A flotation recovery rate prediction method and system based on multi-source data

Abstract

The present invention discloses a flotation recovery rate prediction method and system based on multi-source data. The method comprises: marking the last target data of each cache subsequence in the cache sequence formed after splicing, and being able to directly determine that the last target data in each cache subsequence is dynamic target data; and predicting the predicted data corresponding to the time label by considering the time span between the time label and the time node of the first target data and the time node of the second target data, and being able to more effectively improve the accuracy of the predicted data; and fitting the predicted result by the actual result, and being able to improve the accuracy of the output of the predicted result.

Description

Technical Field

[0001] The invention belongs to the technical field of mineral processing data, and in particular relates to a flotation recovery rate prediction method and system based on multi-source data. Background Art

[0002] Dashan Ore Dressing Plant achieves efficient production of copper concentrate through a series of processes such as ore crushing, grinding, and flotation. Flotation recovery rate is an important indicator for assessing and measuring the mineral processing technology, management level, and recovery degree of useful components in the selected ore of a mining enterprise. The higher the recovery rate, the more useful components are recovered in the mineral processing process (or operation). Flotation recovery rate refers to the ratio of the mass of useful components in the concentrate to the mass of the useful components in the original ore. Generally, under the premise of ensuring the quality requirements of the concentrate, efforts are made to improve the recovery rate of useful components. This recovery rate is the actual recovery rate of valuable elements obtained under certain mineral processing conditions.

[0003] However, in the process of predicting flotation recovery rate in industry, due to the lack of unified data sources, the time of collecting multi-source data is likely to be inconsistent. When it is necessary to predict the flotation recovery rate at a certain moment, there is a phenomenon that the data from a certain source at a certain moment is missing. If the data before a certain moment is directly used as the data at a certain moment, the flotation recovery rate prediction may be inaccurate or have large deviations due to data lag. Summary of the invention

[0004] The present invention provides a flotation recovery rate prediction method and system based on multi-source data, which is used to solve the technical problem that directly using data before a certain moment as data at a certain moment may lead to inaccurate or large deviation in flotation recovery rate prediction due to data lag.

[0005] In a first aspect, the present invention provides a flotation recovery rate prediction method based on multi-source data, comprising:

[0006] Based on the frequency corresponding to each of the multiple data sources, receiving target data sent by each of the data sources, wherein the target data includes a time node for receiving the target data;

[0007] According to a preset cache rule, the target data sent by each data source is stored in a cache library, wherein the cache library contains a cache sequence corresponding to each data source, and a cache sequence only contains at least one target data sent by a data source;

[0008] Obtaining a user's instruction, and determining whether a target data associated with the instruction exists in a cache sequence, wherein the instruction includes a time tag;

[0009] If a certain target data associated with the instruction does not exist in a certain cache sequence, obtaining first target data and second target data corresponding to two time nodes adjacent to the time tag;

[0010] According to the first target data and the second target data, the prediction data corresponding to the time tag is determined by using a preset data allocation rule, and the prediction data is corrected according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences;

[0011] Each target data or target prediction data corresponding to the time tag is input into a flotation recovery rate prediction model pre-constructed based on an improved BP neural network, and the flotation recovery rate prediction model outputs a flotation recovery rate.

[0012] In a second aspect, the present invention provides a flotation recovery rate prediction system based on multi-source data, comprising:

[0013] A receiving module, configured to receive target data sent by each of the multiple data sources based on a frequency corresponding to each of the data sources, wherein the target data includes a time node for receiving the target data;

[0014] A storage module, configured to store the target data sent by each of the data sources into a cache library according to a preset cache rule, wherein the cache library contains a cache sequence corresponding to each of the data sources, and a cache sequence only contains at least one target data sent by a data source;

[0015] A determination module configured to obtain a user's instruction and determine whether there is a certain target data associated with the instruction in a certain cache sequence, wherein the instruction includes a time tag;

[0016] an acquisition module, configured to acquire first target data and second target data corresponding to two time nodes adjacent to the time tag if a target data associated with the instruction does not exist in a cache sequence;

[0017] a correction module, configured to determine the prediction data corresponding to the time tag according to the first target data and the second target data by using a preset data allocation rule, and to correct the prediction data according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences;

[0018] The output module is configured to input each target data or target prediction data corresponding to the time tag into a flotation recovery rate prediction model pre-built based on the improved BP neural network, and the flotation recovery rate prediction model outputs the flotation recovery rate.

