Optimization method and system for technological parameters of manure and straw turning and water-logged composting

Abstract

The invention belongs to the technical field of manure treatment, and provides an optimization method and system for manure straw pile-turning water-logged compost process parameters. The method comprises the following steps: collecting and constructing a piece of training data according to each piece of real pile-turning water-logged compost data; according to compost evaluation data, dividing training data corresponding to the pile-turning compost data into a plurality of groups, and sorting the training data in each group according to a first compost volume; according to the sorting sequence, a plurality of training data are selected from each group to form a training data set, a plurality of training data sets are obtained, and the plurality of training data in each training data set are used for training the process parameter optimization model until the training reaches the standard; and determining second manure straw data and a second compost volume related to the composting, processing the second manure straw data and the second compost volume by using the trained process parameter optimization model, and predicting to obtain a second process parameter. According to the invention, optimization of composting process parameters can be realized.

Description

Technical Field

[0001] This invention relates to the field of manure treatment technology, and more specifically, to a method and system for optimizing process parameters for manure and straw composting. Background Technology

[0002] With the large-scale generation of agricultural waste (such as livestock and poultry manure and crop straw), how to efficiently and environmentally treat these organic wastes has become one of the key issues for sustainable agricultural development. Traditional incineration or landfill methods easily cause environmental pollution and resource waste, while composting manure and straw is an economically feasible organic waste treatment technology that can transform organic matter into stable humus through microbial fermentation, realizing the resource utilization of waste and improving soil fertility at the same time.

[0003] However, in actual composting processes, composting efficiency and quality are affected by various process parameters, such as temperature, carbon-to-nitrogen ratio, pH, additives, moisture content, ventilation method, and ventilation volume. Improper parameter control can lead to problems such as prolonged composting cycles, incomplete decomposition, nutrient loss, or odor emissions. Currently, domestic and international research mainly focuses on optimizing composting processes through the synergistic effects of temperature, oxygen, and microorganisms, but the following technical bottlenecks still exist: Inaccurate turning strategies: Over-turning leads to heat loss, while insufficient turning causes localized anaerobic conditions, affecting the uniformity of composting. Low ventilation and oxygen supply efficiency: Natural ventilation composting has a long cycle, while forced ventilation has high energy consumption, requiring a balance between oxygen supply and temperature control. Incomplete composting evaluation system: A single indicator (such as temperature or C / N ratio) cannot comprehensively reflect compost quality; multi-dimensional dynamic monitoring is necessary. Therefore, optimizing the turning and composting process parameters is crucial for improving composting efficiency, reducing energy consumption, and ensuring composting quality.

[0004] In summary, obtaining optimized composting process parameters to achieve high-level composting results is a pressing technical problem that needs to be solved. Summary of the Invention

[0005] To solve at least one of the above-mentioned technical problems, the present invention specifically provides a method, system, electronic device, computer storage medium, and computer program product for optimizing process parameters of manure and straw composting.

[0006] This invention provides a method for optimizing process parameters of manure and straw composting, comprising the following steps: Collect several real composting data, each of which includes first manure and straw data, first process parameters, composting evaluation data, and first compost volume; A training data point is constructed based on the first manure and straw data, the first process parameters, the composting evaluation data, and the first compost volume in each of the aforementioned composting data points; wherein, the composting evaluation data is used to construct the labels for the training data. The training data corresponding to each composting data is divided into multiple groups based on the composting evaluation data, and the training data in each group is sorted according to the first compost volume. According to the sorting order, several training data are selected from each group to form a training data group, and multiple training data groups are obtained. The process parameter optimization model is trained using several training data in each training data group until the training reaches the target. The second manure and straw data and the second compost volume involved in this composting are determined. The trained process parameter optimization model is used to process the second manure and straw data and the second compost volume to predict the second process parameters.

