Study On Key Technologies Of Soa Based Digital System For Agricultural Affairs

Agriculture is characterized by the interweaving of natural reproduction and economic reproduction. In market economy, farmers always try every means to achieve optimal economic benefits. However, the existing digital farming system, which focused on digital expression and integration of farming information stream, production supervision and precision farming, paid less attention to the aim of optimal economic benefits, commodity circulation during pre-production and post-production, or consultation services. In order to provide farmers with decision-making assistance in the whole process of agricultural production and management, including pre-production, production and post-production, the concept of Digital System for Agricultural Affairs (DSAA) was introduced in this paper. Key techniques of integration of crop models and economic benefit models, combination of web services, management and reusage of agricultural model components were studied on the basis of Service-Oriented Architecture (SOA). Finally, a SOA-based Digital System for Agricultural Affairs (DSAA-SOA) was designed and implemented by integrating agricultural models and WebGIS.Since existing agricultural databases, consultation and information service systems, agricultural models and decision-making support systems developed with different programming languages and on different platforms, were distributed among a lot of administrational or decision-making organizations, it was very difficult for these agricultural information resources to share or integrate. Web service technology is an efficient implementation approach to constructing comprehensively integrated system. By applying service-oriented architecture to the decision-making support domain of agricultural production, management and evaluation, taking the economic benefit and optimization of management practices as objectives, and by integrating the methods of domain decomposition and legacy system analysis, web services were extracted from the process of "model-driven data, crop models, benefit model, spatial information analysis", on field-scale and region-scale, respectively. The framework of web services based on the integration of agricultural models and WebGIS was constructed, which laid a foundation for developing distributed and comprehensive digital system for agricultural affairs.For improving the capability of model resource sharing and management for different agricultural disciplines, the study proposed a new approach to description and representation of agricultural models, which used the description information of model components as meta-data to generate the topic maps at different levels. Firstly, the study analyzed the characteristics of agricultural models, established the mapping relationships between agricultural models and model components, and then extracted information on the facets and relationship descriptions of agricultural model components and constructed the Description Model for Agricultural Model Components (DM-AMC). Further, the DM-AMC was represented as topic type, association type, occurrence type and association role. Finally, the Topic Maps for Agricultural Model Components (TM-AMC) was constructed by taking example and using matching and merging algorithms. With crop phasic development model as example, the study generated the XML documents of atomic topic maps with the characters of unified description and dynamic expansion. The results indicated that a unified description on agricultural model information and generation of multi-level and easy-accessed TM-AMC based on DM-AMC could enhance capability of general sharing and management with agricultural model information, which was helpful in taking full advantage of agricultural model components distributed among different organizations to make candidate web services.Using data web services to access meteorological data and soil data at multi-location was a good solution to settling data source needed in the implementation of regional application of crop models on field-scale, but in the mean time a remarkable increase in the times of data remote call and model computing resulted in a problem of long responding time of crop model. In order to improve the responding time of crop model web services, the concurrent processing scenario of crop model web services based on caching mechanism was proposed in this paper. By comprehensively analyzing the combination structure of crop model web services, communication features of data web services and responding time feature of crop model web services, the concurrent processing scenario of crop model web services was designed based on multi-thread processing and data caching technology. With web services of crop management knowledge model as the testing target, the effectiveness of concurrent processing scenario was verified in a real network environment. The result indicated that the responding time of crop model web service with concurrent processing scenario reduced32%and35%for single-user at multi-location and multi-user at multi-location, respectively, which improved the model ability for multi-user concurrent processing. These results could provide the technical support for regional application of crop model.By integrating crop management knowledge models, economic benefit models, and WebGIS spatial analysis functions on the framework of web service combination, a DSAA-SOA was developed. Web service development approach of contract first was adopted to envelope web services of agricultural models. With ArcGIS Service and Model Builder as development platform, the approach to the development of dynamic thematic map web service and spatial interpolation web service was presented. DSAA-SOA realized comprehensive service functions such as system management, data management, management decision-making, economic benefit calculation, information consultation, digital map, covering the process of pre-production, production and post-production. The case study of DSAA-SOA at Tuqiao town indicated that it was feasible to use SOA technology to integrate agricultural data, models and WebGIS, and enchance the capability of agricultural model sharing in the region-scale application.