Comprehensive evaluation method and system for wet regeneration and cleaning technology of waste lead paste

Abstract

The present invention belongs to the field of comprehensive evaluation technology, and specifically relates to a comprehensive evaluation method and system for wet regeneration and cleaning technology of scrap lead paste. The method includes: obtaining evaluation results on the efficiency of the processing process, the effectiveness of the implementation of environmental protection measures, product benefits, and product quality; selecting one of the final inference algorithm and the final candidate inference algorithm to perform inference processing to obtain equipment inferred operation data at the inference time; generating a difference inference algorithm based on the difference between multiple equipment operation data; and, when the final inference algorithm is used for inference processing, calculating the final equipment inferred operation data at the inference time based on the difference between the equipment inferred operation data at the inference time and the equipment inferred operation data at the inference time, generating an evaluation result on the equipment operation stability, and integrating all the evaluation results. The present invention can provide an intelligent evaluation method.

Description

Technical Field

[0001] The present invention belongs to the field of comprehensive evaluation technology, and in particular relates to a comprehensive evaluation method and system for wet regeneration and cleaning technology of waste lead paste. Background Art

[0002] As people pay more and more attention to environmental protection and resource recycling, wet regeneration technology is used to extract lead from waste lead paste and make lead products. This not only solves environmental pollution problems but also creates economic benefits. However, there is currently a lack of comprehensive evaluation methods for this technology.

[0003] Similar prior art includes Chinese patent application publication number CN114168904A, which discloses a method for evaluating VOCs emission control technologies for low-quality heavy oil processing. The evaluation method includes the following steps: S100, conducting on-site and experimental testing on the low-quality heavy oil processing process to obtain VOCs emission characteristics for each sub-functional area; S200, screening important VOCs pollution sources corresponding to each sub-functional area as evaluation targets based on the VOCs emission characteristics; S300, establishing an evaluation index system framework for the evaluation targets; S400, establishing an evaluation matrix and assessment model for VOCs emission control technologies for each sub-functional area of ​​the low-quality heavy oil processing process, and selecting the optimal control technology for each sub-functional area to control its VOCs emissions. Furthermore, similar prior art includes Chinese patent application publication number CN103646147A, which proposes a comprehensive evaluation method for aerospace components based on a maturity model. This method quantifies component performance indicators into digital grades, providing a judgment method to ensure the selection of aerospace components. First, the concept of aerospace component maturity is defined, then a clear classification of component maturity levels is made, the main work content of each level classification is given, and then component evaluation criteria, comprehensive scoring methods, and evaluation implementation steps are given. Finally, a component maturity improvement process is provided, providing an important theoretical basis for the continuous improvement of aerospace component product performance. However, the evaluation methods proposed in the above two patent applications lack a certain degree of intelligence. Summary of the Invention

[0004] The present invention generates multiple inference algorithms and multiple candidate inference algorithms in advance, selects a final inference algorithm and a final candidate inference algorithm that are more suitable for inference processing, and further selects the most suitable algorithm for inference processing from the final inference algorithm and the final candidate inference algorithm, thereby obtaining the device inference operation data at the inference time. In addition, if the device inference operation data is generated by the final inference algorithm, the final device inference operation data at the inference time is calculated based on the difference between the device inference operation data at the inference time and the device inference operation data at the inference time, and then integrates the evaluation results of the device operation stability with all other evaluation results. The present invention aims to provide an intelligent evaluation method.

[0005] In order to achieve the above-mentioned object of the invention, the present invention provides the following comprehensive evaluation method for the wet regeneration and cleaning technology of waste lead paste, which mainly includes the following steps:

[0006] The preparation module collects a process data set, an environmental protection-related data set, an equipment operation data set, a cost data set, and a product test data set, and the analysis module analyzes and processes the process data set to obtain an evaluation result on the process efficiency, analyzes and processes the environmental protection-related data set to obtain an evaluation result on the effectiveness of the environmental protection measures, analyzes and processes the cost data set to obtain an evaluation result on the product benefit, and analyzes and processes the product test data set to obtain an evaluation result on the product quality;

[0007] The analysis module generates a plurality of inference algorithms and a plurality of candidate inference algorithms, determines a final inference algorithm from the plurality of inference algorithms, determines a final candidate inference algorithm from the plurality of candidate inference algorithms, and selects one of the final inference algorithm and the final candidate inference algorithm for inference processing to obtain inference operation data of the device at the inference time;

[0008] The analysis module obtains actual device operation data for a number of past moments and estimated device operation data for a number of past moments, and for each past moment, calculates a device operation data difference between the actual device operation data for the past moment and the estimated device operation data for the past moment. The analysis module also generates a difference estimation algorithm based on the device operation data differences.

