Sampling management method based on big data analysis
Through sampling management methods based on big data analysis, accurate sampling is carried out for different suppliers and carriage categories, the problems of low sampling efficiency and wrong quality judgment in the existing technology are solved, and efficient and accurate coal inlet management and power plant operation efficiency are achieved.
Patent Information
- Application Number
- CN202411859793.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-17
- Publication Date
- 2025-05-13
AI Technical Summary
The existing sampling methods for coal-fired entry into the factory cannot conduct accurate sampling for different suppliers and carriage categories, resulting in low efficiency of coal sampling and incorrect quality judgment, resulting in economic losses.
The sampling management method based on big data analysis is adopted, feature tags are extracted through historical sampling data, different carriage categories are constructed, and targeted samples are carried out according to factors such as suppliers, batches, coal types, etc., and the carriages to be inspected are dynamically adjusted to eliminate distorted data.
It improves the sampling efficiency and sampling quality of coal materials entering the factory, enhances the accuracy and entry efficiency of fuel quality judgment, and improves fuel management and power plant operation efficiency.
Smart Images
Figure CN119990855A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of fuel management technology, and in particular to a sampling management method based on big data analysis. Background Art
[0002] The current common form of sampling in carriages when transporting coal into the factory is to divide the carriage area into 3*6 grid areas, and the plant-side fuel management system randomly designates 3 sampling points and sends drilling instructions to the carriage sampler (screw sampler).
[0003] However, the current sampling method cannot conduct accurate spot checks on different suppliers and car types, which reduces the efficiency of coal sampling. At the same time, there is also the problem of improper sampling operation, which affects the final coal quality and leads to incorrect judgment, thus causing unnecessary economic losses. Summary of the invention
[0004] The purpose of this application is: to solve the above-mentioned technical problems, this application provides a sampling management method based on big data analysis, aiming to improve the sampling efficiency of coal entering the factory and the accuracy of sample data, and provide data support for coal quality judgment.
[0005] In some embodiments of the present application, multiple feature labels are extracted based on historical sampling data, and different carriage categories are constructed, so that targeted sampling can be performed on incoming fuel from different suppliers, batches, and types of coal, thereby improving the sampling efficiency and quality of incoming fuel, thereby improving the accuracy of fuel quality judgment and the efficiency of entry into the factory, and improving the fuel management efficiency of the factory.
[0006] In some embodiments of the present application, the overall sampling efficiency and fuel entry efficiency are improved by dynamically adjusting all the carriages to be sampled, and at the same time, the sample data of each carriage is screened to eliminate distorted data to avoid misjudgment of fuel quality due to improper sampling. The management efficiency of fuel is improved, thereby improving the operating efficiency of the power plant.
[0007] In some embodiments of the present application, a sampling management method based on big data analysis is provided, including: Generate multiple carriage categories based on historical sampling data, and generate sub-sampling strategies for each carriage category; Generate a first-level sampling inspection plan based on the parameters of all coal carriages in the current plant area, and obtain sampling data of each coal carriage based on the first-level sampling inspection plan; Determine whether to generate a secondary sampling plan based on all sampling data; Among them, when generating multiple carriage categories, it includes: Establish a car category sequence A, A=(a1, a2…ai…an), where ai is the i-th car category; n is the number of car categories.
[0008] In some embodiments of the present application, when generating a sub-sampling strategy for each car category, it includes: Set ai as the target car category in turn; Obtain all historical sampling results of the target carriage category and generate a sampling evaluation value f of the target carriage category; According to the sampling evaluation value f, the number of grid areas and the number of sampling points in the target car category are set, and a sub-sampling strategy for the target car category is established; Generate sub-sampling strategies for each carriage category in turn; Establish a sub-sampling strategy sequence T, T = (t1, t2…ti…tn), where ti is the sub-sampling strategy of the i-th car category; Set the update time node of the sub-adopted strategy sequence T.
[0009] In some embodiments of the present application, when generating the sampling evaluation value f of the target compartment category, it includes: f=e1*Q1*[ (βi*ki)]+e2*Q2*[ (µi*pi)]; Among them, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; θ1 is the number of sampling evaluation indicators; βi is the weight coefficient of the i-th sampling evaluation index; ki is the reference value of the i-th sampling evaluation index in the target car category; θ2 is the number of historical evaluation indicators; µi is the weight coefficient of the i-th historical evaluation index; pi is the reference value of the i-th historical evaluation index in the target car category.
