CN-122022687-A - Reagent storage linkage warehouse entry system based on improved transducer large model

CN122022687ACN 122022687 ACN122022687 ACN 122022687ACN-122022687-A

Abstract

The invention discloses a reagent storage linkage warehouse-in system based on an improved transducer large model, which comprises the steps of collecting reagent attribute, storing real-time operation and use demand data, preprocessing and constructing a characteristic set, generating a warehouse-in priority and an adaptive scheme, controlling an execution assembly to automatically warehouse in, synchronizing data and pre-judging demand, adjusting environmental parameters and checking data in two directions. The system solves the problem of prejudgment and dislocation of the existing warehouse-in and warehouse-in requirements, realizes the linkage of warehouse-in and warehouse-in, ensures the accurate inventory, and avoids the shortage of goods or improper storage.

Inventors

Zi Chenrui
ZHOU JINGLIANG
LIU XIAOMING
ZHANG FEIFEI
ZHANG TIAN

Assignees

碳金智瑞(天津)科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (10)

1. Reagent warehouse linkage warehouse entry system based on improvement transform big model, characterized by that includes following steps: s1, collecting reagent attribute data, real-time operation data of a warehouse system and reagent use demand data; s2, carrying out standardized pretreatment on the acquired reagent attribute data, the real-time operation data of the warehousing system and the reagent use requirement data to construct a multi-dimensional comprehensive feature set; s3, extracting key features by improving a transducer large model based on the multi-dimensional comprehensive feature set, and generating a reagent warehouse-in priority list and an adaptive storage scheme; S4, controlling a warehousing execution assembly to complete automatic transportation and warehousing positioning of the reagent based on the warehousing priority list and the adaptive storage scheme; s5, synchronizing the warehousing progress data and the real-time operation data of the warehousing system in real time, and pre-judging the reagent requirements in a future preset time period of the warehousing system according to the synchronization result to generate a pre-judging warehousing plan; S6, monitoring environmental parameters of a target storage area in the warehousing system in real time, and if the environmental parameters do not meet the requirements of an adaptive storage scheme, adjusting the environmental parameters by an environmental control system of the linked warehousing system; and S7, after the warehousing operation is finished, bidirectional checking is carried out on the warehousing data and the inventory data of the warehousing system, so that the data are ensured to be completely consistent.
2. The reagent warehouse-in linkage warehouse-in system based on the improved Transformer large model according to claim 1, wherein the reagent attribute data in S1 includes chemical stability, temperature and humidity sensitivity, remaining duration of validity period, unsealing status, reagent type, physical status of the reagent.
3. The reagent warehouse linkage warehouse entry system based on the improved transducer large model according to claim 2, wherein the real-time operation data of the warehouse system in S1 comprises the real-time inventory number of the warehouse system, the current temperature and humidity value of the storage area, the air flow disturbance influence parameter of the storage area, the occupancy rate of the storage area and the operation state of the warehouse equipment.
4. The reagent warehouse linkage warehouse entry system based on the improved Transformer large model of claim 3, wherein the reagent usage demand data in S1 includes usage urgency, demand frequency, single usage amount, usage department priority, turnover frequency weight of the reagent.
5. The reagent warehouse-in linkage system based on the improved transducer large model of claim 4, further comprising calculating a storage risk coefficient of the reagent by a storage risk coefficient formula, wherein the storage risk coefficient formula is constructed based on chemical stability, temperature and humidity sensitivity, unsealing state, physical state, remaining duration of the validity period and total duration of the validity period of the reagent.
6. The reagent warehouse-in linkage warehouse-in system based on the improved Transformer large model of claim 5, further comprising calculating a warehouse-in priority comprehensive score of the reagent by a warehouse-in priority comprehensive score formula, wherein the warehouse-in priority comprehensive score formula is constructed based on reagent storage risk coefficients, inventory gap coefficients, utilization urgency, turnover frequency weights, demand frequency, utilization department priority, occupancy of a storage area and airflow disturbance influencing parameters.
7. The reagent warehouse linkage warehouse entry system based on the improved Transformer large model of claim 6, further comprising calculating a temperature and humidity adjustment quantity of the target storage area through a warehouse environment adaptation adjustment quantity formula, wherein the warehouse environment adaptation adjustment quantity formula comprises a temperature adjustment quantity formula and a humidity adjustment quantity formula, which are constructed based on a warehouse entry priority comprehensive score, a target temperature and humidity reference value required by an adaptation storage scheme, a temperature and humidity current value of the storage area, an air flow disturbance influence parameter of the storage area, an occupancy rate of the storage area and a reagent storage risk coefficient.
8. The reagent warehouse-in linkage system based on the improved transducer large model of claim 1, wherein the improved transducer large model in S3 comprises a characteristic attention layer, a space-time fusion layer and an output layer, wherein the characteristic attention layer is used for distributing attention weights to all dimensional characteristics in a multi-dimensional comprehensive characteristic set and extracting key static characteristics, the space-time fusion layer is used for fusing the key static characteristics with dynamic change characteristics in real-time operation data of the warehouse-in system to generate space-time fusion characteristics, and the output layer is used for outputting a warehouse-in priority list of reagents based on the space-time fusion characteristics and an adaptive storage scheme, and the adaptive storage scheme comprises a target storage area number, a target temperature and humidity reference value and storage mode requirements.
9. The reagent warehouse linkage warehouse entry system based on the improved Transformer large model of claim 6, wherein the generating process of the pre-determined warehouse entry plan in S5 comprises: S51, constructing a demand prediction sample set based on real-time inventory quantity, warehousing priority comprehensive scores and historical demand data of a warehousing system; s52, processing the demand prediction sample set by adopting a sliding window prediction algorithm; And S53, determining the pre-judging warehouse-in quantity and warehouse-in time nodes of each type of reagent in a preset time period in the future according to the output result of the sliding window prediction algorithm and combining the weight coefficient corresponding to the warehouse-in priority comprehensive score to form a pre-judging warehouse-in plan.
10. The reagent warehouse linkage warehouse entry system based on the improved Transformer large model of claim 1, wherein the process of bi-directional verification in S7 comprises: s71, checking the integrity of the data of the warehouse-in data and the warehouse-in data of the warehouse system, and checking whether the key field is complete; s72, checking the numerical consistency of the complete warehouse-in data and the inventory data of the warehouse system; And S73, carrying out logic relevance verification on the data with consistent values, verifying the suitability of reagent types and storage areas in the warehouse-in data and the logicality of warehouse-in quantity and inventory variation, judging that the data are consistent when all the verification passes, and triggering a data correction flow when the verification does not pass.