[0009] When the final speculation algorithm is used for speculation processing, the analysis module calculates the final device speculation operating data at the speculation moment based on the device speculation operating data at the speculation moment and the difference between the device speculation operating data at the speculation moment output by the difference speculation algorithm. When the final candidate speculation algorithm is used for speculation processing, the analysis module directly uses the device speculation operating data at the speculation moment as the final device speculation operating data. Subsequently, the analysis module generates an evaluation result on the device operation stability based on the final device speculation operating data, and the comprehensive evaluation module integrates different evaluation results to generate a comprehensive evaluation result.

[0010] As a preferred technical solution of the present invention, the process of the analysis module generating several inference algorithms includes: first, the analysis module determines several factors that affect the inference results; secondly, the analysis module generates several factor combinations, each factor combination is composed of different factors, and generates several inference algorithms corresponding to the several factor combinations.

[0011] As a preferred technical solution of the present invention, the process of the analysis module generating several candidate inference algorithms includes: the analysis module generates several candidate inference algorithms corresponding to the several factor combinations based on the several factor combinations that have been generated.

[0012] As a preferred technical solution of the present invention, the inference algorithm is pre-trained by the analysis module using a training data set, and the training data set is composed of different training data, including the actual operation data of the equipment at the past moment, and different factor data corresponding to the past moment.

[0013] As a preferred technical solution of the present invention, the analysis module determines a final inference algorithm from a plurality of inference algorithms, and determines a final candidate inference algorithm from a plurality of candidate inference algorithms, including the following steps:

[0014] The analysis module selects a speculation algorithm from a plurality of speculation algorithms, and simultaneously selects a candidate speculation algorithm corresponding to the selected speculation algorithm from a plurality of candidate speculation algorithms;

[0015] The analysis module determines a number of past moments, and causes the selected inference algorithm to perform inference processing a number of times to obtain inferred operating data of the first device at the number of past moments, and causes the selected candidate inference algorithm to perform inference processing a number of times to obtain inferred operating data of the second device at the number of past moments;

[0016] The analysis module obtains actual device operation data at a certain number of past moments, calculates a first accuracy of a selected estimation algorithm based on the actual device operation data at the certain number of past moments and first device estimated operation data at the certain number of past moments using a preset formula, calculates a second accuracy of a selected candidate estimation algorithm based on the actual device operation data at the certain number of past moments and second device estimated operation data at the certain number of past moments using a preset formula, and calculates a comprehensive accuracy based on the first accuracy and the second accuracy;

[0017] The analysis module determines whether there is an unselected inference algorithm. If so, it jumps to the initial step of selecting the inference algorithm and the candidate inference algorithm. If not, the selected inference algorithm and the selected candidate inference algorithm corresponding to the maximum comprehensive accuracy are respectively regarded as the final inference algorithm and the final candidate inference algorithm.

[0018] As a preferred technical solution of the present invention, the analysis module selects one of the final inference algorithm and the final candidate inference algorithm to perform inference processing to obtain the device inference operation data at the inference time, including the following steps:

[0019] The analysis module obtains actual operation data of the device at several past moments within a past time period before the current moment, and for each past moment, the analysis module obtains different factor data corresponding to the past moment, and the analysis module also performs training processing on the final inference algorithm;

[0020] The analysis module determines a plurality of past moments in a past time period two before the current moment, and causes the final speculation algorithm to perform a plurality of speculation processes to obtain the third device speculated operating data at the plurality of past moments, and causes the final candidate speculation algorithm to also perform a plurality of speculation processes to obtain the fourth device speculated operating data at the plurality of past moments;

[0021] The analysis module obtains the actual operation data of the device at the determined several past moments, calculates a third accuracy of the final estimation algorithm based on the actual operation data of the device at the determined several past moments and the estimated operation data of the third device at the determined several past moments using a preset formula, and calculates a fourth accuracy of the final candidate estimation algorithm based on the actual operation data of the device at the determined several past moments and the estimated operation data of the fourth device at the determined several past moments using a preset formula;

[0022] The analysis module determines whether the third accuracy is greater than or equal to the fourth accuracy. If yes, the analysis module selects a final inference algorithm; if not, the analysis module selects a final candidate inference algorithm. The analysis module uses the selected algorithm to perform inference processing to obtain device inference operation data at the inference time.

[0023] As a preferred technical solution of the present invention, the preset formula is Among them, δ is the accuracy, k is the total number of real operation data of the equipment, α i is the actual operation data of the i-th device, β i is the estimated operation data of the i-th device, and χ is the maximum device real operation data among all the device real operation data.