[0007] The present invention also provides an optimization system for the process parameters of composting manure and straw, the system comprising a collection unit, a training unit, and an optimization unit; The collection unit is used to collect several real composting data, each of which includes first manure and straw data, first process parameters, composting evaluation data, and first compost volume. The training unit is used to: construct a training data based on the first manure and straw data, the first process parameters, the composting evaluation data, and the first compost volume in each piece of composting data; wherein, the composting evaluation data is used to construct the label of the training data; The training data corresponding to each composting data is divided into multiple groups based on the composting evaluation data, and the training data in each group is sorted according to the first compost volume. According to the sorting order, several training data are selected from each group to form a training data group, and multiple training data groups are obtained. The process parameter optimization model is trained using several training data in each training data group until the training reaches the target. The optimization unit is used to determine the second manure and straw data and the second compost volume involved in this composting, and to process the second manure and straw data and the second compost volume using the trained process parameter optimization model to predict the second process parameters.

[0008] The present invention also provides an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calling the executable program code stored in the memory to perform the method as described in any of the preceding claims.

[0009] The present invention also provides a computer storage medium storing a computer program that, when executed by a processor, performs the method described in any of the preceding claims.

[0010] The present invention also provides a computer program product for being invoked and executed by a processor of an electronic device to implement the method described in any of the preceding claims.

[0011] The beneficial effects of this invention are as follows: The present invention uses training data constructed based on real composting data to train the process parameter optimization model. The decision strategy of the training data group improves the diversity of the training data, which significantly improves the training effect and efficiency of the process parameter optimization model. This is conducive to obtaining more efficient process parameters and thus improving the effect of composting manure and straw. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a schematic flowchart of a method for optimizing process parameters of manure and straw composting, as disclosed in an embodiment of the present invention.

[0014] Figure 2 This is a schematic diagram of grouping training data and forming training data groups as disclosed in an embodiment of the present invention.

[0015] Figure 3 This is a schematic diagram of the structure of an optimization system for manure and straw composting process parameters disclosed in an embodiment of the present invention. Detailed Implementation

[0016] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0017] See Figure 1 The flowchart shown illustrates a method for optimizing process parameters in manure and straw composting, comprising the following steps: S10, collect several real composting data, each of which includes first manure and straw data, first process parameters, composting evaluation data, and first compost volume.

[0018] This invention pre-constructs a process parameter optimization model to predict the optimal composting process parameters. However, this optimization model requires pre-training to be capable of predicting these parameters. To address this, this invention collects a suitable number (e.g., 200 data points) of real composting data, which includes at least the following: Firstly, data on manure and straw: Because different types of manure and straw (such as the compositional differences of different livestock and poultry manure and the characteristics of different crop straw) can affect the composting process and results, it is necessary to collect relevant data. This data includes information such as the type and composition content of manure and straw.

[0019] The first process parameters, such as temperature, carbon-nitrogen ratio, pH value, auxiliary materials, moisture content, ventilation method, and ventilation volume, will affect the efficiency and quality of composting in actual composting.

[0020] Compost evaluation data: Compost evaluation data can be scores or grades used to characterize its quality. These scores can be generated manually or automatically by an automated evaluation model; this invention does not impose specific limitations on this. Compost evaluation data is derived from a comprehensive evaluation of at least physical, chemical, maturity, biological and hygiene indicators, as well as other indicators. For example, it may be obtained by weighted calculation of multiple sub-scores. The explanations of each indicator are as follows: Physical indicators: Well-rotted compost should have no foul odor, be dark brown, loose, and free of visible raw materials (such as straw, manure residue).

[0021] Chemical indicators: The organic matter degradation rate after composting should be reduced by 30%-50%, assessed by the reduction in total organic carbon (TOC) or volatile solids (VS). The carbon-to-nitrogen ratio after composting should be less than 20:1, close to the soil level of 10-15:1. The pH range should be 6.5-8.5. The electrical conductivity (EC) of the composted manure should be <4 mS / cm; excessively high EC may cause salinization.