[0010] In some embodiments of the present application, when a first-level sampling inspection plan is generated according to the parameters of all coal carriages in the current plant area, it includes: Establish a sequence of carriages to be inspected B, B = (b1, b2…bi…bm), where bi is the i-th carriage to be inspected in the current factory area; m is the number of carriages to be inspected; Set the primary sampling strategy for each carriage to be inspected in turn; Set the inspection order of each carriage to be inspected; Generate a first-level sampling plan based on the sampling order and all first-level sampling strategies.
[0011] In some embodiments of the present application, when the primary sampling strategy of each compartment to be inspected is set in sequence, it includes: Sequentially select the i-th carriage to be inspected as the target carriage to be inspected; Based on the similarity evaluation value between the target carriage to be sampled and each carriage category; Establish a similarity evaluation value sequence J, J = (j1, j2...ji...jn), where ji is the similarity between the target carriage to be inspected and the i-th carriage category; Set the sub-sampling strategy of the carriage category corresponding to the maximum value jmax in the similarity evaluation value sequence J as the primary sampling strategy of the target carriage to be inspected; Set the primary sampling strategy for each carriage to be inspected in turn; Set the inspection order of each carriage to be inspected; Generate a first-level sampling plan based on the sampling order and all first-level sampling strategies.
[0012] In some embodiments of the present application, when obtaining sampling data of each coal carriage according to the primary sampling plan, it includes: Obtain all sampling samples from the target carriages to be inspected; Establish a sampling sample sequence C, C=(c1, c2…c i …c m1 ), where c i is the i-th sample of the target carriage to be inspected; m1 is the number of samples; Generate the fluctuation evaluation value of each sampling sample in turn; Establish a series of fluctuation evaluation values D, D = (d1, d2…d i …d m1 ), where d i is the fluctuation evaluation value of the i-th sampling sample; Preset the first fluctuation evaluation value threshold D1; If d i ≥D1, then the i-th sample is eliminated; According to the elimination results, the uneliminated samples in the sample sequence C are set as the first-level samples of the target carriage to be inspected; Generate the first-level samples of each carriage to be inspected in turn.
[0013] In some embodiments of the present application, when the fluctuation evaluation value of each sample is generated in sequence, it includes: Sequentially set the i-th sampling sample as the target sampling sample; Generate a fluctuation evaluation value d of the target sampling sample; d= gi*(si-∆si); Among them, v is the number of coal material detection indicators; gi is the fixed coefficient of the i-th coal material detection indicator; si is the real-time value of the i-th coal material detection indicator in the target sampling sample; ∆si is the average value of the i-th coal material detection indicator in all sampling samples.
[0014] In some embodiments of the present application, when determining whether to generate a secondary sampling plan based on all sampling data, it includes: Obtain all first-level samples of the target carriage to be inspected; Establish a first-level sample sequence C1, C1=(c 11 , c 12 …c 1i …c 1m2 ), where c 1i is the i-th first-level sample in the target carriage to be inspected; m2 is the number of first-level samples; Generate a distortion evaluation value of the target carriage to be inspected based on all first-level samples; Generate the distortion evaluation value of each carriage to be inspected in sequence, and establish a distortion evaluation value sequence H, H = (h1, h2...hi...hm), where hi is the distortion evaluation value of the i-th carriage to be inspected; Preset distortion evaluation value threshold H1; If hi ≥ H1, generate a secondary inspection instruction for the i-th carriage to be inspected; Generate a secondary sampling plan based on all secondary sampling instructions.
[0015] In some embodiments of the present application, when generating the distortion evaluation value of the target compartment to be sampled, it includes: h=U*[ (d'i-∆d'i) 2 ]; d'i= gr*(sr-∆sr); Among them, h is the distortion evaluation value of the target carriage to be inspected; d'i is the fluctuation evaluation value of the i-th first-level sample of the target carriage to be inspected; ∆d'i is the average value of the fluctuation evaluation values of all first-level samples of the target carriage to be inspected; U is the compensation coefficient set based on the number of sampling samples eliminated from the target carriage to be inspected; ∆sr is the average value of the r-th coal detection index in all first-level samples of the target carriage to be inspected; sr is the real-time value of the r-th coal quantity monitoring index in the i-th first-level sample of the target carriage to be inspected; gr is the fixed coefficient of the r-th coal detection index; v is the number of coal detection indicators.
