CN-122009771-A - Tomato unloading flow optimization method and system based on big data analysis

CN122009771ACN 122009771 ACN122009771 ACN 122009771ACN-122009771-A

Abstract

The invention relates to the technical field of tomato unloading optimization, and discloses a tomato unloading flow optimization method and system based on big data analysis, wherein the method comprises the steps of collecting multi-source data in a historical unloading process and constructing a historical operation database; setting a plurality of flows to be optimized in the unloading process based on a historical operation database, constructing an unloading optimization model according to all flows to be optimized, collecting real-time data of the current flows to be optimized, generating an initial optimization strategy of the current flows to be optimized by combining the unloading optimization model, collecting feedback data according to a preset feedback time node, judging whether to perform automatic execution on the initial optimization strategy of the current flows to be optimized and a correction instruction of the unloading optimization model, adapting to different requirements in various complex scenes, improving the unloading efficiency and effect of tomatoes, and reducing resource consumption.

Inventors

LU HAIXIA
Hou Binghong
JI WEI

Assignees

新疆铭鼎高科投资发展有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (10)

1. The tomato unloading process optimization method based on big data analysis is characterized by comprising the following steps of: The method comprises the steps of collecting multi-source data in a historical unloading process, and constructing a historical operation database, wherein the multi-source data comprises vehicle characteristic data, logistics state data, operation control data and operation result data; setting a plurality of flows to be optimized in the unloading process based on a historical operation database, and constructing an unloading optimization model according to all flows to be optimized; collecting real-time data of a current flow to be optimized, and generating an initial optimization strategy of the current flow to be optimized by combining a discharging optimization model; and collecting feedback data according to a preset feedback time node, and judging whether to perform initial optimization strategy on the current flow to be optimized and a correction instruction of the unloading optimization model.
2. The method for optimizing a tomato discharge flow based on big data analysis according to claim 1, wherein the discharge process comprises vehicle approach, vehicle stopping and information identification, discharge preparation, intelligent flushing and discharge, crane tube reset and discharge completion vehicle departure; the intelligent flushing and unloading comprises tomato region identification and crane tube movement track generation.
3. The tomato discharge flow optimization method based on big data analysis as claimed in claim 2, wherein the setting of a plurality of flows to be optimized in the discharge process based on the historical operation database comprises: Randomly selecting one unloading process as a target unloading process; extracting historical control class data of a target unloading process from a historical operation data packet, and calculating the influence degree of different historical control class data on other unloading processes and operation result data; If the influence degree of the historical control data is larger than a preset influence degree threshold, setting a target unloading process as a process to be optimized; and sequentially setting a plurality of flows to be optimized in the unloading process.
4. A method for optimizing tomato discharge flow based on big data analysis as claimed in claim 3, wherein calculating the influence degree of different historical control class data on other discharge processes and operation result data comprises: Presetting a plurality of process indexes in the unloading process and a plurality of result indexes in operation result data, wherein each index is mapped with a corresponding weight coefficient; Analyzing each historical control class data and each index based on a correlation analysis algorithm to obtain a correlation coefficient of each historical control class data and each index; Performing weight average value processing according to the correlation coefficients of the same historical control class data and all indexes and the weight coefficients of the corresponding indexes, and generating the influence degree of the corresponding historical control class data according to the processing result; The degree of influence of each historical control class data is generated in turn.
5. A method for optimizing a tomato discharge flow based on big data analysis as claimed in claim 3, wherein constructing a discharge optimization model based on all flows to be optimized comprises: Setting historical control class data with the influence degree larger than a preset influence degree threshold value in a flow to be optimized as data to be optimized; acquiring a control duration interval of data to be optimized, and setting the number of control points according to the influence degree; Generating a plurality of control duration subintervals of the data to be optimized according to the control duration intervals and the control point number; Determining a plurality of static scene information of a process to be optimized based on a historical operation database, and performing cluster analysis to obtain a plurality of similar scene clusters; Extracting a plurality of historical value sequences of data to be optimized meeting standard operation result data in each similar scene cluster, and performing control point time alignment and weight processing to obtain a standard value interval of each control duration subinterval; constructing a three-dimensional optimization parameter matrix taking a scene cluster as an index, a control duration subinterval as a horizontal axis and a standard value interval as a vertical axis; and training based on the three-dimensional optimization parameter matrix and a machine learning algorithm to obtain a discharging optimization model.
6. The method for optimizing a tomato discharge flow based on big data analysis according to claim 5, wherein collecting real-time data of a current flow to be optimized and generating an initial optimization strategy of the current flow to be optimized by combining a discharge optimization model comprises: Collecting real-time data of a current flow to be optimized, and dividing the real-time data into real-time static scene information and real-time control duration of a plurality of data to be optimized; Inputting real-time static scene information and real-time control time lengths of a plurality of data to be optimized into a discharge optimization model to obtain a real-time standard value interval and a prediction standard value interval of a future control time length subinterval of each data to be optimized of a current flow to be optimized; And generating an initial optimization strategy of the current flow to be optimized according to the real-time standard value interval of all the data to be optimized of the current flow to be optimized and the prediction standard value interval of the future control duration subinterval.
7. The tomato discharge flow optimization method based on big data analysis as claimed in claim 1, wherein the steps of collecting feedback data according to a preset feedback time node, and judging whether to generate an initial optimization strategy of a current flow to be optimized and a correction instruction of a discharge optimization model, include: The feedback data comprise actual value intervals of the data to be optimized in all control time intervals and characteristic result data of the current flow to be optimized; Calculating a first deviation value of an actual value interval and a corresponding standard value interval of each control duration subinterval, and calculating a second deviation value of characteristic result data and a standard result data interval of a current flow to be optimized; Generating a deviation coefficient according to the first deviation value and the second deviation value; and judging whether to generate an initial optimization strategy of the current flow to be optimized and a correction instruction of the unloading optimization model according to the deviation coefficient.
8. The method for optimizing a tomato discharge flow based on big data analysis according to claim 7, wherein the step of determining whether to generate an initial optimization strategy of a current flow to be optimized and a correction instruction of a discharge optimization model according to the deviation coefficient comprises: Presetting a deviation coefficient threshold; when the deviation coefficient is larger than the deviation coefficient threshold value, generating an initial optimization strategy of the current flow to be optimized and a correction instruction of the unloading optimization model; And when the deviation coefficient is not greater than the deviation coefficient threshold value, not generating an initial optimization strategy of the current flow to be optimized and a correction instruction of the unloading optimization model.
9. The method for optimizing a tomato discharge flow based on big data analysis according to claim 8, wherein the initial optimization strategy for the current flow to be optimized and the correction instruction for the discharge optimization model comprise: calculating a deviation coefficient difference value; And according to the corresponding relation between the difference value of the deviation coefficient and the preset difference value interval, selecting a corresponding correction strategy to dynamically adjust the initial optimization strategy, and updating parameters of the unloading optimization model according to the correction strategy.
10. Tomato unloading flow optimization system based on big data analysis, characterized by comprising: The system comprises an acquisition module, a data processing module and a data processing module, wherein the acquisition module is used for acquiring multi-source data in a historical unloading process and constructing a historical operation database, and the multi-source data comprises vehicle characteristic data, logistics state data, operation control data and operation result data; the construction module is used for setting a plurality of to-be-optimized flows in the unloading process based on the historical operation database, and constructing an unloading optimization model according to all to-be-optimized flows; The generating module is used for collecting real-time data of the current flow to be optimized and generating an initial optimizing strategy of the current flow to be optimized by combining the unloading optimizing model; the correction module is used for collecting feedback data according to a preset feedback time node and judging whether to perform correction instructions on an initial optimization strategy and a discharging optimization model of the current flow to be optimized.

