the form of OWL and XML, and the knowledge is expressed by: And (3) establishing a fruit tree fertilization strategy library by combining the professional knowledge of the fruit tree water and fertilizer field, the region of the fruit tree water and fertilizer instance, the fruit tree variety, the target yield and the fruit tree growth period attribute factors.
2. The method according to claim 1, wherein the knowledge graph is obtained according to the professional knowledge in the water and fertilizer field of the fruit tree, specifically: and constructing a knowledge graph in a bottom-up mode according to the professional knowledge in the water and fertilizer field of the fruit trees.
3. The method according to claim 1, wherein the knowledge graph is obtained according to the professional knowledge in the water and fertilizer field of the fruit tree, specifically: acquiring professional knowledge in the water and fertilizer field of the fruit trees, and marking parts of speech of the professional terms in the water and fertilizer field of the fruit trees; The hierarchical relationship and attribute relationship among the professional terms are analyzed in a summary way, knowledge attributes in the water and fertilizer field of the fruit trees are described, and attribute types are defined; And constructing classes and attributes of domain knowledge and combining the examples to obtain a knowledge graph.
4. The method according to claim 1, wherein the knowledge graph and strategy library are based on the similarity of semantic structures of the body of the orchard field, the similarity of the attribute elements of the water and fertilizer irrigation of the orchard and the similarity of the examples of the water and fertilizer irrigation of the orchard, and a similarity calculation model of the strategy of the water and fertilizer irrigation of the orchard is constructed, and the formula is as follows: ; In the above-mentioned method, the step of, The weight value corresponds to the irrigation attribute element, the semantic structure of the body and the fertilization irrigation example respectively; The method refers to the similarity of water and fertilizer irrigation attribute elements of an orchard, wherein the irrigation attribute comprises irrigation time, region where the irrigation attribute is positioned, fruit tree varieties and expected yield; meaning semantic structure similarity of the body in the orchard field; the similarity of the fertilization and irrigation examples in the strategy library is referred; 。
5. The method according to claim 1, wherein the matching of optimal irrigation strategies by a similarity calculation model in combination with an inference mechanism when receiving fertilizer elements comprises the following: according to the received fertilization elements, irrigation strategy screening is carried out, and the screened irrigation strategy is calculated through a similarity calculation model to obtain at least one fertilization strategy; The fertilization strategy is deduced according to preset reasoning to obtain a recommended fertilization strategy; and evaluating the recommended fertilization strategy according to a preset evaluation standard, and outputting the recommended fertilization strategy as an optimal irrigation strategy when the evaluation standard is met.
6. The method of claim 5, wherein the fertilizer elements are fruit tree period, expected yield and area, terrain.
7. The method according to claim 5, wherein the fertilization strategy is based on preset reasoning to obtain a recommended fertilization strategy by reasoning, specifically: And (3) combining the agenda of the Drools engine, and adopting a local variable mode to combine the weight of the element to conduct rule reasoning.
8. The method according to claim 7, wherein the agenda combined with the Drools engine adopts a local variable mode to combine weights of elements to perform rule reasoning, specifically: Adding a fertilization strategy into the agenda, and iteratively executing rules in the agenda until a recommended fertilization strategy is obtained; If there is a conflict between the matched rules, the conflicting rules need to be temporarily stored in the conflict set, and the rules are activated and added to the agenda after the conflict is resolved.
9. The method according to claim 5, wherein the recommended fertilization strategy is evaluated according to a preset evaluation criterion, and when the evaluation criterion is met, the recommended fertilization strategy is output as an optimal irrigation strategy, specifically: And according to a preset evaluation standard, evaluating whether the recommended fertilization strategy is reasonable or not, outputting the recommended fertilization strategy as an optimal irrigation strategy if the recommended fertilization strategy is reasonable, directly applying the strategy, and continuously adjusting the recommended fertilization strategy by combining the knowledge of the water and fertilizer field of the fruit trees if the recommended fertilization strategy is not reasonable until a reasonable water and fertilizer strategy is obtained.

