CN-122022327-A - Agricultural socialization service resource optimal allocation method and system based on supply and demand intelligent matching

CN122022327ACN 122022327 ACN122022327 ACN 122022327ACN-122022327-A

Abstract

The invention relates to the technical field of agricultural social service, in particular to an agricultural social service resource optimal allocation method and system based on supply and demand intelligent matching, wherein the method collects full-dimension multi-source data of stored grains through a starry sky network integrated sensing network, establishes a simplified matching index CRI and an adaptation degree evaluation index Sij cooperation mechanism based on an intelligent algorithm after standardized processing and blockchain encryption storage to form a dynamic matching logic with primary screening grade and accurate optimization, and links storage, preprocessing equipment, agricultural materials, transportation agricultural machinery and agricultural technology resources to realize full-chain cooperation scheduling; the invention breaks through the traditional rough matching mode by combining the Internet of things, beidou positioning and RFID technology to construct a full-flow monitoring traceability system and performing closed-loop evaluation iteration optimization on model parameters, obviously improves the utilization rate of stored grain resources and the safety of stored grains, reduces postpartum loss, provides technical support for large-scale green stored grains, and has practicability and popularization value.

Inventors

YANG YONGBIN
Xun Yongtao
SUN DASHUANG
ZHOU YANMIN
KONG LING
LI SONG
LI CHUNYU
Hu Shuwang
DOU YANLI
ZHANG YUGUO

Assignees

金乡县乡村振兴事务中心(金乡县农村经济管理服务中心)

Dates

Publication Date: 20260512
Application Date: 20260129

Claims (8)

