Compost maturity determination method, equipment and medium

Abstract

The invention discloses a compost maturity determination method and device and a medium, and relates to the technical field of biomass fermentation composting, the method comprises the following steps: sequentially carrying out pretreatment and characteristic wave band extraction treatment on real-time spectral information of compost to be evaluated to obtain real-time characteristic spectral data; the real-time total organic carbon content of the to-be-evaluated compost is obtained based on the total organic carbon content detection model, then the real-time organic matter reduction rate is determined, the real-time total nitrogen content of the to-be-evaluated compost is obtained based on the total nitrogen content detection model, and then the real-time carbon-nitrogen ratio reduction rate is determined. And inputting the real-time organic matter reduction rate and the real-time carbon-nitrogen ratio reduction rate into the compost maturity determination model to obtain a compost maturity detection result. According to the application, the spectral information characteristic wave band is extracted, and the total organic carbon content detection model, the total nitrogen content detection model and the compost maturity determination model are constructed, so that the compost maturity detection result is obtained. The efficiency and the accuracy of whether the compost is thoroughly decomposed can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of biomass fermentation composting, and in particular to a method, device and medium for determining the maturity of compost. Background Art

[0002] Composting technology can convert biodegradable waste into hygienic, pollution-free and humus-like substances, thereby recovering valuable nutrients and improving soil fertility and crop production quality. In agricultural production, the use of uncomposted organic fertilizers will cause the reproduction of microorganisms, leading to soil hypoxia, and will also attract underground pests to eat plant roots and stems. Parasites and pathogens will spread in the land, which will also increase the risk of large-scale outbreaks of crop diseases. Therefore, it is crucial to ensure the safety and availability of compost products through compost maturity assessment.

[0003] During the composting process, microorganisms convert carbon and nitrogen in organic matter into gases such as carbon dioxide and ammonia, which changes the carbon and nitrogen content in the compost. The change of carbon and nitrogen substances in the compost becomes an important indicator for evaluating the maturity of the product. However, the traditional carbon and nitrogen determination method is costly, time-consuming and labor-intensive. At present, with the development of near-infrared spectroscopy analysis technology, a variety of methods for determining the chemical components of compost based on near-infrared spectroscopy have been developed. With the development of machine learning, a variety of different machine learning models have been proposed and applied in this regard. A variety of feature extraction algorithms have also been used to screen feature bands to eliminate interference factors, reduce data dimensions, and improve model accuracy. In most studies, it has been shown that the machine learning model combined with feature band extraction can achieve good prediction accuracy for indicators such as total carbon and total nitrogen in compost through near-infrared spectral data.

[0004] However, due to the complex composition of compost raw materials in different places, the carbon and nitrogen components in the materials vary significantly. Traditional indicators used to evaluate the maturity of compost, such as carbon-nitrogen ratio, organic matter reduction rate, ammonia nitrogen, nitric nitrogen, etc., are often one-sided and difficult to form a unified standard. It is urgent to develop a fast, accurate, simple and efficient method for evaluating the maturity of compost. Therefore, a fast, accurate, simple and efficient method for evaluating the maturity of compost was constructed by combining near-infrared spectroscopy analysis technology and maturity evaluation model. Summary of the invention

[0005] The purpose of the present application is to provide a method, device and medium for determining the maturity of compost, which can improve the efficiency and accuracy of determining whether the compost is mature.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides a method for determining compost maturity, comprising:

[0008] Obtaining the initial total organic carbon content, initial total nitrogen content and real-time spectral information of the compost to be evaluated; the compost to be evaluated is the compost treated by the mature composting method; the initial total organic carbon content is the total organic carbon content of the compost to be evaluated before the mature composting method is used; the initial total nitrogen content is the total nitrogen content of the compost to be evaluated before the mature composting method is used;

[0009] The real-time spectrum information is sequentially preprocessed and subjected to characteristic band extraction processing to obtain real-time characteristic spectrum data;

[0010] The real-time characteristic spectrum data is input into a total organic carbon content detection model to obtain the real-time total organic carbon content of the compost to be evaluated; the total organic carbon content detection model is constructed based on the historical characteristic spectrum data and historical total organic carbon content of multiple composts using a gradient boosting regression algorithm;