[0024] The present invention also provides a comprehensive evaluation system for the wet regeneration and cleaning technology of waste lead paste, which mainly includes the following modules:

[0025] Preparation module, used to collect process data sets, environmental protection related data sets, equipment operation data sets, cost data sets, and product test data sets;

[0026] The analysis module is used to analyze and process the process data set to obtain the evaluation results on the efficiency of the process, analyze and process the environmental protection related data set to obtain the evaluation results on the implementation effect of the environmental protection measures, analyze and process the cost data set to obtain the evaluation results on the product benefits, and analyze and process the product test data set to obtain the evaluation results on the product quality; at the same time, it is used to generate several inference algorithms and several candidate inference algorithms, determine the final inference algorithm from the several inference algorithms, determine the final candidate inference algorithm from the several candidate inference algorithms, select one of the final inference algorithm and the final candidate inference algorithm for inference processing to obtain the equipment inference operation data at the inference time; and is used to obtain the equipment true value at several past times. actual operating data, and estimated operating data of the device at several past moments, calculating, for each past moment, a difference in device operating data between the actual operating data of the device at the past moment and the estimated operating data of the device at the past moment, and generating a difference estimation algorithm based on the several differences in device operating data; further configured to, when a final estimation algorithm is used for estimation processing, calculate final estimated operating data of the device at the estimation moment based on the estimated operating data of the device at the estimation moment and the difference in the estimated operating data of the device at the estimation moment output by the difference estimation algorithm; when a final candidate estimation algorithm is used for estimation processing, directly use the estimated operating data of the device at the estimation moment as the final estimated operating data of the device, and generate an evaluation result on the stability of the device operation based on the final estimated operating data of the device;

[0027] Comprehensive evaluation module, used to integrate different evaluation results to generate comprehensive evaluation results.

[0028] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0029] In the present invention, first, the evaluation results of the processing efficiency, the evaluation results of the implementation effect of the environmental protection measures, the evaluation results of the product benefits, and the evaluation results of the product quality are obtained; secondly, a final speculation algorithm is determined among several speculation algorithms, and a final candidate speculation algorithm is determined among several candidate speculation algorithms, and one of the final speculation algorithm and the final candidate speculation algorithm is selected for speculation processing to obtain the equipment speculation operation data at the speculation moment; thirdly, for each past moment, the equipment operation data difference between the actual equipment operation data at the past moment and the equipment speculation operation data at the past moment is calculated, and a difference speculation algorithm is generated based on several equipment operation data differences; finally, when the final speculation algorithm is used for speculation processing, the final equipment speculation operation data at the speculation moment is calculated based on the equipment speculation operation data at the speculation moment and the difference between the equipment speculation operation data at the speculation moment; when the final candidate speculation algorithm is used for speculation processing, the equipment speculation operation data at the speculation moment is directly used as the final equipment speculation operation data, and then an evaluation result on the equipment operation stability is generated based on the final equipment speculation operation data, and different evaluation results are integrated to generate a comprehensive evaluation result. The present invention not only provides an intelligent, comprehensive and scientific evaluation method, but also ensures the accuracy of the evaluation results on the equipment operation stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flow chart of the comprehensive evaluation method for the wet regeneration and cleaning technology of waste lead paste of the present invention;

[0031] Figure 2 This is a structural diagram of the comprehensive evaluation system for the wet regeneration and cleaning technology of waste lead paste according to the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0033] It is understood that the terms "first," "second," etc., used herein may be used to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script without departing from the scope of this application.

[0034] refer to Figure 1 The present invention provides a comprehensive evaluation method for the wet regeneration and cleaning technology of waste lead paste, which is mainly achieved by performing the following steps:

[0035] The preparation module collects a process data set, an environmental protection-related data set, an equipment operation data set, a cost data set, and a product test data set, and the analysis module analyzes and processes the process data set to obtain an evaluation result on the process efficiency, analyzes and processes the environmental protection-related data set to obtain an evaluation result on the effectiveness of the environmental protection measures, analyzes and processes the cost data set to obtain an evaluation result on the product benefit, and analyzes and processes the product test data set to obtain an evaluation result on the product quality;

[0036] The analysis module generates a plurality of inference algorithms and a plurality of candidate inference algorithms, determines a final inference algorithm from the plurality of inference algorithms, determines a final candidate inference algorithm from the plurality of candidate inference algorithms, and selects one of the final inference algorithm and the final candidate inference algorithm for inference processing to obtain inference operation data of the device at the inference time;

[0037] The analysis module obtains actual device operation data for a number of past moments and estimated device operation data for a number of past moments, and for each past moment, calculates a difference between the actual device operation data for the past moment and the estimated device operation data for the past moment. The analysis module also generates a difference estimation algorithm based on the differences in the device operation data.

[0038] When the final speculation algorithm is used for speculation processing, the analysis module calculates the final device speculation operating data at the speculation time based on the device speculation operating data at the speculation time and the difference between the device speculation operating data at the speculation time output by the difference speculation algorithm. When the final candidate speculation algorithm is used for speculation processing, the analysis module directly uses the device speculation operating data at the speculation time as the final device speculation operating data. The analysis module then generates an evaluation result on the device operation stability based on the final device speculation operating data. The comprehensive evaluation module integrates different evaluation results to generate a comprehensive evaluation result.