1. The agricultural socialization service resource optimization configuration method based on intelligent supply and demand matching is characterized by comprising the following steps: S1, grain storage multi-source data acquisition and standardization processing, namely acquiring grain storage demand end, resource end, environment end and auxiliary data through a starry sky and ground network integrated sensing network and a multi-terminal access module, and storing the acquired data in a distributed data center after standardized conversion and cleaning; S2, extracting supply and demand characteristics and sequencing demand priorities, extracting supply and demand core characteristics to construct characteristic vectors, and determining the demand priorities based on a hierarchical analysis method in combination with grain safety requirements and preprocessing urgency; S3, dynamically and intelligently matching stored grains and optimizing and scheduling resources, wherein an intelligent algorithm is adopted, and a logic generation from primary screening classification to accurate optimization is realized through a matching index CRI and an adaptation degree evaluation index Sij, and a matching scheme is dynamically adjusted; The optimization scheduling path is characterized in that grain moisture content, impurity content, pretreatment requirement and time limit, warehouse empty period and turnover requirement are defined as core conditions, quantitative calculation is needed, various variable value ranges and calculation rules are defined, an adaptation degree calculation formula and CRI classification mechanism are constructed, multiple types of grain storage resources are accurately adapted through exponential sequencing, removed resources are defined through rigid screening rules, and finally an optimal scheme is locked by combining the conditions that storage residual capacity is more than or equal to total grain storage amount multiplied by 1.1, and agricultural resource adaptation degree is more than or equal to 0.8, and total consumption time is less than or equal to pretreatment time limit Ti; S4, dynamically monitoring and tracing data in the grain storage service process, and acquiring whole process data through the Internet of things, beidou positioning and RFID technology to generate a tracing file binding with a unique ID; And S5, performing closed-loop evaluation and model optimization, quantifying a service effect based on a multidimensional evaluation system, and feeding back and optimizing parameters of a matched model to form a closed-loop mechanism.
2. The optimal allocation method for agricultural social service resources based on intelligent supply and demand matching according to claim 1, wherein in the step S1, a starry sky and ground network integrated sensing network comprises a satellite remote sensing unit, an unmanned aerial vehicle field inspection unit and an internet of things sensing unit, and a multi-terminal access module comprises a farmer APP, a grain storage service main body terminal and a government platform interface, wherein the internet of things sensing unit is a warehouse temperature and humidity sensor.
3. The optimization configuration method for agricultural social service resources based on intelligent supply and demand matching according to claim 1, wherein in S3, the specific implementation steps of the primary screening classification-to-accurate optimization collaborative matching logic are as follows: 1) CRI preliminary screening, namely calculating CRI indexes and eliminating invalid resources with CRI less than 0.7, and locking corresponding types of grain storage resources according to CRI classification to reduce the searching range, wherein the step is that large-scale orders are selected; 2) Rigid screening, namely eliminating resources with Sωij <0.5, Sζij <0.5 and STij <0.6 in the locked resource category, and reserving a candidate set meeting the core requirement; 3) The method comprises the steps of accurately optimizing, incorporating Sij into algorithm fitness value calculation, iteratively screening the first 3 candidate schemes, correcting scheduling time sequence by combining the conditions that the storage residual capacity is more than or equal to the total stored grain amount multiplied by 1.1 and the agricultural resource fitness is more than or equal to 0.8, ensuring that Tzij + Trij is less than or equal to Ti, and locking the optimal scheme; 4) Dynamic switching, the medium-small-scale order can skip the CRI preliminary screening step, the accurate matching is directly completed through Sij, and the preliminary screening and optimization two-step method is strictly executed for the super-large-scale order.
4. The method for optimizing and configuring agricultural social service resources based on intelligent matching of supply and demand according to claim 1, wherein the intelligent algorithm in step S3 is as follows: Matching index CRI: CRI=0.3×(2-X)+0.25×(2-Y)+0.2×Z+0.15×W+0.1×V; wherein, X is the water content exceeding coefficient, Y is the impurity content exceeding coefficient, Z is the pretreatment aging coefficient, W is the warehouse adapting coefficient and V is the turnover demand coefficient in the formula, each variable definition and calculation rule is defined according to the technical scheme, and the weight is fixed and can not be modified; the quick preliminary screening of the resource category is realized through 5 variable input/output CRI indexes, and the method is specifically as follows: The water content exceeding coefficient X is X=ωi/ω s , wherein ωi is the actual water content of the grain, the unit is%and the value is 5% -30%, ω s is the critical value of the water content of the corresponding product type safety stored grain, the unit is%and the value is 8% -15%, the wheat 13%, the corn 14% and the rice 12% are preset values, when X is less than or equal to 1, 1 is taken, the state is no exceeding standard, and