[0022] Indicators of humification maturity: A humic acid (HA) / fulvic acid (FA) ratio >1.5 indicates good humification.

[0023] Biological and health indicators Microbial community: Microbial activity is assessed by PCR or enzyme activity (such as dehydrogenase, urease).

[0024] Pathogens and parasite eggs: must meet organic fertilizer standards, such as a mortality rate of ≥95% for roundworm eggs and a coliform count of ≤0.01g⁻¹.

[0025] Other indicators: The concentration of volatile organic compounds (VOCs, such as ammonia and hydrogen sulfide) should be significantly reduced after decomposition. The heavy metal content must meet the standards for organic fertilizers, for example, Cd≤3mg / kg, Pb≤100mg / kg.

[0026] First, compost volume: Different compost volumes may result in different internal environmental conditions (such as oxygen distribution and temperature diffusion), which will affect the composting process. Therefore, this is also one of the data that needs to be collected.

[0027] S20, a training data is generated based on the first manure and straw data, the first process parameters, the composting evaluation data, and the first compost volume in each of the composting data; wherein, the composting evaluation data is used to construct the label of the training data.

[0028] After collecting a sufficient amount of composting data, the composting evaluation data is used as data labels, while the first manure and straw data, the first process parameters, and the first compost volume are used as data entities. The integration of data entities and data labels yields a training data set.

[0029] Understandably, the initial data on manure and straw, the initial process parameters, and the initial compost volume serve as input features, reflecting the initial conditions and process control factors of composting. The compost evaluation data, on the other hand, provides a comprehensive assessment of the composting results. The model's training objective is to learn the relationship between input features and labels, thereby predicting appropriate process parameters based on new inputs to achieve better composting results.

[0030] S30, the training data corresponding to each of the composting data is divided into multiple groups according to the composting evaluation data, and the training data in each group is sorted according to the first compost volume.

[0031] Different composting evaluation results (such as different degrees of maturity, quality grades, etc.) represent different composting effects. This invention sets up a grouping system based on the composting evaluation data, for example, dividing the data into three groups: low, medium, and high. Each group contains a corresponding number of training data points. Figure 2 As shown. Different training strategies can be formulated subsequently when training the process parameter optimization model to improve the model's adaptability and accuracy. After grouping, the training data in each group are sorted within the group based on the first compost volume.

[0032] This completes the preparation of the training data.

[0033] S40, according to the sorting order, select several training data from each group to form a training data group, obtain multiple training data groups, and use several training data from each training data group to train the process parameter optimization model until the training reaches the target.

[0034] Next, as Figure 2 As shown, several training data points at the top of the sorted list are selected from each group, and these selected training data points are grouped into a single training data group. After multiple selections, multiple training data groups are obtained.

[0035] This setup allows the training data in each training data set to have more significant diversity, which is beneficial for improving the training effect and efficiency of the process parameter optimization model, enabling the model to learn more comprehensive composting rules.

[0036] Using training data sets as units, the process parameter optimization model is trained using the training data contained therein. The model parameters are continuously adjusted so that the model can accurately predict appropriate process parameters based on the input data such as manure and straw, compost volume, etc., until the model training reaches the preset standard (such as accuracy, error and other indicators meet the requirements). At this point, the model has a certain predictive ability.

[0037] S50, determine the second manure and straw data and the second compost volume involved in this composting, and use the trained process parameter optimization model to process the second manure and straw data and the second compost volume to predict the second process parameters.

[0038] In practical applications, once the second set of manure and straw data (specific types and composition of manure and straw, etc.) and the second compost volume are determined, this data is input into a pre-trained process parameter optimization model. The process parameter optimization model can then predict suitable second process parameters for this composting operation based on the relationship between the previously learned input features and the composting results. This optimizes the composting process, improves composting efficiency and quality, and addresses some technical problems currently existing in the composting process.

[0039] It is understandable that the first process parameter and the second process parameter should contain the same parameter content.