[0039] Specifically, the present invention proposes a scientific, comprehensive and intelligent method for comprehensively evaluating the wet regeneration and cleaning technology of waste lead paste. First, a preparation module collects a process data set, an environmental protection-related data set, an equipment operation data set, a cost data set, and a product test data set, wherein the process data set includes the particle size distribution data of the lead paste crushing stage, the screening pass rate data of the screening stage, and the fineness data after grinding in the grinding stage, etc. The environmental protection-related data set includes the concentration data of various pollutants in the wastewater and the content data of harmful gases in the exhaust gas, etc. The cost data set includes raw material cost data, energy consumption cost data, labor cost data, lead product price data, and lead product demand data, etc. The product test data set includes the purity data of the lead product, the impurity content data of the lead product, and Mechanical performance data, etc. The equipment operation data set includes various operating status data of the equipment, such as temperature data, pressure data, current data, etc. when the equipment is running. On this basis, the analysis module analyzes and processes the process data set to obtain an evaluation result on the processing process efficiency. The analysis module analyzes and processes the environmental protection related data set to obtain an evaluation result on the implementation effect of environmental protection measures. The analysis module analyzes and processes the cost data set to obtain an evaluation result on product benefits. The analysis module analyzes and processes the product test data set to obtain an evaluation result on product quality. The machine learning method in the existing technology can be used to obtain the above-mentioned evaluation results, which will not be repeated here. The following focuses on how to generate an evaluation result on the stability of equipment operation.

[0040] Secondly, the analysis module generates several inference algorithms and several candidate inference algorithms in advance, determines the final inference algorithm among the several inference algorithms, and determines the final candidate inference algorithm among the several candidate inference algorithms. The final inference algorithm and the final candidate inference algorithm are both more suitable algorithms for inference processing. Then, the algorithm that is most suitable for inference processing is selected from the final inference algorithm and the final candidate inference algorithm, thereby obtaining the device inference operation data at the time of inference.

[0041] Again, the analysis module obtains the actual operation data of the equipment at several past moments, as well as the estimated operation data of the equipment at several past moments. The estimated operation data of the equipment here is actually output by the final estimation algorithm. Then, the analysis module calculates the difference between the actual operation data of the equipment at the past moment and the estimated operation data of the equipment at the past moment for each past moment. The analysis module also generates a difference estimation algorithm based on the difference of several equipment operation data. The difference estimation algorithm can be generated through existing technology, which will not be elaborated on.

[0042] Finally, if the final speculation algorithm is used for speculation processing, the analysis module calculates the final device speculation operating data at the speculation moment based on the device speculation operating data at the speculation moment and the difference between the device speculation operating data at the speculation moment output by the difference speculation algorithm. The specific method can be to add the device speculation operating data at the speculation moment and the difference between the device speculation operating data at the speculation moment. If the final candidate speculation algorithm is used for speculation processing, the analysis module directly uses the device speculation operating data at the speculation moment as the final device speculation operating data, and then the analysis module generates an evaluation result on the stability of the device operation based on the final device speculation operating data. For ease of understanding, the present invention takes the speculation of the temperature data of the device in the operating state as an example. When the speculated temperature data of the device exceeds the threshold, it is determined that the device operation is unstable. After that, the comprehensive evaluation module integrates different evaluation results to generate a comprehensive evaluation result. Specifically, the expert evaluation method can be used to integrate different evaluation results, which will not be repeated here.

[0043] The above method comprehensively considers the efficiency of the processing process, the effectiveness of the implementation of environmental protection measures, product benefits, product quality, and equipment operation stability to evaluate the wet regeneration and cleaning technology of waste lead paste, which can ensure the intelligence, scientificity and comprehensiveness of the evaluation method.

[0044] Furthermore, the process of the analysis module generating several inference algorithms includes: first, the analysis module determines several factors that affect the inference results; second, the analysis module generates several factor combinations, each factor combination is composed of different factors, and generates several inference algorithms corresponding to the several factor combinations.

[0045] Specifically, the process of generating several inference algorithms is introduced. Before the introduction, it should be noted that the present invention infers the temperature data of the device in the running state. Of course, the present invention can also infer other data of the device in the running state. In the first step, the analysis module determines several factors that affect the inference result, such as ambient temperature, heat dissipation conditions, workload, material properties, process parameters, etc. In the second step, the analysis module generates several factor combinations, each factor combination is composed of different factors, that is, the total number of different factors in each factor combination is not fixed. For example, a factor combination can be "ambient temperature, heat dissipation conditions, process parameters", and then several inference algorithms corresponding to the several factor combinations are generated respectively, for example, an inference algorithm corresponding to the factor combination "ambient temperature, heat dissipation conditions, process parameters" is generated, that is, the inference algorithm infers the temperature data of the device in the running state based on the ambient temperature data, heat dissipation condition data, and process parameter data.