when X is more than 1, the actual value is calculated; The impurity content exceeding coefficient Y is Y=ζi/ζ s , wherein ζi is the actual impurity content unit of grain, the value is 0.1% -10%, ζ s is the impurity content critical value of corresponding product safety stored grain, the unit is 1% -3%, the preset value of wheat 2%, corn 2.5% and rice 1.5%, when Y is less than or equal to 1, 1 is no exceeding state, when Y is more than 1, the actual value is calculated; the pretreatment aging coefficient Z is Z=Ti/(Tzij +2), wherein Ti is the time limit for finishing the pretreatment requirement, the unit is h, tzij is the estimated pretreatment time length, the unit is h, and 2h is added for buffering, when Z is more than or equal to 1, 1 is taken, the aging is sufficient, and when Z is less than 1, the actual value is calculated; the warehouse adapting coefficient W is W=1-delta Tij/72, wherein delta Tij is the deviation between the planned warehouse-in time and the warehouse space initial time, the unit is h,72 is the maximum allowable deviation threshold value, when W is more than or equal to 0.6, according to the actual value, when W is less than 0.6, 0.5 is taken, and the time sequence adaptation is insufficient; The turnover demand coefficient V is set based on a turnover coefficient Ri, ri=0, V=1 and Ri=1 when no turnover exists, V=0.8 when the turnover is single, ri is more than or equal to 2, V=0.6 when the turnover is repeated, and the more complex the turnover is, the lower the coefficient is; Wherein, (2-X) (2-Y) reversely constrains the quality index, and the more serious the exceeding standard, the lower the score is; all variables are partitioned into 0.5-1, the CRI value range is 0.7-1.8, and the higher the index is, the stronger the adaptation degree is; CRI grading matching rules are that CRI value ranges are 0.7-1.8, the grading rules are fixed to CRI is larger than or equal to 1.5 and is a primary resource, CRI is larger than or equal to 1.2 and is smaller than or equal to 1.5 and is a secondary resource, CRI is larger than or equal to 0.9 and is smaller than or equal to 1.2 and is a tertiary resource, CRI is smaller than or equal to 0.9 and is a quaternary resource, a matching process is fixed to CRI primary screening, and then according to grading locking resource types, the weight ratio of multiple objective functions is fixed to be 40% of matching precision, 30% of resource utilization, 20% of service cost and 10% of time consumption.
5. The agricultural social service resource optimizing configuration method based on intelligent supply and demand matching according to claim 3, wherein the CRI hierarchical matching rule divides 4 grades according to indexes, and the rapid classification matching is realized corresponding to differentiated resource combinations: High-end resource combination, configuration of intelligent temperature and humidity control warehouse, drying and cleaning integrated equipment, special agricultural materials, special agricultural technology whole course guidance, and priority of nearby warehouse and high-speed transportation; Secondary matching (CRI < 1.2 is less than or equal to 1.5), namely, medium-end resource combination, conventional temperature and humidity regulation warehouse and independent pretreatment equipment are configured, standard agricultural materials are used, agricultural technology inspection is timed, and agricultural machinery is reasonably allocated and transported; Three-level matching (CRI < 1.2) is carried out, wherein, the basic resource combination is provided with a common ventilation warehouse, simple pretreatment equipment, general agricultural materials and necessary agricultural technology consultation, and the non-turnover requirement is preferentially met; and four-stage matching (CRI < 0.9), namely, emergency resource combination, configuration of emergency pretreatment equipment and temporary storage, management and control by special personnel, priority treatment of core quality problems and reduction of loss.
6. The optimization configuration method for agricultural social service resources based on intelligent supply and demand matching according to claim 1, wherein in S3, the fitness evaluation index Sij is used for precise optimization after CRI preliminary screening, and the detail of the adaptation is quantified by a precise term coefficient, which is specifically as follows: setting a grain storage order as i and a storage resource as j; Δωi is grain moisture content deviation, the unit is%, the calculation formula is Δωi= |ωi- ω s |, wherein ωi is actual moisture content of the grain of the order i, and ω s is a critical value of the moisture content of the safe stored grain of the grain corresponding to the order i; Delta zeta i is grain impurity content deviation, the unit is%, the calculation formula is delta zeta i= |zeta i-zeta s |, wherein zeta i is actual impurity content of the grain of the order i, and zeta s is a safety grain storage impurity content critical value of the grain corresponding to the order i; Pi is the pretreatment demand coefficient of the order i, the coefficient is 0, no pretreatment demand exists, the coefficient is 1, only the cleaning demand is selected, the coefficient is 2, only the drying demand is selected, and the coefficient is 3, and the cleaning and drying compound demand is selected; Qi is total grain storage amount of the order i, wherein the unit is ton, tzij is estimated pretreatment total time of the order i at the warehouse resource j, the unit is h, the calculation formula is Tzij =qi×tj, wherein Tj is single unit treatment time of pretreatment equipment at the warehouse resource j, and the unit is h/ton; Trij is the turnover time consumption of the grain round-trip warehouse resource j of the order i, and the unit is h; The calculation formula is Trij = (2×lij×ri)/Vi; The system comprises a storage resource j, an order i grain production place, a storage resource j, a