[0016] Compared with the prior art, the sampling management method based on big data analysis in the embodiment of the present application has the following beneficial effects: Based on historical sampling data, multiple feature labels are extracted and different carriage categories are constructed, so that targeted sampling can be carried out on the incoming fuel from different suppliers, batches and types of coal, thereby improving the sampling efficiency and quality of the incoming fuel, thereby improving the quality judgment accuracy of the fuel and the efficiency of entry into the factory, and improving the fuel management efficiency of the factory.
[0017] By dynamically adjusting all the carriages to be sampled, the overall sampling efficiency and fuel entry efficiency are improved. At the same time, the sample data of each carriage is screened to eliminate distorted data to avoid misjudgment of fuel quality due to improper sampling. The management efficiency of fuel is improved, thereby improving the operating efficiency of the power plant. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 It is a flow chart of a sampling management method based on big data analysis in the preferred embodiment of the present application. DETAILED DESCRIPTION
[0019] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.
[0020] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.
[0021] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.
[0022] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.
[0023] like Figure 1 As shown, a sampling management method based on big data analysis in a preferred embodiment of the present application includes: S101: Generate multiple carriage categories according to historical sampling data, and generate sub-sampling strategies for each carriage category; S102: Generate a first-level sampling inspection plan according to the parameters of all coal carriages in the current plant area, and obtain sampling data of each coal carriage according to the first-level sampling inspection plan; S103: Determine whether to generate a secondary sampling plan based on all sampling data; Among them, when generating multiple carriage categories, it includes: Establish a car category sequence A, A=(a1, a2…ai…an), where ai is the i-th car category; n is the number of car categories.
[0024] Specifically, multiple carriage categories are constructed based on the carriage supplier, batch, coal type in the carriage, and carriage equipment parameters, and sub-adoption strategies corresponding to each carriage category are constructed to accurately sample incoming carriages, improve sampling efficiency, and ensure the accuracy of sample data.
[0025] Specifically, when generating the subsampling strategy for each car category, it includes: Set ai as the target car category in turn; Obtain all historical sampling results of the target carriage category and generate a sampling evaluation value f of the target carriage category; According to the sampling evaluation value f, the number of grid areas and the number of sampling points in the target car category are set, and a sub-sampling strategy for the target car category is established; Generate sub-sampling strategies for each carriage category in turn; Establish a sub-sampling strategy sequence T, T = (t1, t2…ti…tn), where ti is the sub-sampling strategy of the i-th car category; Set the update time node of the sub-adopted strategy sequence T.
[0026] Specifically, the larger the sampling rating value, the more samples the current car category needs, the more detailed the grid division needs to be, and the more corresponding sampling points are required. The sampling points need to be generated randomly by computer. The larger the sampling evaluation value, the higher the corresponding randomness should be.
[0027] Specifically, when generating the sampling evaluation value f of the target car category, it includes: f=e1*Q1*[ (βi*ki)]+e2*Q2*[ (µi*pi)]; Among them, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; θ1 is the number of sampling evaluation indicators; βi is the weight coefficient of the i-th sampling evaluation index; ki is the reference value of the i-th sampling evaluation index in the target car category; θ2 is the number of historical evaluation indicators; µi is the weight coefficient of the i-th historical evaluation index; pi is the reference value of the i-th historical evaluation index in the target car category.
[0028] Specifically, the sampling evaluation index includes multiple parameters such as coal type, supplier, and carriage equipment parameters, and the historical evaluation index includes multiple parameters such as the historical sampling times of the target carriage category and the qualified rate. The larger the sampling evaluation value, the more coal samples are needed for the current carriage category.
[0029] Specifically, by presetting the first fixed coefficient and the second fixed coefficient, all parameters in the model are normalized so that all parameters are in the same value range.
[0030] It can be understood that in the above embodiment, the corresponding sub-sampling strategy is dynamically adjusted according to different car categories, and the sampling number is reduced for the coal-carrying car category with good long-term performance, thereby improving the overall sampling efficiency and further improving the management efficiency of fuel entering the factory. At the same time, each car category is periodically evaluated, and the corresponding sub-sampling strategy is corrected in time to ensure the overall sampling efficiency and the quality of the coal entering the factory.