Description

Knowledge-graph-based water and fertilizer integrated intelligent irrigation method for fruit trees Technical Field The application belongs to the technical field of intelligent agriculture, and particularly relates to a knowledge-graph-based fruit tree water and fertilizer integrated intelligent irrigation method. Background The water and fertilizer integrated technology can improve the water resource utilization rate, can effectively control the fertilizer concentration in an optimal range during operation, better promotes the root development of crops and the absorption of nutrient substances, and is widely applied in fields, orchards, greenhouses and the like. However, the practice finds that the existing water and fertilizer integration technology is difficult to accurately and quickly match and call a fertilizer application strategy, and causes unreasonable conditions such as water shortage and fertilizer shortage of fruit trees. Disclosure of Invention The embodiment of the application provides a knowledge-graph-based water and fertilizer integrated intelligent irrigation method for fruit trees, which aims to solve the technical problems. The first aspect of the embodiment of the application provides a knowledge-graph-based fruit tree water and fertilizer integrated intelligent irrigation method, which comprises the following steps: According to the water and fertilizer field profession of fruit trees knowledge is obtained to obtain a knowledge graph; based on the knowledge graph and the fruit tree fertilization example, establishing a strategy library containing fertilization elements and professional knowledge in the fruit tree water and fertilizer field; Based on the knowledge graph and the strategy library, the semantic structure similarity of the body of the orchard field, the similarity of the water and fertilizer irrigation attribute elements of the orchard and the similarity of the fertilization and irrigation examples are synthesized, and a fertilization strategy similarity calculation model of the water and fertilizer field of the orchard is constructed; when the fertilization element is received, the optimal irrigation strategy is matched through a similarity calculation model and an inference mechanism. Further, the knowledge graph is obtained according to the professional knowledge in the water and fertilizer field of the fruit trees, and specifically comprises the following steps: and constructing a knowledge graph in a bottom-up mode according to the professional knowledge in the water and fertilizer field of the fruit trees. Further, the knowledge graph is obtained according to the professional knowledge in the water and fertilizer field of the fruit trees, and specifically comprises the following steps: acquiring professional knowledge in the water and fertilizer field of the fruit trees, and marking parts of speech of the professional terms in the water and fertilizer field of the fruit trees; The hierarchical relationship and attribute relationship among the professional terms are analyzed in a summary way, knowledge attributes in the water and fertilizer field of the fruit trees are described, and attribute types are defined; And constructing classes and attributes of domain knowledge and combining the examples to obtain a knowledge graph. Further, the strategy library comprising each fertilization element and the professional knowledge in the water and fertilizer field of the fruit tree is established based on the knowledge graph and the fruit tree fertilization example, specifically, Classifying and storing water and fertilizer structured, semi-structured and unstructured knowledge of an orchard, wherein the structured knowledge is stored by MySQL and Oracle relational data, the structured part of the semi-structured knowledge is stored by a relational database, the unstructured part of the semi-structured knowledge is stored by a hard disk and a server, the unstructured knowledge is stored by the hard disk, knowledge representation is completed by using ontology languages based on XML and OWL, the knowledge is mainly expressed after the OWL and XML forms are analyzed by an analysis tool, and: And (3) establishing a fruit tree fertilization strategy library by combining the professional knowledge of the fruit tree water and fertilizer field, the region of the fruit tree water and fertilizer instance, the fruit tree variety, the target yield and the fruit tree growth period attribute factors. Further, based on the knowledge graph and the strategy library, the similarity of semantic structures of the body of the orchard field, the similarity of the water and fertilizer irrigation attribute elements of the orchard and the similarity of the fertilization irrigation examples are synthesized, and a fertilization strategy similarity calculation model of the water and fertilizer field of the orchard is constructed, wherein the formula is as follows: In the above-mentioned method, the st