Search

Assessing the sustainable development of agriculture is a hot topic in global agricultural research and an important task of rural revitalization and agricultural upgrading in the new era. Reform and opening up promote the rapid development of the agricultural economy, but the conflict between agricultural development and agricultural resources and ecological environment is serious. Since the 18th National Congress of the CPC, the sustainable development of agriculture has been given more prominence. It is of great significance to study the evaluation of agricultural sustainable development. Firstly, this paper studies the relevant literature on agricultural sustainable development evaluation at home and abroad. Based on wavelet transform, support vector machine(SVM), genetic algorithm (GA)and ant colony optimization(ACO) algorithm, a combination evaluation method of wavelet and intelligent optimization support vector machine is proposed. Secondly, by referring to the existing evaluation index system and combining with the actual situation of China's agricultural development, the alternative index set of agricultural sustainable development evaluation is established. Using SPSS software, the significance and correlation of 150 primary indexes are analyzed, and an evaluation index system including 62 indexes is established. Thirdly, using one-dimensional discrete stationary wavelet analysis, data denoising processing, genetic algorithm, ant colony algorithm, optimize the parameters of support vector machine, get better penalty parameters, kernel function, insensitivity coefficient, and then train the support vector machine. Fourth, taking 31 provinces (municipalities) in China as an example, this paper evaluates the agricultural sustainable development by applying the wavelet-GASVM model and the wavelet-ACOSVM model, and using the GA-SVM model and the ACO-SVM model to evaluate the agricultural sustainable development of 31 provinces (municipalities) in China. Fifthly, the reliability of the wavelet intelligent optimization SVM method is verified by changing the number of test sets and conducting multiple tests. The evaluation and simulation results show that: 1) The mean square error of agricultural sustainable development evaluation of Wavelte-GASVM is 9.641×10-5, and the correlation coefficient is 0.980; however, under the same training set and test set, the mean square error and correlation coefficient obtained by GASVM are 0.006 and 0.979, respectively. 2)The mean square error of Wavelet-ACOSVM agricultural sustainable development evaluation was 9.318×10-5, and the correlation coefficient was 0.972; the mean square error MSE and correlation coefficient of ACO-SVM are 0.016 and 0.953, respectively, under the same training set and test set. 3) Change the number of test sets and perform several tests and compare the results of GASVM model and ACO-SVM model before and after wavelet denoising respectively. It can be seen that under the GASVM model, the mean square error of the running results after wavelet denoising is smaller than that before wavelet denoising, and the correlation coefficient is larger than that before wavelet denoising, and the running time is shorter. Under the ACOSVM model, the mean square error of the running result after wavelet denoising is also smaller than that before wavelet denoising, and the correlation coefficient is larger than that before wavelet denoising, and the running time is shorter. Conclusion: The prediction accuracy and convergence speed of agricultural sustainable development evaluation by wavelet-GASVM and wavelet-ACOSVM are better than GA-SVM and ACO-SVM, respectively, and the effect of agricultural sustainable development evaluation by wavelet-GASVM is more ideal. This study provides a method reference for agricultural sustainable development assessment. [ABSTRACT FROM AUTHOR]

Modern mechanical equipment is very important to agricultural production. The introduction of modern machinery and equipment into the production process of traditional Chinese medicine is an effective way to solve the contradiction between the expansion of production scale of traditional Chinese medicine industry and the reduction of labor force. The mechanization development of traditional Chinese medicinal materials production is in the initial stage. The main goal at this stage is to realize the "from scratch" of the relevant technical equipments. However, due to the varieties of traditional Chinese medicine, complex production process, regional differences and other factors, it is difficult to judge the demand for mechanized production of traditional Chinese medicine and lack of methods. How to evaluate and analyze the the differences in the demand for mechanized development is of great significance to the research of technical equipment, the implementation of supply side reform and the promotion of the production and development of traditional Chinese medicine. According to the characteristics of traditional Chinese medicinal materials production, combined with the analysis of degree of mechanized operation, status of technology and equipment and levels of user demand, this paper constructs the evaluation index system for the development of technical and equipment of traditional Chinese medicinal materials production. The degree of mechanized operation indicates the ratio of mechanized operation area to the total planting area. The status of technology and equipment is alculated by weighting the five indexes of advance ment, economy, applicability, reliability and security. The levels of user demand is divided according to the proportion of users with demand, users with demand but no machinery available and the proportion of users with demand in different production process. Then an evaluation model combined with entropy weight method and hierarchical cluster analysis method is formed, and after classifying by the hierarchical cluster analysis method, the model is expressed functionally based on the Fisher discriminant method. In empirical research, the demonstration base of national comprehensive experimental station of traditional Chinese medicinal materials is selected as the research object, eight production processes of sowing, transplanting, fertilization, intertillage, plant protection, harvesting, cleaning and drying are selected for evaluation Through the investigation in 23 provinces of China, the production processes of traditional Chinese medicinal materials production are categorized into three types. The harvest, intertillage, and transplanting belonged to the type A, which have the highest demand level. The seeding is classified as type B, and the drying, cleaning, plants protection and fertilization are type C. The phenomenons of “no machinery available” and “unsuitable to use” are obvious for type A, and there is an urgent demand for the development of technology and equipment, the evaluation results are consistent with the investigation results. According to the further calculation results based on discriminant Function I, the central values of type A, type B and type C are -2.868, 1.904 and 1.675 respectively, which shows that the evaluation model of development demands classification can be accurately expressed by quantitative mathematical functions. The machinery of the type A production processes represented by harvest should be developed in priority in traditional Chinese medicinal materials production. The research and development of special machinery for the production of traditional Chinese medicinal materials should take the road of combining independent innovation research with the introduction of other main grain production machinery. The model proposed in this paper reflects the reasons and details at the same time of forming classification, which can provide reference for the development demand evaluation of other crops or regional agricultural machinery technology and equipment with backward mechanization development. [ABSTRACT FROM AUTHOR]