[0031] In a preferred embodiment of the present application, when a first-level sampling inspection plan is generated according to the parameters of all coal carriages in the current plant area, it includes: Establish a sequence of carriages to be inspected B, B = (b1, b2…bi…bm), where bi is the i-th carriage to be inspected in the current factory area; m is the number of carriages to be inspected; Set the primary sampling strategy for each carriage to be inspected in turn; Set the inspection order of each carriage to be inspected; Generate a first-level sampling plan based on the sampling order and all first-level sampling strategies.
[0032] Specifically, the first-level sampling strategy for each carriage to be inspected is set in sequence, including: Sequentially select the i-th carriage to be inspected as the target carriage to be inspected; Based on the similarity evaluation value between the target carriage to be sampled and each carriage category; Establish a similarity evaluation value sequence J, J = (j1, j2...ji...jn), where ji is the similarity between the target carriage to be inspected and the i-th carriage category; Set the sub-sampling strategy of the carriage category corresponding to the maximum value jmax in the similarity evaluation value sequence J as the primary sampling strategy of the target carriage to be inspected; Set the primary sampling strategy for each carriage to be inspected in turn; Set the inspection order of each carriage to be inspected; Generate a first-level sampling plan based on the sampling order and all first-level sampling strategies.
[0033] Specifically, all carriages to be inspected within a single factory entry cycle are uniformly scheduled, and the corresponding sub-sampling strategy is selected according to the category of each carriage to be inspected. At the same time, the corresponding inspection sequence is set according to the entry order of the carriages to be inspected and the number of sampling stations, and an overall first-level inspection plan is constructed to achieve overall scheduling, ensure the overall sampling efficiency, and improve the efficiency of fuel entry into the factory.
[0034] In a preferred embodiment of the present application, when obtaining sampling data of each coal carriage according to the first-level sampling plan, it includes: Obtain all sampling samples from the target carriages to be inspected; Establish a sampling sample sequence C, C=(c1, c2…c i …c m1 ), where c i is the i-th sample of the target carriage to be inspected; m1 is the number of samples; Generate the fluctuation evaluation value of each sampling sample in turn; Establish a series of fluctuation evaluation values D, D = (d1, d2…d i …d m1 ), where d i is the fluctuation evaluation value of the i-th sampling sample; Preset the first fluctuation evaluation value threshold D1; If d i ≥D1, then the i-th sample is eliminated; According to the elimination results, the uneliminated samples in the sample sequence C are set as the first-level samples of the target carriage to be inspected; Generate the first-level samples of each carriage to be inspected in turn.
[0035] Specifically, when generating the fluctuation evaluation value of each sampling sample in sequence, it includes: Sequentially set the i-th sampling sample as the target sampling sample; Generate a fluctuation evaluation value d of the target sampling sample; d= gi*(si-∆si); Among them, v is the number of coal material detection indicators; gi is the fixed coefficient of the i-th coal material detection indicator; si is the real-time value of the i-th coal material detection indicator in the target sampling sample; ∆si is the average value of the i-th coal material detection indicator in all sampling samples.
[0036] Specifically, based on the fluctuation evaluation value of each sampling sample, some sampling samples with large deviations are eliminated to eliminate the interference of sample data caused by improper sampling operations.
[0037] Specifically, when judging whether to generate a secondary sampling plan based on all sampling data, it includes: Obtain all first-level samples of the target carriage to be inspected; Establish a first-level sample sequence C1, C1=(c 11 , c 12 …c 1i …c 1m2 ), where c 1i is the i-th first-level sample in the target carriage to be inspected; m2 is the number of first-level samples; Generate a distortion evaluation value of the target carriage to be inspected based on all first-level samples; Generate the distortion evaluation value of each carriage to be inspected in sequence, and establish a distortion evaluation value sequence H, H = (h1, h2...hi...hm), where hi is the distortion evaluation value of the i-th carriage to be inspected; Preset distortion evaluation value threshold H1; If hi ≥ H1, generate a secondary inspection instruction for the i-th carriage to be inspected; Generate a secondary sampling plan based on all secondary sampling instructions.