[0046] Furthermore, the process of the analysis module generating a plurality of candidate inference algorithms includes: the analysis module generates a plurality of candidate inference algorithms corresponding to the plurality of factor combinations according to the plurality of factor combinations that have been generated.

[0047] Specifically, the process of the analysis module generating several candidate inference algorithms is continued to be introduced. The analysis module generates several candidate inference algorithms corresponding to the several factor combinations based on the several factor combinations that have been generated, that is, the several factor combinations generated when generating several inference algorithms. For example, candidate inference algorithms corresponding to the factor combination of "ambient temperature, heat dissipation conditions, and process parameters" are also generated. It should be noted that each inference algorithm always has a corresponding candidate inference algorithm, and the factor combinations corresponding to them are the same.

[0048] Furthermore, the inference algorithm is pre-trained by the analysis module using a training data set, where the training data set is composed of different training data, including actual operation data of the equipment at past moments, and different factor data corresponding to the past moments.

[0049] Specifically, the generation process of the inference algorithm and the candidate inference algorithm has been explained above. It should also be noted that the inference algorithm needs to be pre-trained by the analysis module using a training data set. The inference algorithm can be, for example, a machine learning algorithm, while the candidate inference algorithm does not need to be pre-trained by the analysis module. The candidate inference algorithm can, for example, calculate the product of each factor data and its weight and calculate the sum of all products. The training data includes the actual operation data of the equipment at the past moment, and different factor data corresponding to the past moment. For ease of understanding, for example, a training data is "the actual temperature data at nine o'clock in the morning of the previous day, and the ambient temperature data, heat dissipation condition data, and process parameter data at nine o'clock in the morning of the previous day."

[0050] Furthermore, the analysis module determines a final inference algorithm from among several inference algorithms, and determines a final candidate inference algorithm from among several candidate inference algorithms, including the following steps:

[0051] The analysis module selects a speculation algorithm from a plurality of speculation algorithms, and simultaneously selects a candidate speculation algorithm corresponding to the selected speculation algorithm from a plurality of candidate speculation algorithms;

[0052] The analysis module determines a number of past moments, and causes the selected inference algorithm to perform inference processing a number of times to obtain inferred operating data of the first device at the number of past moments, and causes the selected candidate inference algorithm to perform inference processing a number of times to obtain inferred operating data of the second device at the number of past moments;

[0053] The analysis module obtains actual device operation data for a certain number of past moments, calculates a first accuracy of a selected estimation algorithm based on the actual device operation data for the certain number of past moments and first device estimated operation data for the certain number of past moments using a preset formula, calculates a second accuracy of a selected candidate estimation algorithm based on the actual device operation data for the certain number of past moments and second device estimated operation data for the certain number of past moments using a preset formula, and calculates a comprehensive accuracy based on the first accuracy and the second accuracy;

[0054] The analysis module determines whether there is an inference algorithm that has not been selected. If so, it jumps to the initial step of selecting the inference algorithm and the candidate inference algorithm. If not, the selected inference algorithm and the selected candidate inference algorithm corresponding to the maximum comprehensive accuracy are respectively regarded as the final inference algorithm and the final candidate inference algorithm.

[0055] Specifically, the process of the analysis module determining the final inference algorithm from several inference algorithms and determining the final candidate inference algorithm from several candidate inference algorithms is explained. In the first step, the analysis module selects an inference algorithm from several inference algorithms, and at the same time selects a candidate inference algorithm corresponding to the selected inference algorithm from several candidate inference algorithms. In the second step, the analysis module determines several past moments, and makes the selected inference algorithm perform inference processing several times to obtain the first device inference operation data of several past moments, and at the same time makes the selected candidate inference algorithm perform inference processing several times to obtain the second device inference operation data of several past moments. In the third step, the analysis module also obtains the actual operation data of the equipment at the determined several past moments, and uses a preset formula to calculate and select the actual operation data of the equipment at the several past moments and the first device inference operation data at the several past moments. The first accuracy of the inference algorithm is calculated, and the second accuracy of the selected candidate inference algorithm is calculated using a preset formula based on the actual operation data of the device at several past moments and the inferred operation data of the second device at several past moments. The preset formula will be described in detail below. Then, the comprehensive accuracy is calculated based on the first accuracy and the second accuracy. For example, the product of the first accuracy and its weight and the product of the second accuracy and its weight can be calculated first, and then the sum of the two products can be calculated. In the fourth step, the analysis module determines whether there is an inference algorithm that has not been selected. If so, jump to the initial step of selecting the inference algorithm and the candidate inference algorithm, that is, the first step, to select the inference algorithm and the candidate inference algorithm that have not been selected. If not, among all the comprehensive accuracies, the selected inference algorithm corresponding to the largest comprehensive accuracy and the selected candidate inference algorithm will be respectively regarded as the final inference algorithm and the final candidate inference algorithm.