[0019] In a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the flotation recovery rate prediction method based on multi-source data of any embodiment of the present invention.

[0020] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the flotation recovery rate prediction method based on multi-source data of any embodiment of the present invention.

[0021] The flotation recovery rate prediction method and system based on multi-source data of the present application can directly determine that the last target data in each cache subsequence is dynamic target data by setting cache subsequences, then simply splicing each cache subsequence based on time order, and marking the last target data of each cache subsequence in the cache sequence formed after splicing, and predict the predicted data corresponding to the time label by considering the first time span between the time label and the time node of the first target data and the second time span between the time label and the time node of the second target data, so as to effectively improve the accuracy of the predicted data, and fit the predicted result through the actual result, so as to improve the accuracy of the output of the predicted result. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 A flowchart of a flotation recovery rate prediction method based on multi-source data provided by one embodiment of the present invention;

[0024] Figure 2 A structural block diagram of a flotation recovery rate prediction system based on multi-source data provided by an embodiment of the present invention;

[0025] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0027] See also Figure 1 , which shows a flow chart of a flotation recovery rate prediction method based on multi-source data of the present application.

[0028] like Figure 1 As shown, the flotation recovery rate prediction method based on multi-source data specifically includes the following steps:

[0029] Step S101 : based on the frequency corresponding to each of the multiple data sources, receiving target data sent by each of the data sources, wherein the target data includes a time node for receiving the target data.

[0030] The process flow of a mine flotation workshop is affected by many factors, so when processing large-scale flotation process data, there will be a problem of data set dimension redundancy. Selecting key flotation condition data can simplify the data set, reduce the difficulty of model learning, and accelerate convergence without reducing the accuracy of flotation recovery rate prediction.

[0031] In this step, four flotation data, namely, raw ore grade data, lime dosage data, butyl ammonium nitrophenol dosage data and pine oil dosage data, are used as input variables of the flotation recovery prediction model. Therefore, based on the frequency corresponding to each data source in the multiple data sources, the raw ore grade data, lime dosage data, butyl ammonium nitrophenol dosage data and pine oil dosage data sent by each data source are received.

[0032] Step S102, storing the target data sent by each of the data sources in a cache library according to a preset cache rule, wherein the cache library contains a cache sequence corresponding to each of the data sources, and a cache sequence only contains at least one target data sent by a data source.

[0033] In this step, the cache library includes a transmission queue corresponding to each data source and a cache sequence connected to the transmission queue. For example, one data source corresponds to one transmission queue and one cache sequence connected to one transmission queue.

[0034] It should be noted that the target data sent by a certain data source at a first time node is transmitted to a certain transmission queue, and it is determined whether there are other target data corresponding to the first time node in other transmission queues, and the other target data are the target data sent by other data sources except the certain data source; if there are no other target data corresponding to the first time node in other transmission queues, then a certain transmission queue continues to receive the target data sent by a certain data source; if there are other target data corresponding to the first time node in other transmission queues, then the target data corresponding to the first time node and the previous time nodes in a certain transmission queue are stored in the corresponding cache subsequence based on time order, and each cache subsequence is spliced ​​to obtain a cache sequence.

[0035] In this embodiment, by setting cache subsequences, and then simply splicing each cache subsequence based on time order, and marking the last target data of each cache subsequence in the cache sequence formed after splicing, it is possible to directly determine that the last target data in each cache subsequence is dynamic target data. In this way, when steps S103 to S106 are executed, two dynamic target data adjacent to the time tag can be found more quickly.

[0036] Step S103, obtaining a user's instruction, and determining whether there is a certain target data associated with the instruction in a certain cache sequence, wherein the instruction includes a time tag.

[0037] For example, the user's instruction includes voice input or text input "obtain the flotation recovery rate prediction result of time node A". At this time, it is determined whether there is a certain target data associated with time node A in a certain cache sequence.

[0038] In a specific embodiment, if there is certain target data associated with the instruction, the certain target data is directly input into a flotation recovery rate prediction model pre-built based on the improved BP neural network.

[0039] Step S104: if a certain target data associated with the instruction does not exist in a certain cache sequence, then obtain the first target data and the second target data corresponding to two time nodes adjacent to the time tag.

[0040] Step S105, based on the first target data and the second target data, a preset data allocation rule is used to determine the prediction data corresponding to the time tag, and the prediction data is corrected according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences.