[0038] Specifically, when generating the distortion evaluation value of the target compartment to be sampled, it includes: h=U*[ (d'i-∆d'i) 2 ]; d'i= gr*(sr-∆sr); Among them, h is the distortion evaluation value of the target carriage to be inspected; d'i is the fluctuation evaluation value of the i-th first-level sample of the target carriage to be inspected; ∆d'i is the average value of the fluctuation evaluation values of all first-level samples of the target carriage to be inspected; U is the compensation coefficient set based on the number of sampling samples eliminated from the target carriage to be inspected; ∆sr is the average value of the r-th coal detection index in all first-level samples of the target carriage to be inspected; sr is the real-time value of the r-th coal quantity monitoring index in the i-th first-level sample of the target carriage to be inspected; gr is the fixed coefficient of the r-th coal detection index; v is the number of coal detection indicators.
[0039] Specifically, the larger the number of samples eliminated, the larger the corresponding compensation coefficient.
[0040] Specifically, the larger the distortion evaluation value, the greater the degree of data discreteness in each current sampling sample. The distortion evaluation value threshold can be set according to the historical sampling data. When the distortion evaluation value is greater than the threshold, it means that there are serious errors in the sampling of the current target carriage to be inspected, and the authenticity of the data is in doubt. Re-sampling is required to avoid interference in the detection results due to data distortion.
[0041] It is understandable that in the above embodiment, through multi-level judgment of sampling samples, distorted data can be eliminated in time to avoid wrong judgment of fuel quality due to improper sampling, thereby improving the management efficiency of fuel and thus improving the operating efficiency of the power plant.
[0042] According to the first concept of the present application, multiple feature labels are extracted based on historical sampling data, and different carriage categories are constructed, so that targeted sampling can be carried out on the incoming fuel from different suppliers, different batches, and different types of coal, thereby improving the sampling efficiency and sampling quality of the incoming fuel, thereby improving the quality judgment accuracy of the fuel and the efficiency of entry into the factory, and improving the fuel management efficiency of the factory area.
[0043] According to the second concept of this application, by dynamically adjusting all the carriages to be sampled, the overall sampling efficiency and fuel entry efficiency are improved, and at the same time, the sample data of each carriage is screened to eliminate distorted data to avoid misjudgment of fuel quality due to improper sampling. The management efficiency of fuel is improved, thereby improving the operating efficiency of the power plant.
[0044] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.
Claims
1. A sampling management method based on big data analysis, characterized in that: include: Generate multiple carriage categories based on historical sampling data, and generate sub-sampling strategies for each carriage category; Generate a first-level sampling inspection plan based on the parameters of all coal carriages in the current plant area, and obtain sampling data of each coal carriage based on the first-level sampling inspection plan; Determine whether to generate a secondary sampling plan based on all sampling data; Among them, when generating multiple carriage categories, it includes: Establish a car category sequence A, A=(a1, a2…ai…an), where ai is the i-th car category; n is the number of car categories.
2. The sampling management method based on big data analysis according to claim 1, characterized in that: When generating subsampling strategies for each car class, include: Set ai as the target car category in turn; Obtain all historical sampling results of the target carriage category and generate a sampling evaluation value f of the target carriage category; According to the sampling evaluation value f, the number of grid areas and the number of sampling points in the target car category are set, and a sub-sampling strategy for the target car category is established; Generate sub-sampling strategies for each carriage category in turn; Establish a sub-sampling strategy sequence T, T = (t1, t2…ti…tn), where ti is the sub-sampling strategy of the i-th car category; Set the update time node of the sub-adopted strategy sequence T.
3. The sampling management method based on big data analysis according to claim 2, characterized in that: When generating the sampling evaluation value f of the target carriage category, it includes: f=e1*Q1*[ (βi*ki)]+e2*Q2*[ (µi*pi)]; Among them, e1 is the preset first weight coefficient; e2 is the preset second weight coefficient; Q1 is the preset first fixed coefficient; Q2 is the preset second fixed coefficient; θ1 is the number of sampling evaluation indicators; βi is the weight coefficient of the i-th sampling evaluation index; ki is the reference value of the i-th sampling evaluation index in the target car category; θ2 is the number of historical evaluation indicators; µi is the weight coefficient of the i-th historical evaluation index; pi is the reference value of the i-th historical evaluation index in the target car category.