[0011] Obtaining a real-time organic matter reduction rate according to the real-time total organic carbon content and the initial total organic carbon content;

[0012] The real-time characteristic spectrum data is input into a total nitrogen content detection model to obtain the real-time total nitrogen content of the compost to be evaluated; the total nitrogen content detection model is constructed based on the historical characteristic spectrum data and historical total nitrogen content of multiple composts using a gradient boosting regression algorithm;

[0013] Obtaining a real-time carbon-nitrogen ratio reduction rate according to the real-time total organic carbon content, the initial total organic carbon content, the real-time total nitrogen content and the initial total nitrogen content;

[0014] The real-time organic matter reduction rate and the real-time carbon-nitrogen ratio reduction rate are input into a compost maturity determination model to obtain a compost maturity detection result; the compost maturity determination model is obtained by training a support vector machine model using historical organic matter reduction rates and historical carbon-nitrogen ratio reduction rates of multiple composts; the compost maturity detection result is whether the compost is mature.

[0015] Optionally, the real-time spectrum information is subjected to preprocessing and characteristic band extraction processing in sequence to obtain real-time characteristic spectrum data, including:

[0016] Processing the real-time spectrum information using a polynomial smoothing algorithm to obtain real-time smoothed spectrum data;

[0017] Performing multivariate scattering correction on the real-time smoothed spectral data to obtain real-time denoised spectral data;

[0018] Performing first-order differential processing on the real-time denoised spectral data to obtain real-time differential spectral data;

[0019] The real-time differential spectrum data is subjected to characteristic band extraction processing to obtain real-time characteristic spectrum data.

[0020] Optionally, before obtaining the initial total organic carbon content, initial total nitrogen content and real-time spectrum information of the compost to be evaluated, the method further includes:

[0021] Prepare multiple compost samples and label the compost samples according to whether they are mature or not;

[0022] Determine the historical total organic carbon content, initial total organic carbon content, historical total nitrogen content, and initial total nitrogen content of each compost sample;

[0023] Obtain historical spectral information for each compost sample;

[0024] The historical spectrum information is sequentially preprocessed and subjected to characteristic band extraction processing to obtain historical characteristic spectrum data;

[0025] The first training set is constructed using historical characteristic spectral data and historical total organic carbon content as data pairs;

[0026] The second training set is constructed using historical characteristic spectrum data and historical total nitrogen content as data pairs;

[0027] Determine the historical organic matter reduction rate for each compost sample based on the historical total organic carbon content and the initial total organic carbon content of the compost sample;

[0028] Determine the historical carbon-nitrogen ratio reduction rate for each compost sample based on the historical total organic carbon content, initial total organic carbon content, historical total nitrogen content, and initial total nitrogen content of the compost sample;

[0029] Determine the historical organic matter reduction rate and the historical carbon-nitrogen ratio reduction rate for the same compost sample as input data pairs;

[0030] The third training set is constructed using the input data pairs and labels as data pairs.

[0031] Optionally, after constructing the first training set using the historical characteristic spectrum data and the historical total organic carbon content as a data pair, the method further includes:

[0032] Taking historical characteristic spectral data as input and historical total organic carbon content as output, a total organic carbon content detection model was constructed using the gradient boosting regression algorithm.

[0033] Optionally, after constructing the second training set using the historical characteristic spectrum data and the historical total nitrogen content as a data pair, the method further includes:

[0034] Taking historical characteristic spectral data as input and historical total nitrogen content as output, a total nitrogen content detection model was constructed using the gradient boosting regression algorithm.

[0035] Optionally, after constructing the third training set with the input data pair and the label as the data pair, the following steps are further included:

[0036] With the input data pairs as input and the labels as output, the support vector machine model is trained to obtain the compost maturity determination model.

[0037] Optionally, the historical organic matter reduction rate is:

[0038]

[0039] Among them, C i,R is the historical organic matter reduction rate of the i-th compost sample, C0 is the initial total organic carbon content of the i-th compost sample, C i is the historical total organic carbon content of the i-th compost sample.