[0040] The present invention uses training data constructed based on real composting data to train the process parameter optimization model. The decision strategy of the training data group improves the diversity of the training data, which significantly improves the training effect and efficiency of the process parameter optimization model. This is conducive to obtaining better process parameters and thus improving the effect of composting manure and straw.

[0041] Further, the step of dividing the training data corresponding to each of the composting data into multiple groups based on the composting evaluation data, and sorting the training data in each group according to the first compost volume, includes: The training data corresponding to each of the aforementioned composting data are sorted from high to low according to the composting evaluation data, and the sorted training data are divided into corresponding groups based on a preset division ratio. Calculate the absolute value of the volume difference between the first compost volume and the second compost volume, and determine the sorting number of each training data in the corresponding group based on the absolute value; wherein, the smaller the absolute value, the smaller the corresponding sorting number, that is, the earlier it is.

[0042] In this example, sorting the training data from highest to lowest based on compost evaluation data ensures that training data corresponding to high-quality compost is prioritized, while data of lower quality is placed later. This allows for the extraction of higher-quality training data first when assembling subsequent training datasets. It's understandable that training the process parameter optimization model may not require all training data; that is, the model may converge after training with a subset of the data. Therefore, prioritizing higher-quality training data is practically meaningful.

[0043] The partitioning ratios are pre-defined to divide the sorted training data into different groups. For example, the first 30% of the training data might be assigned to a high-quality group, the next 40% to a medium-quality group, and the last 30% to a low-quality group. This is done to allow for different training strategies for different quality levels of composting when training the process parameter optimization model, enabling the model to learn the characteristics of the process parameters corresponding to different quality levels of composting and improving the model's adaptability and accuracy.

[0044] The process parameter optimization model in this invention can be specifically applied to a single batch of composting, where the compost volume within the same batch is identical. Therefore, this invention prioritizes training data with volumes closer to the second batch of compost within the group (e.g., [missing information]). Figure 2 The training data m1, n1, t1 can be used to increase the probability of this type of training data being selected. The process parameter optimization model trained with training data under more "similar" conditions will have a higher probability of predicting a better second process parameter corresponding to this batch of composting.

[0045] It should be noted that when performing each batch of composting operations, the sorting method of the training data within the group can be redefined as described above, thereby retraining or re-training the process parameter optimization model. However, the training data used in this case may not change significantly. This ensures that the process parameter optimization model can always adapt to the composting operations with different compost volumes in the current batch.

[0046] Furthermore, before deriving a training data point based on the first manure and straw data, the first process parameters, the composting evaluation data, and the first compost volume structure in each of the composting data points, the method further includes: Obtain the data source information for each of the aforementioned composting data, evaluate the process compliance of the data source information, and obtain the process compliance value; If the process compliance value is lower than the preset value and the composting evaluation data of the turning and composting data is of medium level or above, then an adjustment factor is determined based on the difference between the process compliance value and the preset value, and the composting evaluation data is optimized using the adjustment factor.

[0047] In this example, to ensure the training effect of the process parameter optimization model, it is necessary to obtain as much composting data as possible to construct training data. However, in reality, it is very difficult to obtain composting data, so it is necessary to make the best use of the available composting data.

[0048] The data on composting and turning comes from different sources, namely, composting plants of different sizes and in different regions. These plants vary in their strict adherence to the pre-defined composting and turning process parameters. For example, some small-scale composting plants may have insufficient ventilation time for composting, resulting in ventilation volumes that are lower than the standard values.

[0049] In view of the unavoidable realities mentioned above, the present invention first identifies the data source information of each composting plant (i.e., the composting plant itself), and then obtains relevant information about these composting plants, such as obtaining actual composting operation data through on-site visits, personnel interviews, and review of work records. Based on this relevant information, the process compliance of the composting plant is evaluated, and the degree to which the composting plant strictly adheres to the composting process standards in actual operation can be determined.