4. The sampling management method based on big data analysis according to claim 2, characterized in that: When generating the first-level sampling inspection plan based on the parameters of all coal carriages in the current plant area, it includes: Establish a sequence of carriages to be inspected B, B = (b1, b2…bi…bm), where bi is the i-th carriage to be inspected in the current factory area; m is the number of carriages to be inspected; Set the primary sampling strategy for each carriage to be inspected in turn; Set the inspection order of each carriage to be inspected; Generate a first-level sampling plan based on the sampling order and all first-level sampling strategies.
5. The sampling management method based on big data analysis according to claim 4, characterized in that: When setting the primary sampling strategy for each carriage to be inspected in turn, it includes: Sequentially select the i-th carriage to be inspected as the target carriage to be inspected; Based on the similarity evaluation value between the target carriage to be sampled and each carriage category; Establish a similarity evaluation value sequence J, J = (j1, j2...ji...jn), where ji is the similarity between the target carriage to be sampled and the i-th carriage category; Set the sub-sampling strategy of the carriage category corresponding to the maximum value jmax in the similarity evaluation value sequence J as the primary sampling strategy of the target carriage to be inspected; Set the primary sampling strategy for each carriage to be inspected in turn; Set the inspection order of each carriage to be inspected; Generate a first-level sampling plan based on the sampling order and all first-level sampling strategies.
6. The sampling management method based on big data analysis according to claim 5, characterized in that: When obtaining sampling data for each coal carriage according to the primary sampling plan, it includes: Obtain all sampling samples from the target carriages to be inspected; Establish a sampling sample sequence C, C=(c1, c2…c i …c m1 ), where c i is the i-th sample of the target carriage to be inspected; m1 is the number of samples; Generate the fluctuation evaluation value of each sampling sample in turn; Establish a series of fluctuation evaluation values D, D = (d1, d2…d i …d m1 ), where d i is the fluctuation evaluation value of the i-th sampling sample; Preset the first fluctuation evaluation value threshold D1; If d i ≥D1, then the i-th sample is eliminated; According to the elimination results, the uneliminated samples in the sample sequence C are set as the first-level samples of the target carriage to be inspected; Generate the first-level samples of each carriage to be inspected in turn.
7. The sampling management method based on big data analysis according to claim 6, characterized in that: When generating the fluctuation evaluation value of each sampling sample in sequence, it includes: Sequentially set the i-th sampling sample as the target sampling sample; Generate a fluctuation evaluation value d of the target sampling sample; d= gi*(si-∆si); Among them, v is the number of coal material detection indicators; gi is the fixed coefficient of the i-th coal material detection indicator; si is the real-time value of the i-th coal material detection indicator in the target sampling sample; ∆si is the average value of the i-th coal material detection indicator in all sampling samples.
8. The sampling management method based on big data analysis according to claim 7, characterized in that: When judging whether to generate a secondary sampling plan based on all sampling data, it includes: Obtain all first-level samples of the target carriage to be inspected; Establish a first-level sample sequence C1, C1=(c 11 , c 12 …c 1i …c 1m2 ), where c 1i is the i-th first-level sample in the target carriage to be inspected; m2 is the number of first-level samples; Generate a distortion evaluation value of the target carriage to be inspected based on all first-level samples; Generate the distortion evaluation value of each carriage to be inspected in sequence, and establish a distortion evaluation value sequence H, H = (h1, h2...hi...hm), where hi is the distortion evaluation value of the i-th carriage to be inspected; Preset distortion evaluation value threshold H1; If hi ≥ H1, generate a secondary inspection instruction for the i-th carriage to be inspected; Generate a secondary sampling plan based on all secondary sampling instructions.
9. The sampling management method based on big data analysis according to claim 8, characterized in that: When generating the distortion evaluation value of the target compartment to be sampled, it includes: h=U*[ (d'i-∆d'i) 2 ]; d'i= gr*(sr-∆sr); Among them, h is the distortion evaluation value of the target carriage to be inspected; d'i is the fluctuation evaluation value of the i-th first-level sample of the target carriage to be inspected; ∆d'i is the average value of the fluctuation evaluation values of all first-level samples of the target carriage to be inspected; U is the compensation coefficient set based on the number of sampling samples eliminated from the target carriage to be inspected; ∆sr is the average value of the r-th coal detection index in all first-level samples of the target carriage to be inspected; sr is the real-time value of the r-th coal quantity monitoring index in the i-th first-level sample of the target carriage to be inspected; gr is the fixed coefficient of the r-th coal detection index; v is the number of coal detection indicators.