[0040] Optionally, the historical carbon-nitrogen ratio reduction rate is:

[0041]

[0042]

[0043]

[0044] Among them, R i,R is the historical carbon-nitrogen ratio reduction rate of the i-th composting sample, R0 is the initial carbon-nitrogen ratio of the i-th composting sample, R i is the historical carbon-nitrogen ratio of the i-th compost sample, N i is the historical total nitrogen content of the i-th compost sample, and N0 is the initial total nitrogen content of the i-th compost sample.

[0045] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method for determining compost maturity.

[0046] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned method for determining compost maturity when executed by a processor.

[0047] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0048] The present application provides a method, device and medium for determining the maturity of compost. By exploring the physical and chemical index characteristics that affect the changes in organic matter and total nitrogen content during the composting process, near-infrared spectral data of compost samples are collected at the same time. Through the relationship between the spectral data of the compost samples and the organic matter and total nitrogen content, a rapid prediction model for the organic matter and total nitrogen content of the compost during the composting process is established. It is proposed to use the compost organic matter reduction rate and the carbon-nitrogen ratio reduction rate to jointly establish a compost maturity determination model, so as to judge the maturity of the compost, provide technical support for judging the quality of the compost, and provide a new idea for judging the maturity of the compost, which has important practical significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 This is a flow chart of a method for determining the maturity of compost in one embodiment of the present application;

[0051] Figure 2 A technical roadmap of a method for determining compost maturity in an embodiment of the present application;

[0052] Figure 3 This is a graph showing the average reflectance spectrum of compost samples at different composting times after treatment in an embodiment of the present application;

[0053] Figure 4 This is a schematic diagram of the division results of the compost maturity determination model in one embodiment of the present application;

[0054] Figure 5 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0057] In an exemplary embodiment, Figure 1 and Figure 2 As shown, a method for determining the maturity of compost is provided, comprising:

[0058] Step 101: Obtain the initial total organic carbon content, initial total nitrogen content and real-time spectral information of the compost to be evaluated. The compost to be evaluated is the compost treated by the mature composting method. The initial total organic carbon content is the total organic carbon content of the compost to be evaluated before the mature composting method is used. The initial total nitrogen content is the total nitrogen content of the compost to be evaluated before the mature composting method is used.

[0059] Step 102: pre-processing and characteristic band extraction processing are performed on the real-time spectrum information in sequence to obtain real-time characteristic spectrum data.

[0060] Step 103: Input the real-time characteristic spectrum data into the total organic carbon content detection model to obtain the real-time total organic carbon content of the compost to be evaluated. The total organic carbon content detection model is constructed based on the historical characteristic spectrum data and historical total organic carbon content of multiple composts using a gradient boosting regression algorithm.

[0061] Step 104: Obtain a real-time organic matter reduction rate according to the real-time total organic carbon content and the initial total organic carbon content.

[0062] Step 105: Input the real-time characteristic spectrum data into the total nitrogen content detection model to obtain the real-time total nitrogen content of the compost to be evaluated. The total nitrogen content detection model is constructed based on the historical characteristic spectrum data and historical total nitrogen content of multiple composts using a gradient boosting regression algorithm.

[0063] Step 106: Obtain a real-time carbon-nitrogen ratio reduction rate according to the real-time total organic carbon content, the initial total organic carbon content, the real-time total nitrogen content and the initial total nitrogen content.

[0064] Step 107: Input the real-time organic matter reduction rate and the real-time carbon-nitrogen ratio reduction rate into the compost maturity determination model to obtain the compost maturity detection result. The compost maturity determination model is obtained by training the support vector machine model using the historical organic matter reduction rate and the historical carbon-nitrogen ratio reduction rate of multiple composts. The compost maturity detection result is whether the compost is mature.

[0065] Step 102 includes:

[0066] Step 102 - 1 : Process the real-time spectrum information using a polynomial smoothing algorithm to obtain real-time smoothed spectrum data.

[0067] Step 102 - 2 : Perform multivariate scattering correction on the real-time smoothed spectral data to obtain real-time denoised spectral data.

[0068] Step 102 - 3 : Perform first-order differential processing on the real-time denoised spectral data to obtain real-time differential spectral data.

[0069] Step 102 - 4 : performing characteristic band extraction processing on the real-time differential spectrum data to obtain real-time characteristic spectrum data.