Description

Tomato unloading flow optimization method and system based on big data analysis Technical Field The application relates to the technical field of tomato unloading optimization, in particular to a tomato unloading flow optimization method and system based on big data analysis. Background In the food processing industry such as tomato sauce, the unloading of raw material tomatoes is the first key process of a production line. The traditional unloading mode mainly relies on manually-operated high-pressure water guns or oil filling riser to flush tomatoes in a transportation carriage, and has the problems of high labor intensity, low efficiency, incomplete unloading, water resource waste, potential safety hazard and the like. In the prior art, although some automatic unloading equipment adopting machine vision for positioning can replace manual flushing operation, the systems often only operate independently for a single unloading task and lack deep excavation and utilization of historical operation data. The control parameters (such as flushing track, time and pressure) are usually set based on preset experience, and cannot adapt to the differentiated requirements of different vehicle types and different loading states (such as stacking shape, density and water content of tomatoes), so that a discharging blind area, excessive flushing or efficiency fluctuation can possibly occur, and the dynamic balance of a discharging effect and resource consumption is difficult to realize. In addition, the feedback mechanism of the traditional system is mostly manually estimated after the fact, lacks real-time data acquisition and closed loop optimization capability, and is difficult to adjust strategies in time according to actual operation conditions, so that the optimization space of the discharging flow is limited. Disclosure of Invention In order to solve the technical problems, the application provides a tomato unloading process optimization method and a tomato unloading process optimization system based on big data analysis, which are characterized in that a historical operation database is constructed, a process to be optimized is determined, a three-dimensional optimization parameter matrix is constructed, a machine learning algorithm is combined to train an unloading optimization model, an initial optimization strategy is generated by combining real-time data acquisition and the model, feedback data of a preset feedback time node is introduced to carry out deviation analysis, dynamic correction of the optimization strategy and model parameters is realized, automatic execution of the unloading process is realized, the differentiated requirements under various complex scenes are adapted, the tomato unloading efficiency and effect are improved, and meanwhile, the resource consumption is reduced. In some embodiments of the present application, a method for optimizing a tomato unloading process based on big data analysis is provided, including: The method comprises the steps of collecting multi-source data in a historical unloading process, and constructing a historical operation database, wherein the multi-source data comprises vehicle characteristic data, logistics state data, operation control data and operation result data; setting a plurality of flows to be optimized in the unloading process based on a historical operation database, and constructing an unloading optimization model according to all flows to be optimized; collecting real-time data of a current flow to be optimized, and generating an initial optimization strategy of the current flow to be optimized by combining a discharging optimization model; and collecting feedback data according to a preset feedback time node, and judging whether to perform initial optimization strategy on the current flow to be optimized and a correction instruction of the unloading optimization model. In some embodiments of the application, the unloading process comprises vehicle approach, vehicle stopping and information identification, unloading preparation, intelligent flushing and unloading, crane tube resetting and unloading completion vehicle departure; the intelligent flushing and unloading comprises tomato region identification and crane tube movement track generation. In some embodiments of the present application, setting a number of flows to be optimized in the unloading process based on a historical job database includes: Randomly selecting one unloading process as a target unloading process; extracting historical control class data of a target unloading process from a historical operation data packet, and calculating the influence degree of different historical control class data on other unloading processes and operation result data; If the influence degree of the historical control data is larger than a preset influence degree threshold, setting a target unloading process as a process to be optimized; and sequentially setting a plurality of flows to be optimized