[0041] In this step, a first difference between the time node of the first target data and the time tag, and a second difference between the time node of the second target data and the time tag are calculated; a first predicted value of the predicted data corresponding to the time tag is determined according to the first difference and the second difference, wherein the expression of the first predicted value is: , where is the first predicted value, is the first difference, is the second difference, is the value in the first target data, is the value in the second target data; calculate the third difference between the time node of the first dynamic target data and the time label, and the fourth difference between the time node of the second dynamic target data and the time label, wherein the first dynamic target data is a dynamic target data whose time node is adjacent to the time label, and the second dynamic target data is another dynamic target data whose time node is adjacent to the time label; determine the second predicted value of the predicted data corresponding to the time label according to the third difference and the fourth difference; judge whether the predicted difference between the second predicted value and the first predicted value is greater than a preset threshold; if it is greater than the prediction threshold, take the average of the first predicted value and the second predicted value as the corrected value, that is, obtain the target predicted data; if it is not greater than the prediction threshold, directly take the first predicted value as the corrected value, that is, obtain the target predicted data.

[0042] It should be noted that the first difference, the second difference, the third difference and the fourth difference are all time differences, and the method for calculating the second predicted value is the same as the principle for calculating the first predicted value.

[0043] In this embodiment, the closer the data is to the time tag, the greater the reference value of the data. Therefore, by considering the time span between the time tag and the time node of the first target data and the time node of the second target data to predict the predicted data corresponding to the time tag, the accuracy of the predicted data can be more effectively improved, and the dynamic target data is used to correct the predicted data to obtain the target predicted data. The predicted results corresponding to the target predicted data can be fitted according to the actual results corresponding to the dynamic target data (all dynamic target data input into the flotation recovery prediction model are real data), so that the predicted results may be closer to the real results, thereby further improving the accuracy of the predicted data.

[0044] Step S106, inputting each target data or target prediction data corresponding to the time tag into a flotation recovery rate prediction model pre-constructed based on the improved BP neural network, and the flotation recovery rate prediction model outputs the flotation recovery rate.

[0045] Before inputting each target data and / or target prediction data corresponding to the time tag into the flotation recovery prediction model pre-built based on the improved BP neural network, the BP neural network is optimized according to the improved gray wolf algorithm to obtain the optimal hyperparameters, and the flotation recovery prediction model is built according to the optimal hyperparameters. The improved gray wolf algorithm includes a dynamic convergence factor, and the expression of the dynamic convergence factor is: , where is the dynamic convergence factor, is the initial convergence factor, t is the current iteration number, is the maximum number of iterations.

[0046] Furthermore, the predicted target prediction data and the directly collected target data are output to a flotation recovery rate prediction model pre-constructed based on the improved BP neural network, and the flotation recovery rate is output by the flotation recovery rate prediction model.

[0047] In summary, the method of the present application sets cache subsequences, then simply splices the cache subsequences based on time order, and marks the last target data of each cache subsequence in the cache sequence formed after splicing. It can directly determine that the last target data in each cache subsequence is dynamic target data, and predict the predicted data corresponding to the time label by considering the time span of the time label and the time node of the first target data and the time node of the second target data. It can effectively improve the accuracy of the predicted data, and fit the predicted results through the actual results to improve the accuracy of the predicted result output.

[0048] See also Figure 2, which shows a structural block diagram of a flotation recovery rate prediction system based on multi-source data of the present application.

[0049] like Figure 2 As shown, the flotation recovery rate prediction system 200 includes a receiving module 210 , a storage module 220 , a determination module 230 , an acquisition module 240 , a correction module 250 and an output module 260 .

[0050] The receiving module 210 is configured to receive target data sent by each of the multiple data sources based on the frequency corresponding to each of the data sources, wherein the target data includes a time node for receiving the target data;

[0051] The storage module 220 is configured to store the target data sent by each of the data sources into a cache library according to a preset cache rule, wherein the cache library contains a cache sequence corresponding to each of the data sources, and a cache sequence only contains at least one target data sent by a data source;

[0052] The judgment module 230 is configured to obtain a user's instruction and judge whether there is a certain target data associated with the instruction in a certain cache sequence, wherein the instruction includes a time tag;

[0053] The acquisition module 240 is configured to acquire first target data and second target data corresponding to two time nodes adjacent to the time tag if a target data associated with the instruction does not exist in a cache sequence;

[0054] The correction module 250 is configured to determine the prediction data corresponding to the time tag according to the first target data and the second target data by using a preset data allocation rule, and to correct the prediction data according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences;

[0055] The output module 260 is configured to input each target data or target prediction data corresponding to the time tag into a flotation recovery rate prediction model pre-constructed based on the improved BP neural network, and the flotation recovery rate prediction model outputs the flotation recovery rate.