[0056] Furthermore, the analysis module selects one of the final inference algorithm and the final candidate inference algorithm to perform inference processing to obtain the device inference operation data at the inference time, including the following steps:

[0057] The analysis module obtains the actual operation data of the device at several past moments within a past time period before the current moment, and for each past moment, the analysis module obtains different factor data corresponding to the past moment. The analysis module also performs training processing on the final inference algorithm;

[0058] The analysis module determines a plurality of past moments in a past time period two before the current moment, and causes the final speculation algorithm to perform a plurality of speculation processes to obtain the third device speculated operating data at the plurality of past moments, and causes the final candidate speculation algorithm to also perform a plurality of speculation processes to obtain the fourth device speculated operating data at the plurality of past moments.

[0059] The analysis module obtains actual device operation data at a certain number of past moments, calculates a third accuracy of a final estimation algorithm based on the actual device operation data at the certain number of past moments and estimated third device operation data at the certain number of past moments using a preset formula, and calculates a fourth accuracy of a final candidate estimation algorithm based on the actual device operation data at the certain number of past moments and estimated fourth device operation data at the certain number of past moments using a preset formula;

[0060] The analysis module determines whether the third accuracy is greater than or equal to the fourth accuracy. If yes, it selects the final inference algorithm. If not, it selects the final candidate inference algorithm. The analysis module uses the selected algorithm to perform inference processing to obtain the device inference operation data at the inference time.

[0061] Specifically, the process of selecting one of the final inference algorithm and the final candidate inference algorithm by the analysis module for inference processing to obtain the inferred operation data of the device at the inference time is introduced. Before starting the introduction, it should be noted that a large amount of training data in the training data set used above is generated based on historical equipment of the same type as the device for which temperature data inference is required. The selected inference algorithm and the selected candidate inference algorithm also infer the temperature data of historical equipment of the same type as the device for which temperature data inference is required. When generating the difference inference algorithm above, the actual operation data of the device at the past moment and the inferred operation data of the device at the past moment of the historical equipment of the same type as the device for which temperature data inference is required are also used. However, if the device for which temperature data inference is required is a new model, the algorithm parameters in the final inference algorithm may need to be updated. Since the device operation time is not long enough, the actual operation data of the device is missing. In order to ensure the accuracy of the inference processing in this case, the following method is proposed:

[0062] In the first step, the analysis module obtains the actual device operation data for several past moments within a past time period one preceding the current moment. For each past moment, the analysis module also obtains different factor data corresponding to the past moment. In other words, the analysis module generates a new training dataset based on the actual device operation data for several moments since the device began operating, as well as the different factor data corresponding to each of the past moments. On this basis, the analysis module also performs training processing on the final inference algorithm. In the second step, the analysis module determines several past moments within a past time period two preceding the current moment. It should be noted that past time period two is after past time period one, for example, past time period two begins after past time period one ends, and past time period two is shorter than past time period one. The analysis module then causes the final inference algorithm to perform several inference processes to obtain the third device inferred operation data for several past moments, and causes the final candidate inference algorithm to perform several inference processes to obtain the fourth device inferred operation data for several past moments. In the third step, the analysis module obtains actual device operating data from several past moments. Based on this actual device operating data and estimated third device operating data from several past moments, the analysis module calculates a third accuracy for the final estimation algorithm using a preset formula. Based on this actual device operating data and estimated fourth device operating data from several past moments, the analysis module calculates a fourth accuracy for the final candidate estimation algorithm using a preset formula (described in detail below). In the fourth step, the analysis module determines whether the third accuracy is greater than or equal to the fourth accuracy. If so, the analysis module selects the final estimation algorithm. If not, the analysis module selects the final candidate estimation algorithm. The analysis module then performs estimation processing using the selected algorithm to obtain estimated device operating data at the estimation time.

[0063] Furthermore, the preset formula is Among them, δ is the accuracy, k is the total number of real operation data of the equipment, α i is the actual operation data of the i-th device, β i is the estimated operation data of the i-th device, and χ is the maximum device real operation data among all the device real operation data.

[0064] Specifically, the preset formula The accuracy of the calculation algorithm. The greater the accuracy, the better the algorithm performs inference processing. δ is the accuracy, k is the total number of real running data of the device, which is the same as the total number of inferred running data of the device, and α i is the actual operation data of the i-th device, β i is the estimated operation data of the i-th device, and χ is the maximum device real operation data among all the device real operation data.