[0050] Next, select those composting data whose process compliance values ​​are lower than the preset values ​​and whose composting evaluation data in the turning and composting data is of medium or higher level (e.g., a score of 1-10, with 6 or higher being medium or higher level). Optimize and adjust the composting evaluation data in these turning and composting data. Specifically, calculate the difference between the process compliance value and the preset value, and then match the difference to obtain the corresponding adjustment factor. The adjustment factor is negatively correlated with the difference; for example, the adjustment factor is 1.2 or 1.5. In this way, the adjustment factor can appropriately increase the composting evaluation data.

[0051] The reason for this design is that for cases where the process compliance value is not high enough, but the actual effect of composting is at a medium to high level, it indirectly indicates that the process parameters are more likely to be reasonable. In other words, superior process parameters can overcome the adverse effects of a low process compliance value. In this case, this invention uses an adjustment factor to appropriately increase the compost evaluation data of the composting data (to achieve a more objective evaluation of the corresponding process parameters). This higher-evaluation data label guides the training direction of the process parameter optimization model, thereby improving the training effect of the model. Simultaneously, this also allows the training data corresponding to the composting data to be assigned to a higher group, increasing its probability of being selected for training. However, the adjustment factor should not be too high, otherwise the adjusted compost evaluation data may deviate excessively from the actual situation, leading to a deterioration in the model's training effect.

[0052] It should be noted that the above methods are not suitable for optimizing data with a composting evaluation level of less than medium. This is because the reasons for poor composting results may be complex, and the correlation between these results and the suitability of process parameters is unclear. Blindly increasing these parameters may lead to deviations in the training direction of the process parameter optimization model.

[0053] Further, according to the sorting order, several training data points are selected from each of the aforementioned groups to form a training data group, including: In the first stage of training the process parameter optimization model, a number of training data are selected from each of the groups according to the first selection rule to form a training data group; wherein, the first selection rule refers to selecting training data accounting for a first proportion, a second proportion, and a third proportion of the total number from the groups of high-level, medium-level, and low-level compost evaluation data, respectively, and the first proportion is higher than the second proportion, and the second proportion is higher than the third proportion. In the second stage of training the process parameter optimization model, several training data are selected from each of the groups according to the second selection rule to form a training data group; wherein, the second selection rule refers to selecting training data accounting for the fourth, fifth and sixth proportions of the total from the groups of high-level, medium-level and low-level composting evaluation data, respectively, and the fourth proportion is higher than the fifth proportion, the fifth proportion is higher than the sixth proportion, and the fourth proportion is lower than the first proportion.

[0054] In this example, when training the process parameter optimization model, the training process is divided into different stages, and different selection rules are used to select training data. In the early stages of training, more training data is selected from the group with higher composting evaluation data; in the later stages of training, the proportion of training data selected from the group with lower composting evaluation data is increased. Specifically: Assuming the first proportion is 50%, the second is 30%, and the third is 20%, then during the first stage of training, a larger amount of training data will be selected from the higher-level group, and a relatively smaller amount from the lower-level group. In the early stages of training, the process parameter optimization model has limited understanding of the relationship between composting process parameters and composting effects. The training data corresponding to higher-level composting reflects relatively better combinations of composting process parameters and better composting results. Therefore, this invention sets the process parameter optimization model to learn more from these higher-level data in the early stages of training. This helps the model to more quickly capture the process parameter features that produce better composting effects, thereby rapidly establishing a preliminary and relatively reasonable model framework, laying the foundation for further model optimization.

[0055] Assuming the fourth proportion is 30%, the fifth proportion is 35%, and the sixth proportion is 35%, the proportion of data selected from the higher-level groups decreases in the second stage compared to the first stage, while the proportion of data selected from the lower-level groups increases. After the first stage of training, the process parameter optimization model has gained some understanding of better composting process parameters. Therefore, this invention increases the proportion of training data selected from the lower-level groups in the second stage, allowing the model to access more data from less-than-ideal composting scenarios, learn the characteristics of process parameters that lead to poor composting results, and further optimize the model. This enables it to more comprehensively understand the impact of different process parameters on composting results, improves the model's generalization ability and accuracy, and allows for more precise prediction and optimization of process parameters for various composting situations.