[0070] In spectral signal processing, the polynomial smoothing algorithm (SG) performs polynomial least squares fitting on a spectral curve by setting different polynomial orders and smoothing window widths. Its essence is a weighted average method that emphasizes the central role of the center point. The calculation formula of the SG algorithm is as follows:

[0071]

[0072] Among them, x i，S-G is the spectral reflectance after SG smoothing at wavelength i, C j is the smoothing coefficient, which can be obtained by polynomial fitting, m is the number of smoothing windows on one side of the wavelength, N is the normalization index, and x i+j is the spectral reflectance at wavelength i+j in the smoothing window.

[0073] The purpose of spectral multivariate scatter correction (MSC) is to effectively eliminate the spectral differences caused by different scattering levels in the spectral data obtained after scattering correction, thereby enhancing the correlation between the spectrum and the data.

[0074] The average value of all spectral data is taken as the "ideal spectrum".

[0075] Compute the average spectrum:

[0076]

[0077] in, is the average spectral reflectance at wavelength i, A i,j is the spectral reflectance of the jth spectrum at wavelength i, and n is the total number of spectra.

[0078] The spectrum of each sample is subjected to a univariate linear regression operation with the average spectrum to obtain the linear translation (regression constant) and tilt offset (regression coefficient) of each spectrum relative to the standard spectrum.

[0079] Univariate linear regression:

[0080]

[0081] Among them, A i(msc) is the spectral reflectance after multivariate scattering correction at wavelength i, m i and b i Represents the near infrared spectrum of each sample A iThe relative shift coefficient and translation amount obtained after performing a univariate linear regression with the average spectrum A.

[0082] The linear shift is subtracted from the original spectrum of each sample and divided by the regression coefficient to correct the relative tilt of the spectrum's baseline. In this way, the baseline shift and offset of each spectrum are corrected with reference to the standard spectrum, while the spectral absorption information corresponding to the sample component content has no effect in the entire data processing process, thereby improving the signal-to-noise ratio of the spectrum.

[0083] Before step 101, the method further includes:

[0084] Step 108: Prepare a plurality of compost samples, and label the compost samples according to whether the compost samples are mature or not.

[0085] Compost samples from different stages of different composting processes were collected, and after natural air drying, they were crushed, fully ground, and passed through a 20-mesh sieve.

[0086] Step 109: Determine the historical total organic carbon content, initial total organic carbon content, historical total nitrogen content, and initial total nitrogen content of each compost sample.

[0087] Step 1010: Obtain historical spectrum information of each compost sample.

[0088] Step 1011: pre-processing and characteristic band extraction processing are performed on the historical spectrum information in sequence to obtain historical characteristic spectrum data.

[0089] Preprocessing and transformation of compost spectrum: The collected raw spectral information is subjected to polynomial smoothing algorithm and multivariate scattering correction to remove data noise and eliminate data differences caused by different scattering levels. The spectral data is processed by first-order differential to expand the spectral information difference and improve its correlation with physical and chemical properties.

[0090] Extraction of characteristic wavelengths: Match the spectral information of the compost sample processed in step three with the physical and chemical properties of the compost sample to establish a database; use competitive adaptive reweighted sampling (CARS) to screen characteristic variables, eliminate interference factors, reduce data dimensions, and improve model accuracy.

[0091] Step 1012: construct a first training set using historical characteristic spectrum data and historical total organic carbon content as data pairs.

[0092] Step 1013: construct a second training set using historical characteristic spectrum data and historical total nitrogen content as data pairs.

[0093] Step 1014: Determine the historical organic matter reduction rate of each compost sample based on the historical total organic carbon content and the initial total organic carbon content of the compost sample. The historical organic matter reduction rate is:

[0094]

[0095] Among them, C i,R is the historical organic matter reduction rate of the i-th compost sample, C0 is the initial total organic carbon content of the i-th compost sample, C i is the historical total organic carbon content of the i-th compost sample.