[0056] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0057] In some other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the flotation recovery prediction method based on multi-source data in any of the above method embodiments;

[0058] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0059] Based on the frequency corresponding to each of the multiple data sources, receiving target data sent by each of the data sources, wherein the target data includes a time node for receiving the target data;

[0060] According to a preset cache rule, the target data sent by each data source is stored in a cache library, wherein the cache library contains a cache sequence corresponding to each data source, and a cache sequence only contains at least one target data sent by a data source;

[0061] Obtaining a user's instruction, and determining whether a target data associated with the instruction exists in a cache sequence, wherein the instruction includes a time tag;

[0062] If a certain target data associated with the instruction does not exist in a certain cache sequence, obtaining first target data and second target data corresponding to two time nodes adjacent to the time tag;

[0063] According to the first target data and the second target data, the prediction data corresponding to the time tag is determined by using a preset data allocation rule, and the prediction data is corrected according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences;

[0064] Each target data or target prediction data corresponding to the time tag is input into a flotation recovery rate prediction model pre-constructed based on an improved BP neural network, and the flotation recovery rate prediction model outputs a flotation recovery rate.

[0065] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the flotation recovery rate prediction system based on multi-source data, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the flotation recovery rate prediction system based on multi-source data via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0066] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the electronic device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 The example of the connection via bus is taken. The memory 320 is the computer-readable storage medium mentioned above. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the flotation recovery rate prediction method based on multi-source data of the above method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the flotation recovery rate prediction system based on multi-source data. The output device 340 may include display devices such as display screens.

[0067] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0068] As an embodiment, the electronic device is applied to a flotation recovery rate prediction system based on multi-source data, and is used for a client, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:

[0069] Based on the frequency corresponding to each of the multiple data sources, receiving target data sent by each of the data sources, wherein the target data includes a time node for receiving the target data;

[0070] According to a preset cache rule, the target data sent by each data source is stored in a cache library, wherein the cache library contains a cache sequence corresponding to each data source, and a cache sequence only contains at least one target data sent by a data source;

[0071] Obtaining a user's instruction, and determining whether a target data associated with the instruction exists in a cache sequence, wherein the instruction includes a time tag;

[0072] If a certain target data associated with the instruction does not exist in a certain cache sequence, obtaining first target data and second target data corresponding to two time nodes adjacent to the time tag;

[0073] According to the first target data and the second target data, the prediction data corresponding to the time tag is determined by using a preset data allocation rule, and the prediction data is corrected according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences;

[0074] Each target data or target prediction data corresponding to the time tag is input into a flotation recovery rate prediction model pre-constructed based on an improved BP neural network, and the flotation recovery rate prediction model outputs a flotation recovery rate.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A flotation recovery prediction method based on multi-source data, characterized in that: include: Based on the frequency corresponding to each of the multiple data sources, receiving target data sent by each of the data sources, wherein the target data includes a time node for receiving the target data; According to a preset cache rule, the target data sent by each data source is stored in a cache library, wherein the cache library contains a cache sequence corresponding to each data source, and a cache sequence only contains at least one target data sent by a data source; Obtaining a user's instruction, and determining whether a target data associated with the instruction exists in a cache sequence, wherein the instruction includes a time tag; If a certain target data associated with the instruction does not exist in a certain cache sequence, obtaining first target data and second target data corresponding to two time nodes adjacent to the time tag; According to the first target data and the second target data, the prediction data corresponding to the time tag is determined by using a preset data allocation rule, and the prediction data is corrected according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences; Each target data or target prediction data corresponding to the time tag is input into a flotation recovery rate prediction model pre-constructed based on an improved BP neural network, and the flotation recovery rate prediction model outputs a flotation recovery rate.