factor of 1, a factor of 2, a factor of more than or equal to Vi, a factor of 0, no turnover, a factor of 1, a single turnover, a plurality of turnover, and a factor of Vi, wherein the unit is km, and the unit is km/h; R max is the maximum allowable turnover number set by the system, and the default is 3 times, so that the system can be finely adjusted according to the actual scene; sij is the adaptation degree evaluation index of the order i and the warehouse resource j, the value range is 0-1, the adaptation degree is stronger when the index is higher, wherein: fitness evaluation index Sij: Sij=α×Sωij+β×Sζij+γ×STij+δ×SRij; Wherein, alpha, beta, gamma and delta are the weight coefficients of four adaptation coefficients, which satisfies alpha+beta+gamma+delta=1, the value range alpha epsilon [0.3,0.4], beta epsilon [0.2,0.3], gamma epsilon [0.2,0.3], delta epsilon [0.1,0.2], the default values are respectively 0.35, 0.25 and 0.15, and the values respectively correspond to the following four adaptation coefficients: the water content adaptation coefficient S omega ij, the adaptation score of the order i and the warehouse resource j on the water content processing capacity is 0.3-1, and the calculation rule is as follows: When ωi is less than or equal to ω s , sωij=1, and full-scale adaptation is performed; When ωi > ω s , sωij=1- Δωi/ωi, and is defined forcedly; S omega ij is more than or equal to 0.3, and the invalidation of extreme values is avoided; the impurity content adaptation coefficient S ζij, the adaptation score of order i and warehouse resource j on impurity handling capacity, value range 0.3-1, calculation rule is: When ζi is less than or equal to ζ s , S ζij=1, and fully distributing and adapting; ζi > ζ s , sζij=1- Δζi/ζi, and is defined by default; S ζij is more than or equal to 0.3, so that invalidation of extreme values is avoided; The time sequence adaptation coefficient STij is the adaptation score of the order i and the warehouse resource j on the operation time sequence, the value range is 0.4-1, and the calculation formula is as follows: STij=1-(Tzij+ΔTij+Trij)/(Ti+24); The delta Tij is the deviation between the planned warehousing time of the order i and the vacant starting time of the warehousing resource j, the unit is h, the Ti is the finishing time limit of the preprocessing requirement of the order i, the unit is h, the STij E [0.4,1] is forcedly limited, and STij is more than or equal to 0.6 and is a time sequence qualification standard; The turnover adaptation coefficient SRij is the adaptation score of the order i and the warehouse resource j on the turnover capability, the value range is 0.5-1, and the calculation rule is as follows: Without turnover, ri=0, SRij =1; single turnover, SRij e [0.7,0.9] when ri=1; And when Ri is more than or equal to 2, SRij E [0.5,0.7] and the more complex the turnover is, the lower the adaptation score is.
7. The agricultural socialization service resource optimal allocation system based on the supply and demand intelligent matching is characterized by comprising a multi-source data acquisition module, a data processing and storage module, a grain storage supply and demand analysis and matching module, a grain storage full-chain collaborative scheduling module, a grain storage process monitoring and tracing module, a closed-loop optimization module and a user interaction module; The multi-source data acquisition module comprises a satellite remote sensing unit, an unmanned aerial vehicle field inspection unit, an Internet of things sensing unit and a multi-terminal access unit and is used for acquiring and uploading grain storage full-dimension data in real time; The data processing and storing module comprises a data standardization unit, a cleaning unit and a distributed data center, unifies data formats, eliminates redundant abnormal values, adopts block chain encryption storage and is used for data security and cross-node sharing; the grain storage supply and demand analysis and matching module comprises a feature extraction unit, a demand ordering unit and an intelligent matching unit, and is used for extracting supply and demand core features, determining demand priority, running an intelligent algorithm, loading a CRI and fitness model and generating a dynamic matching scheme; the grain storage full-chain cooperative scheduling module is used for realizing linkage scheduling of storage, pretreatment equipment, agricultural materials, transportation agricultural machinery and agricultural technical personnel, optimizing an operation path and supporting cross-region allocation; The grain storage process monitoring and tracing module comprises an integrated Beidou positioning unit, an RFID (radio frequency identification) identification unit and a video monitoring unit, and is used for collecting process data in real time, generating a tracing file and realizing full-flow visualization and traceability; the closed-loop optimization module comprises an evaluation and model optimization unit and is used for constructing a multi-dimensional evaluation system and iteratively optimizing model parameters based on an evaluation result; The user interaction module comprises an APP (application) at a farmer client, a terminal of a service main body and a platform at a supervision client, and the functions of demand release, order processing and full-flow supervision are respectively supported.
8. The agricultural socialization service resource optimization configuration system based on intelligent supply and demand matching according to claim 7 is characterized in that the data processing and storage module adopts a block chain encryption technology to construct a distributed data center to realize safe storage and cross-node sharing of stored grain data, and the stored grain process monitoring and tracing module integrates a Beidou positioning unit and an RFID radio frequency unit to realize full-flow visualization and traceability of stored grain services.