[0065] References Figure 2 As shown, the present invention also provides a comprehensive evaluation system for the wet regeneration and cleaning technology of waste lead paste, including a preparation module, an analysis module, and a comprehensive evaluation module, which are used to implement the comprehensive evaluation method for the wet regeneration and cleaning technology of waste lead paste as described above. The functions of each module are as follows:

[0066] Preparation module, used to collect process data sets, environmental protection related data sets, equipment operation data sets, cost data sets, and product test data sets;

[0067] The analysis module is used to analyze and process the process data set to obtain the evaluation results on the efficiency of the process, analyze and process the environmental protection related data set to obtain the evaluation results on the implementation effect of the environmental protection measures, analyze and process the cost data set to obtain the evaluation results on the product benefits, and analyze and process the product test data set to obtain the evaluation results on the product quality; at the same time, it is used to generate several inference algorithms and several candidate inference algorithms, determine the final inference algorithm from the several inference algorithms, determine the final candidate inference algorithm from the several candidate inference algorithms, select one of the final inference algorithm and the final candidate inference algorithm for inference processing to obtain the equipment inference operation data at the inference time; and is used to obtain the equipment true value at several past times. actual operating data, and estimated operating data of the device at several past moments, calculating, for each past moment, a difference in device operating data between the actual operating data of the device at the past moment and the estimated operating data of the device at the past moment, and generating a difference estimation algorithm based on the several differences in device operating data; further configured to, when a final estimation algorithm is used for estimation processing, calculate final estimated operating data of the device at the estimation moment based on the estimated operating data of the device at the estimation moment and the difference in the estimated operating data of the device at the estimation moment output by the difference estimation algorithm; when a final candidate estimation algorithm is used for estimation processing, directly use the estimated operating data of the device at the estimation moment as the final estimated operating data of the device, and generate an evaluation result on the stability of the device operation based on the final estimated operating data of the device;

[0068] Comprehensive evaluation module, used to integrate different evaluation results to generate comprehensive evaluation results.

[0069] It should be understood that, although the various steps in the flow chart of each embodiment of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0070] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program. The above-mentioned program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0071] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0072] The above embodiments merely illustrate several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the scope of the present invention, all of which fall within the scope of protection of the present invention.

[0073] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A comprehensive evaluation method for wet regeneration and cleaning technology of waste lead paste, characterized in that: The steps include: The preparation module collects a process data set, an environmental protection-related data set, an equipment operation data set, a cost data set, and a product test data set, and the analysis module analyzes and processes the process data set to obtain an evaluation result on the process efficiency, analyzes and processes the environmental protection-related data set to obtain an evaluation result on the effectiveness of the environmental protection measures, analyzes and processes the cost data set to obtain an evaluation result on the product benefit, and analyzes and processes the product test data set to obtain an evaluation result on the product quality; The analysis module generates a plurality of inference algorithms and a plurality of candidate inference algorithms, determines a final inference algorithm from the plurality of inference algorithms, determines a final candidate inference algorithm from the plurality of candidate inference algorithms, and selects one of the final inference algorithm and the final candidate inference algorithm for inference processing to obtain inference operation data of the device at the inference time; The analysis module obtains actual device operation data for a number of past moments and estimated device operation data for a number of past moments, and for each past moment, calculates a device operation data difference between the actual device operation data for the past moment and the estimated device operation data for the past moment. The analysis module also generates a difference estimation algorithm based on the device operation data differences. When the final speculation algorithm is used for speculation processing, the analysis module calculates the final device speculation operating data at the speculation moment based on the device speculation operating data at the speculation moment and the difference between the device speculation operating data at the speculation moment output by the difference speculation algorithm. When the final candidate speculation algorithm is used for speculation processing, the analysis module directly uses the device speculation operating data at the speculation moment as the final device speculation operating data. Subsequently, the analysis module generates an evaluation result on the device operation stability based on the final device speculation operating data, and the comprehensive evaluation module integrates different evaluation results to generate a comprehensive evaluation result.

2. The method according to claim 1, characterized in that The process of the analysis module generating several inference algorithms includes: first, the analysis module determines several factors that affect the inference results; second, the analysis module generates several factor combinations, each factor combination is composed of different factors, and generates several inference algorithms corresponding to the several factor combinations.

3. The method according to claim 2, characterized in that The process of the analysis module generating a plurality of candidate inference algorithms includes: the analysis module generating a plurality of candidate inference algorithms corresponding to the plurality of factor combinations according to the plurality of factor combinations that have been generated.

4. The method according to claim 3, characterized in that The inference algorithm is pre-trained by the analysis module using a training data set, where the training data set is composed of different training data, including actual equipment operation data at past moments and different factor data corresponding to the past moments.