[0056] It should be noted that the total number of training data points in the training data set can be fixed, such as 20 or 30. Of course, this total number can also be dynamic, for example, 20 data points in the first stage and 15 data points in the second stage. This invention does not impose specific limitations on this.

[0057] Furthermore, the division ratio is determined in the following manner: Acquire several training record data for the process parameter optimization model. The training data includes model test indicators and historical division ratios. Filter the training record data with the best model test indicators and fuse them according to the historical division ratios corresponding to these training record data to obtain the division ratio used for this training.

[0058] In this example, as explained above, the process parameter optimization model can be a one-time event, meaning that the model is trained once for each batch of composting operations. The relevant data from each training session is recorded. This training data contains two key pieces of information: one is the model testing metrics, which reflect the model's performance after a particular training session, such as accuracy, error rate, and recall. These metrics can measure how well the process parameter optimization model performs in handling composting process parameter prediction tasks. The other is the historical partitioning ratio used, which is the ratio used in previous training processes when dividing the training data into different groups according to the composting evaluation data. For example, there might have been instances where the training data was divided into high, medium, and low levels at ratios of 30%, 40%, and 30%, respectively.

[0059] From the numerous training records acquired, a select few with the best model performance metrics were chosen. "Best" here means that, among all training records, these data exhibit the best overall performance in terms of model performance metrics under certain standards, such as highest accuracy or lowest error. Based on the historical partitioning ratios corresponding to these selected training records, a fusion method (e.g., a weighted average of these historical partitioning ratios, with weights determined by the relative merits of the corresponding model performance metrics—higher weights for better metrics; or other reasonable fusion methods) was used to derive the partitioning ratios for this training. This resulting partitioning ratio is based on a comprehensive analysis of historical partitioning that previously optimized model performance. Therefore, it is highly likely that this will also help the model achieve better training results in this training, enabling the model to more effectively learn the process parameter characteristics corresponding to different composting evaluation levels, thereby improving the accuracy and reliability of the model's prediction of composting process parameters and optimizing the composting process.

[0060] Furthermore, this invention compares the training data grouping method described above with conventional methods in terms of training effectiveness. The comparison metrics include training cycle, accuracy, and generalization ability. The conventional method refers to not grouping all training data, but rather randomly selecting training data for training. When the required proportion or quantity of data is reached, a test set is used to test the model to determine whether to continue training.

[0061] Through actual testing, the training cycle of the present invention is shortened by 20%-30% compared with the conventional method; the overall prediction accuracy of the model after training reaches 85%-90%, while the overall prediction accuracy of the conventional method is only 60%-70%, which is significantly effective.

[0062] like Figure 3 The schematic diagram shown in the figure also discloses an optimization system for the process parameters of composting manure and straw, the system including a collection unit, a training unit, and an optimization unit. The collection unit is used to collect several real composting data, each of which includes first manure and straw data, first process parameters, composting evaluation data, and first compost volume. The training unit is used to: construct a training data based on the first manure and straw data, the first process parameters, the composting evaluation data, and the first compost volume in each piece of composting data; wherein, the composting evaluation data is used to construct the label of the training data; The training data corresponding to each composting data is divided into multiple groups based on the composting evaluation data, and the training data in each group is sorted according to the first compost volume. According to the sorting order, several training data are selected from each group to form a training data group, and multiple training data groups are obtained. The process parameter optimization model is trained using several training data in each training data group until the training reaches the target. The optimization unit is used to determine the second manure and straw data and the second compost volume involved in this composting, and to process the second manure and straw data and the second compost volume using the trained process parameter optimization model to predict the second process parameters.