Description

Agricultural socialization service resource optimal allocation method and system based on supply and demand intelligent matching Technical Field The invention relates to the technical field of agricultural social services, in particular to an agricultural social service resource optimization configuration method and system based on supply and demand intelligent matching. Background Agricultural socialization service is a core support for modern agricultural development, and covers full-chain service scenes such as grain storage, cultivation, plant protection, processing, transportation and the like, and service resources of the agricultural socialization service comprise scientific allocation of storage facilities, pretreatment equipment, grain storage agricultural materials, transportation agricultural machinery, agricultural personnel and matched service facilities; With popularization of intelligent agricultural technology, the information barrier of traditional offline agricultural socialization service is broken gradually, an online supply and demand matching platform is created, the problems of unsmooth service resource circulation and low supply and demand butt joint efficiency are relieved to a certain extent, but a mature supply and demand intelligent matching system is not formed yet, and particularly in the aspect of multi-scene service resource overall configuration, the efficiency and the accuracy still have larger promotion space, and the development requirements of large-scale and diversified agricultural socialization service are difficult to adapt; however, the existing agricultural socialization service resource allocation technology still has a plurality of bottlenecks, especially in the aspects of intellectualization, precision and full-scene adaptation of supply and demand matching, and is difficult to support the actual demand of the agricultural socialization service resource optimization allocation based on the supply and demand intelligent matching, and the requirements of large-scale, fine and full-chain agricultural service scenes on efficient resource scheduling cannot be met; The existing matching resource scheme is lack of targeted support of a starry sky network integrated perception network, scattered data and different formats of related links, forms a data island, cannot support cross-link accurate matching, has rough matching logic, does not take the grain storage requirement as a core, does not quantify core conditions such as water content, preprocessing aging and the like, lacks a cooperative mechanism, easily causes resource mismatch and time sequence conflict, aggravates postpartum loss and cost waste, has poor resource scheduling cooperative performance, is limited to a single link, has no flexible switching strategy, and is difficult to consider order requirements of different scales, and based on the method and the system, the invention discloses an agricultural socialization service resource optimization configuration method and system based on intelligent matching of supply and demand. Disclosure of Invention In order to solve the problems that the existing scheme provided in the background technology is different in data dispersion and format in the grain storage and related links, forms a data island, cannot support accurate cross-link matching, has extensive matching logic, does not take the grain storage requirement as a core, does not quantify core conditions such as moisture content, pretreatment aging and the like, lacks a cooperative mechanism, easily causes resource mismatch and time sequence conflict, and aggravates postpartum loss and cost waste. The invention provides an agricultural socialization service resource optimization configuration method based on intelligent supply and demand matching, which comprises the following steps: S1, grain storage multi-source data acquisition and standardization processing, namely acquiring grain storage demand end, resource end, environment end and auxiliary data through a starry sky and ground network integrated sensing network and a multi-terminal access module, and storing the acquired data in a distributed data center after standardized conversion and cleaning; S2, extracting supply and demand characteristics and sequencing demand priorities, extracting supply and demand core characteristics to construct characteristic vectors, and determining the demand priorities based on a hierarchical analysis method in combination with grain safety requirements and preprocessing urgency; S3, dynamically and intelligently matching stored grains and optimizing and scheduling resources, wherein an intelligent algorithm is adopted, and a logic generation from primary screening classification to accurate optimization is realized through a matching index CRI and an adaptation degree evaluation index Sij, and a matching scheme is dynamically adjusted; The optimization scheduling path is characterized in that grain moisture conte