[0096] Step 1015: Determine the historical carbon-nitrogen ratio reduction rate of each compost sample based on the historical total organic carbon content, initial total organic carbon content, historical total nitrogen content, and initial total nitrogen content of the compost sample. The historical carbon-nitrogen ratio reduction rate is:

[0097]

[0098]

[0099]

[0100] Among them, R i,R is the historical carbon-nitrogen ratio reduction rate of the i-th composting sample, R0 is the initial carbon-nitrogen ratio of the i-th composting sample, R i is the historical carbon-nitrogen ratio of the i-th compost sample, N i is the historical total nitrogen content of the i-th compost sample, and N0 is the initial total nitrogen content of the i-th compost sample.

[0101] Step 1016: Determine the historical organic matter reduction rate and the historical carbon-nitrogen ratio reduction rate of the same compost sample as input data pairs.

[0102] Step 1017: Construct a third training set using the input data pairs and labels as data pairs.

[0103] Step 1018: Using historical characteristic spectral data as input and historical total organic carbon content as output, a total organic carbon content detection model is constructed using a gradient boosting regression algorithm.

[0104] Step 1019: Using historical characteristic spectral data as input and historical total nitrogen content as output, a total nitrogen content detection model is constructed using a gradient boosting regression algorithm.

[0105] The gradient boosting regression (GBR) algorithm was used to construct the total organic carbon content detection model and the total nitrogen content detection model for the selected characteristic variables, and their prediction effects were evaluated.

[0106] Step 1020: Using the input data pairs as input and the labels as output, the support vector machine model is trained to obtain a compost maturity determination model.

[0107] The output of the physical and chemical property prediction model is converted into organic matter reduction rate and carbon-nitrogen ratio reduction rate, which are used as input to the compost maturity determination model, thereby forming a model composed of multiple machine learning models for judging compost maturity based on near-infrared spectroscopy technology.

[0108] The present invention collects spectral information of compost through a hyperspectral camera, and the spectral information contains a large amount of information about the organic matter content of the compost. Due to its large data volume and high dimension, interference data reduction and dimensionality reduction algorithms are adopted, and a detection model is established in combination with the physical and chemical data obtained from physical and chemical experiments. The optimal spectral bands and the optimal model for predicting compost organic matter and total nitrogen are screened out to achieve accurate prediction of the compost organic matter content and total nitrogen through near-infrared spectral data, and then the compost organic matter reduction rate and carbon-nitrogen ratio reduction rate are obtained through the predicted value pairs, and a machine learning model is constructed to judge the maturity of the compost.

[0109] The following two evaluation indicators were used to evaluate the prediction ability of the model: Coefficient of Determination (R 2 ), Root Mean Square Error (RMSE).

[0110] Taking the composting experiment as an example, modeling and verification were carried out respectively. 70% of the data were randomly selected for modeling, and the rest were used for verification. During verification, the following two evaluation indicators were used to evaluate the prediction ability of the model: Coefficient of Determination (R 2 ), root mean square error (Root Mean Square Error, RMSE).

[0111] Composting Information:

[0112] The raw materials for high temperature fermentation of livestock and poultry manure were taken from Lingyan Breeding Farm in Gongzhuling City, Jilin Province, and the corn straw was sourced from Junping Agricultural Cooperative in Gongzhuling City, Jilin Province. The specific parameters are shown in Table 1.

[0113] Table 1 Composting raw materials

[0114]

[0115]

[0116] Cow dung and pig dung were mixed evenly with corn stalks according to the experimental setting ratio, and the moisture content of the pile was adjusted between 60% and 70%, and then placed in a 100L PVC barrel for composting. During the composting period, the oxygen content was ensured by manual turning. The pile was turned every two days in the first ten days, and once a week in the next four weeks. After entering the mature stage, the pile was not turned. All samples were collected before turning. From the beginning to the end of the composting, a total of 51 days, the temperature of the upper, middle and lower layers of the pile was measured and recorded at 1 pm every day. The pile samples were collected at different positions in the middle layer of the pile on the 1st, 3rd, 5th, 7th, 15th, 21st, 35th and 51st days after the start of composting and mixed evenly. The mixed sample was divided into three parts, two of which were stored at minus 20℃, and one was naturally air-dried and crushed into powder. The samples were used for analysis of various indicators.