[0063] Furthermore, the training unit is used for: The training data corresponding to each of the aforementioned composting data are sorted from high to low according to the composting evaluation data, and the sorted training data are divided into corresponding groups based on a preset division ratio. Calculate the absolute value of the volume difference between the first compost volume and the second compost volume, and determine the sorting number of each training data in the corresponding group based on the absolute value; wherein, the smaller the absolute value, the smaller the corresponding sorting number, that is, the earlier it is.

[0064] Furthermore, the training unit is also used for: Obtain the data source information for each of the aforementioned composting data, evaluate the process compliance of the data source information, and obtain the process compliance value; If the process compliance value is lower than the preset value and the composting evaluation data of the turning and composting data is of medium level or above, then an adjustment factor is determined based on the difference between the process compliance value and the preset value, and the composting evaluation data is optimized using the adjustment factor.

[0065] Furthermore, the training unit is used for: In the first stage of training the process parameter optimization model, a number of training data are selected from each of the groups according to the first selection rule to form a training data group; wherein, the first selection rule refers to selecting training data accounting for a first proportion, a second proportion, and a third proportion of the total number from the groups of high-level, medium-level, and low-level compost evaluation data, respectively, and the first proportion is higher than the second proportion, and the second proportion is higher than the third proportion. In the second stage of training the process parameter optimization model, several training data are selected from each of the groups according to the second selection rule to form a training data group; wherein, the second selection rule refers to selecting training data accounting for the fourth, fifth and sixth proportions of the total from the groups of high-level, medium-level and low-level composting evaluation data, respectively, and the fourth proportion is higher than the fifth proportion, the fifth proportion is higher than the sixth proportion, and the fourth proportion is lower than the first proportion.

[0066] Furthermore, the division ratio is determined in the following manner: Acquire several training record data for the process parameter optimization model. The training data includes model test indicators and historical division ratios. Filter the training record data with the best model test indicators and fuse them according to the historical division ratios corresponding to these training record data to obtain the division ratio used for this training.

[0067] This invention also discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calling the executable program code stored in the memory to execute the method described in the foregoing embodiments.

[0068] This invention also discloses a computer storage medium storing a computer program, which is executed by a processor to perform the methods described in the foregoing embodiments.

[0069] This invention also discloses a computer program product for being called and run by a processor of an electronic device to implement the method described in any of the preceding embodiments.

[0070] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to the processor or controller of a general-purpose computer, special-purpose computer, or other programmable load balancing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0071] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0072] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for optimizing process parameters for composting of manure and straw, characterized in that: The steps include: Collecting a number of real compost turning and composting data, each of which includes first manure and straw data, first process parameters, composting evaluation data and first composting volume; A training data is obtained according to the first manure straw data, the first process parameter, the compost evaluation data and the first compost volume in each piece of the compost turning and composting data; wherein the compost evaluation data is used to construct a label of the training data; Dividing the training data corresponding to each of the compost turning and fermentation data into a plurality of groups according to the compost evaluation data, and sorting the training data in each group according to the first compost volume; According to the sorting order, a number of the training data are respectively selected from each of the groups to form a training data group, and a plurality of the training data groups are obtained, and a process parameter optimization model is trained using a number of the training data in each of the training data groups until the training meets the standard; The second manure straw data and the second compost volume involved in this composting are determined, and the second manure straw data and the second compost volume are processed using the trained process parameter optimization model to predict the second process parameters.

2. The method for optimizing process parameters for composting of manure and straw according to claim 1, characterized in that: The step of dividing the training data corresponding to each of the compost turning and fermentation data into a plurality of groups according to the compost evaluation data, and sorting the training data in each group according to the first compost volume includes: The training data corresponding to each of the compost turning and composting data are sorted from high to low according to the compost evaluation data, and each of the sorted training data is divided into corresponding groups based on a preset division ratio; The absolute value of the volume difference between the first compost volume and the second compost volume is calculated, and the sorting sequence number of each training data in the corresponding group is determined according to the absolute value; wherein, the smaller the absolute value is, the smaller the corresponding sorting sequence number is, that is, the closer it is to the front.