Description

Reagent storage linkage warehouse entry system based on improved transducer large model Technical Field The invention relates to the technical field of warehouse management, in particular to a reagent warehouse linkage warehouse entry system based on an improved transducer large model. Background In the conventional reagent warehouse-in management, a core technical problem is that warehouse-in flow and warehouse-in demand pre-judging are seriously disjointed. The current system mainly depends on a warehousing system to send out a clear backorder signal before starting a warehousing process, and lacks an advanced pre-judging mechanism based on the real-time state of warehousing and the use rule of the reagent. In the mode, when the warehouse is subjected to sudden demands such as increased reagent consumption in emergency scientific research experiments, conventional reagents are deteriorated in advance due to fluctuation of storage environments, and the like, the warehouse-in process cannot respond quickly, and backout or untimely warehouse-in is easily caused. The reagent is used as special materials, part of types have the characteristics of short validity period, harsh storage conditions and the like, the lack of stock can influence the experimental progress or the medical diagnosis efficiency, the stock gap can be further aggravated if the stock is not in time, and the prior art can not effectively solve the linkage response problem caused by the requirement pre-judgment lag. Based on the above problems, a technical scheme capable of pre-judging the storage requirement in advance and realizing the dynamic linkage of storage and warehouse is needed. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides a reagent storage linkage warehouse-in system based on an improved transducer large model, which comprises the following steps: s1, collecting reagent attribute data, real-time operation data of a warehouse system and reagent use demand data; s2, carrying out standardized pretreatment on the acquired reagent attribute data, the real-time operation data of the warehousing system and the reagent use requirement data to construct a multi-dimensional comprehensive feature set; s3, extracting key features by improving a transducer large model based on the multi-dimensional comprehensive feature set, and generating a reagent warehouse-in priority list and an adaptive storage scheme; S4, controlling a warehousing execution assembly to complete automatic transportation and warehousing positioning of the reagent based on the warehousing priority list and the adaptive storage scheme; s5, synchronizing the warehousing progress data and the real-time operation data of the warehousing system in real time, and pre-judging the reagent requirements in a future preset time period of the warehousing system according to the synchronization result to generate a pre-judging warehousing plan; S6, monitoring environmental parameters of a target storage area in the warehousing system in real time, and if the environmental parameters do not meet the requirements of an adaptive storage scheme, adjusting the environmental parameters by an environmental control system of the linked warehousing system; and S7, after the warehousing operation is finished, bidirectional checking is carried out on the warehousing data and the inventory data of the warehousing system, so that the data are ensured to be completely consistent. Preferably, the reagent attribute data in S1 includes chemical stability, temperature and humidity sensitivity, remaining duration of validity period, unsealing state, reagent type, and physical state of the reagent. Further preferably, the real-time operation data of the warehousing system in S1 includes a real-time inventory number of the warehousing system, a current temperature and humidity value of a storage area, an air current disturbance influence parameter of the storage area, an occupancy rate of the storage area, and an operation state of warehousing equipment. Further preferably, the reagent usage demand data in S1 includes urgency of usage of the reagent, demand frequency, single usage amount, usage department priority, turnover frequency weight. Further preferably, the method further comprises the step of calculating a storage risk coefficient of the reagent through a storage risk coefficient formula, wherein the storage risk coefficient formula is constructed based on the chemical stability, the temperature and humidity sensitivity, the unsealing state, the physical state, the remaining duration of the validity period and the total duration of the validity period of the reagent. Further preferably, the method further comprises the step of calculating the comprehensive grading of the warehousing priority of the reagent through a comprehensive grading formula of the warehousing priority, wherein the comprehensive grading formula of the warehousing priority is const