2. The flotation recovery prediction method based on multi-source data according to claim 1, characterized in that: The cache library includes a transmission queue corresponding to each of the data sources and a cache sequence connected to the transmission queue; The storing the target data sent by each data source into the cache according to the preset cache rule comprises: Transmitting target data sent by a data source at a first time node to a transmission queue, and determining whether other target data corresponding to the first time node exists in other transmission queues, wherein the other target data is target data sent by other data sources other than the data source; If there is no other target data corresponding to the first time node in other transmission queues, the certain transmission queue continues to receive the target data sent by a certain data source; If other target data corresponding to the first time node exists in other transmission queues, the target data corresponding to the first time node and the previous time nodes in the certain transmission queue will be stored in the corresponding cache subsequence based on time order, and each cache subsequence will be spliced ​​to obtain a cache sequence.

3. The flotation recovery prediction method based on multi-source data according to claim 1 is characterized in that: After determining whether there is a certain target data associated with the instruction in a certain cache sequence, the method further includes: If there is a certain target data associated with the instruction, the certain target data is directly input into the flotation recovery rate prediction model pre-built based on the improved BP neural network.

4. The method for predicting flotation recovery rate based on multi-source data according to claim 1, characterized in that: The step of determining the prediction data corresponding to the time tag by using a preset data allocation rule according to the first target data and the second target data, and correcting the prediction data according to two adjacent dynamic target data of the time tag in a certain cache sequence to obtain the target prediction data includes: Calculating a first difference between a time node of the first target data and the time tag, and a second difference between a time node of the second target data and the time tag; A first predicted value of the predicted data corresponding to the time tag is determined according to the first difference and the second difference, wherein the expression of the first predicted value is: , In the formula, is the first predicted value, is the first difference, is the second difference, is the value in the first target data, is the value in the second target data; Calculate a third difference between a time node of a first dynamic target data and the time tag, and a fourth difference between a time node of a second dynamic target data and the time tag, wherein the first dynamic target data is a dynamic target data whose time node is adjacent to the time tag, and the second dynamic target data is another dynamic target data whose time node is adjacent to the time tag; Determine a second predicted value of the predicted data corresponding to the time tag according to the third difference and the fourth difference; Determine whether a predicted difference between the second predicted value and the first predicted value is greater than a preset threshold; If it is greater than the preset threshold, the average of the first predicted value and the second predicted value is taken as the corrected value, that is, the target predicted data is obtained; If it is not greater than the preset threshold, the first predicted value is directly used as the corrected value to obtain the target predicted data.

5. The method for predicting flotation recovery rate based on multi-source data according to claim 1, characterized in that: After determining the prediction data corresponding to the time tag according to the first target data and the second target data by using a preset data allocation rule, and correcting the prediction data according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain the target prediction data, the method further includes: The time nodes corresponding to the target prediction data are stored in the certain cache sequence to obtain an updated certain cache sequence.

6. The method for predicting flotation recovery rate based on multi-source data according to claim 1, characterized in that: Before inputting each target data or target prediction data corresponding to the time tag into the flotation recovery rate prediction model pre-built based on the improved BP neural network, the method further includes: The BP neural network is optimized according to the improved grey wolf algorithm to obtain the optimal hyperparameters, and a flotation recovery rate prediction model is constructed according to the optimal hyperparameters.

7. A flotation recovery prediction system based on multi-source data, characterized in that: include: A receiving module, configured to receive target data sent by each of the multiple data sources based on a frequency corresponding to each of the data sources, wherein the target data includes a time node for receiving the target data; A storage module, configured to store the target data sent by each of the data sources into a cache library according to a preset cache rule, wherein the cache library contains a cache sequence corresponding to each of the data sources, and a cache sequence only contains at least one target data sent by a data source; A determination module configured to obtain a user's instruction and determine whether there is a certain target data associated with the instruction in a certain cache sequence, wherein the instruction includes a time tag; an acquisition module, configured to acquire first target data and second target data corresponding to two time nodes adjacent to the time tag if a certain target data associated with the instruction does not exist in a certain cache sequence; a correction module, configured to determine the prediction data corresponding to the time tag according to the first target data and the second target data by using a preset data allocation rule, and to correct the prediction data according to two dynamic target data adjacent to the time tag in the certain cache sequence to obtain target prediction data, wherein the dynamic target data is the target data corresponding to a certain time node, the time span between the certain time node and the time tag is the smallest, and the certain time node exists in other cache sequences; The output module is configured to input each target data or target prediction data corresponding to the time tag into a flotation recovery rate prediction model pre-built based on the improved BP neural network, and the flotation recovery rate prediction model outputs the flotation recovery rate.

8. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images