3. The method for optimizing process parameters for composting of manure and straw according to claim 2, characterized in that: Before constructing a piece of training data based on the first manure and straw data, the first process parameter, the compost evaluation data and the first compost volume in each piece of the compost turning and composting data, the method further includes: Obtaining data source information of each of the compost turning and fertilizer fermentation data, evaluating the process compliance of the data source information, and obtaining a process compliance value; If the process compliance value is lower than a preset value and the compost evaluation data of the compost turning and fermentation data is evaluated as medium level or above, an increase factor is determined according to the difference between the process compliance value and the preset value, and the compost evaluation data is optimized using the increase factor.

4. The method for optimizing process parameters for composting of manure and straw according to claim 3, characterized in that: According to the sorting order, a number of the training data are selected from each of the groups to form a training data group, including: In the first stage of training the process parameter optimization model, a number of the training data are selected from each of the groups according to a first selection rule to form a training data group; wherein the first selection rule refers to selecting training data accounting for a first proportion, a second proportion, and a third proportion of the total number from the groups with high-level, medium-level, and low-level compost evaluation data, respectively, and the first proportion is higher than the second proportion, and the second proportion is higher than the third proportion; In the second stage of training the process parameter optimization model, a number of the training data are selected from each of the groups according to the second selection rule to form a training data group; wherein the second selection rule refers to selecting the training data accounting for the fourth proportion, the fifth proportion, and the sixth proportion of the total from the groups with high, medium, and low levels of the compost evaluation data, respectively, and the fourth proportion is higher than the fifth proportion, the fifth proportion is higher than the sixth proportion, and the fourth proportion is lower than the first proportion.

5. The method for optimizing process parameters for composting of manure and straw according to claim 2, characterized in that: The division ratio is determined in the following manner: Acquire a number of training record data for the process parameter optimization model, wherein the training data include model test indicators and the historical division ratios adopted, select a number of training record data with the best model test indicators, and obtain the division ratio used for this training based on the historical division ratios corresponding to these training record data.

6. A system for optimizing process parameters of composting of manure and straw, characterized by: The system includes a collection unit, a training unit, and an optimization unit; The collecting unit is used to collect a number of real compost turning and composting data, each of which includes first manure and straw data, first process parameters, composting evaluation data and first composting volume; The training unit is used to construct a piece of training data according to the first manure straw data, the first process parameter, the compost evaluation data and the first compost volume in each piece of the compost turning and composting data; wherein the compost evaluation data is used to construct a label of the training data; Dividing the training data corresponding to each of the compost turning and fermentation data into a plurality of groups according to the compost evaluation data, and sorting the training data in each group according to the first compost volume; According to the sorting order, a number of the training data are respectively selected from each of the groups to form a training data group, and a plurality of the training data groups are obtained, and a process parameter optimization model is trained using a number of the training data in each of the training data groups until the training meets the standard; The optimization unit is used to determine the second manure straw data and the second compost volume involved in this composting, use the trained process parameter optimization model to process the second manure straw data and the second compost volume, and predict the second process parameters.

7. The optimization system for process parameters of manure and straw composting according to claim 6, characterized in that: The training unit is used to: The training data corresponding to each of the compost turning and composting data are sorted from high to low according to the compost evaluation data, and each of the sorted training data is divided into corresponding groups based on a preset division ratio; The absolute value of the volume difference between the first compost volume and the second compost volume is calculated, and the sorting sequence number of each training data in the corresponding group is determined according to the absolute value; wherein, the smaller the absolute value is, the smaller the corresponding sorting sequence number is, that is, the closer it is to the front.

8. An electronic device comprising: A memory storing executable program code; A processor coupled to the memory; characterized in that: the processor calls the executable program code stored in the memory to execute the method according to any one of claims 1-5.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is executed.

10. A computer program product, characterized in that: The computer program product is used to be called and executed by a processor of an electronic device to implement the method according to any one of claims 1 to 5.

Images