5. The method according to claim 4, characterized in that The analysis module determines a final inference algorithm from among several inference algorithms, and determines a final candidate inference algorithm from among several candidate inference algorithms, including the following steps: The analysis module selects a speculation algorithm from a plurality of speculation algorithms, and simultaneously selects a candidate speculation algorithm corresponding to the selected speculation algorithm from a plurality of candidate speculation algorithms; The analysis module determines a number of past moments, and causes the selected inference algorithm to perform inference processing a number of times to obtain inferred operating data of the first device at the number of past moments, and causes the selected candidate inference algorithm to perform inference processing a number of times to obtain inferred operating data of the second device at the number of past moments; The analysis module obtains actual device operation data at a certain number of past moments, calculates a first accuracy of a selected estimation algorithm based on the actual device operation data at the certain number of past moments and first device estimated operation data at the certain number of past moments using a preset formula, calculates a second accuracy of a selected candidate estimation algorithm based on the actual device operation data at the certain number of past moments and second device estimated operation data at the certain number of past moments using a preset formula, and calculates a comprehensive accuracy based on the first accuracy and the second accuracy; The analysis module determines whether there is an unselected inference algorithm. If so, it jumps to the initial step of selecting the inference algorithm and the candidate inference algorithm. If not, the selected inference algorithm and the selected candidate inference algorithm corresponding to the maximum comprehensive accuracy are respectively regarded as the final inference algorithm and the final candidate inference algorithm.

6. The method according to claim 5, characterized in that The analysis module selects one of the final inference algorithm and the final candidate inference algorithm to perform inference processing to obtain the device inference operation data at the inference time, including the following steps: The analysis module obtains actual operation data of the device at several past moments within a past time period before the current moment, and for each past moment, the analysis module obtains different factor data corresponding to the past moment, and the analysis module also performs training processing on the final inference algorithm; The analysis module determines a plurality of past moments in a past time period two before the current moment, and causes the final speculation algorithm to perform a plurality of speculation processes to obtain the third device speculated operating data at the plurality of past moments, and causes the final candidate speculation algorithm to also perform a plurality of speculation processes to obtain the fourth device speculated operating data at the plurality of past moments; The analysis module obtains the actual operation data of the device at the determined several past moments, calculates a third accuracy of the final estimation algorithm based on the actual operation data of the device at the determined several past moments and the estimated operation data of the third device at the determined several past moments using a preset formula, and calculates a fourth accuracy of the final candidate estimation algorithm based on the actual operation data of the device at the determined several past moments and the estimated operation data of the fourth device at the determined several past moments using a preset formula; The analysis module determines whether the third accuracy is greater than or equal to the fourth accuracy. If yes, the analysis module selects a final inference algorithm; if not, the analysis module selects a final candidate inference algorithm. The analysis module uses the selected algorithm to perform inference processing to obtain device inference operation data at the inference time.

7. The method according to claim 6, characterized in that The default formula is Among them, δ is the accuracy, k is the total number of real operation data of the equipment, α i is the actual operation data of the i-th device, β i is the estimated operation data of the i-th device, and χ is the maximum device real operation data among all the device real operation data.

8. A comprehensive evaluation system for wet regeneration and cleaning technology of waste lead paste, used to implement the method according to any one of claims 1 to 7, characterized in that: Includes the following modules: Preparation module, used to collect process data sets, environmental protection related data sets, equipment operation data sets, cost data sets, and product test data sets; an analysis module for analyzing and processing a process data set to obtain an evaluation result on the efficiency of the process, analyzing and processing an environmental-related data set to obtain an evaluation result on the effectiveness of the implementation of environmental protection measures, analyzing and processing a cost data set to obtain an evaluation result on the product benefit, and analyzing and processing a product test data set to obtain an evaluation result on the product quality; and for generating a plurality of inference algorithms and a plurality of candidate inference algorithms, determining a final inference algorithm from among the plurality of inference algorithms, determining a final candidate inference algorithm from among the plurality of candidate inference algorithms, selecting one of the final inference algorithm and the final candidate inference algorithm for inference processing to obtain the inferred operation data of the device at the inference moment; and for obtaining the actual operation data of the device at a plurality of past moments, and the inferred operation data of the device at a plurality of past moments, calculating, for each past moment, a difference in the operation data between the actual operation data of the device at the past moment and the inferred operation data of the device at the past moment, and generating a difference inference algorithm based on the plurality of device operation data differences; Further configured to, when a final speculation algorithm is used for speculation processing, calculate final speculation data for the device at the speculation time based on the difference between the speculated device operating data at the speculation time and the speculated device operating data at the speculation time output by the difference speculation algorithm; and, when a final candidate speculation algorithm is used for speculation processing, directly use the speculated device operating data at the speculation time as the final speculated device operating data, and generate an evaluation result regarding device operation stability based on the final speculated device operating data; Comprehensive evaluation module, used to integrate different evaluation results to generate comprehensive evaluation results.

Images