[0117] Acquisition of compost physicochemical properties and spectral data:

[0118] The total organic carbon of the compost sample was detected using the acid potassium dichromate volumetric method. The total nitrogen of the compost sample was determined using an elemental analyzer. The near-infrared spectral data were collected using the ASD Field Spec 4 portable ground feature spectrometer, with a detection wavelength of 350-2500nm and a spectral resolution of 3nm@700nm and 6nm@1400nm / 2100nm. The near-infrared spectral data were collected in a darkroom, and the ground compost sample was placed in a cuvette. During the collection, the cuvette containing the compost sample was placed under the collection device. When the light source reached the surface of the cuvette through the lens, the diffuse reflected light was received by the collection device, and the collection device then transmitted the spectral data to the computer for storage to form a data set.

[0119] Preprocessing and transformation of compost spectrum:

[0120] The polynomial smoothing algorithm and multivariate scattering correction were performed on the collected 200 compost spectral data, and then the near-infrared spectral data of the compost samples were transformed by the first-order differential transformation. The average reflectance spectral curves of the compost samples at different composting times after preprocessing and first-order differential transformation are shown in Figure 2. Figure 3 shown.

[0121] Extraction of characteristic wavelength:

[0122] The obtained spectral curve was processed by competitive adaptive reweighting algorithm to obtain the spectral characteristic bands of total organic carbon and total nitrogen in the compost sample. The spectral characteristic bands are shown in Table 2.

[0123] Table 2 Spectral characteristic band table

[0124]

[0125] Construction of total organic carbon and total nitrogen detection model:

[0126] 70% of the samples were randomly selected as training sets and 30% of the samples were selected as test sets. The processed data were uniformly transferred into Python, and the total organic carbon and total carbon data were fitted with the corresponding characteristic spectral data using the gradient boosting regression (GBR) algorithm for the training set data. After parameter adjustment and training, the optimal model was finally obtained. The above model was used to verify the prediction model of total organic carbon and total nitrogen in compost samples. The specific prediction results are shown in Table 3.

[0127] Table 3 Model evaluation index table

[0128]

[0129] Obtaining input values ​​of the compost maturity determination model and constructing the model:

[0130] Similarly, 70% of the samples were randomly selected as the training set and 30% of the samples were used as the test set. The predicted values ​​of the compost sample organic matter reduction rate and carbon-nitrogen ratio reduction rate calculated by the formula were uniformly transferred into python, and the support vector classifier (SVC) and radial basis function (RBF) were used as kernel functions to fit the organic matter reduction rate and carbon-nitrogen ratio reduction rate data with the maturity of the compost sample. After parameter adjustment and training, the optimal model was finally obtained. The above model was used to predict and verify the maturity of the compost sample. Figure 4 This is the SVM model classification result. The specific prediction results are shown in Table 4.

[0131] Table 4 Evaluation index table of maturity evaluation model

[0132]

[0133] Evaluation of the combination of models and the effect of judging maturity:

[0134] The output values ​​of the total organic carbon and total nitrogen prediction models were converted into organic matter reduction rate and carbon-nitrogen ratio reduction rate through formula calculation, and used as the input of the compost maturity determination model. The two models were combined to form a model for directly judging the compost maturity through the near-infrared spectral information of the compost. The effect of the model was evaluated, and the results are shown in Table 5.

[0135] Table 5 Model effect table

[0136]

[0137] The present application also provides an application scenario, which applies the above-mentioned compost maturity determination method. Specifically: the compost maturity determination method provided in this embodiment can be applied in a greenhouse control scenario. The scenario includes: fertilization control link, irrigation control link and greenhouse automation control link; the compost maturity determination method provided in this embodiment belongs to the fertilizer quality detection link in the fertilization control link.

[0138] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for determining the maturity of compost is implemented.

[0139] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components. In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0140] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0141] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0142] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0143] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0144] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0145] The technical features of the above embodiments may be arbitrarily combined. 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.

[0146] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for determining compost maturity, characterized in that: include: Obtaining the initial total organic carbon content, initial total nitrogen content and real-time spectral information of the compost to be evaluated; the compost to be evaluated is the compost treated by the mature composting method; the initial total organic carbon content is the total organic carbon content of the compost to be evaluated before the mature composting method is used; the initial total nitrogen content is the total nitrogen content of the compost to be evaluated before the mature composting method is used; The real-time spectrum information is sequentially preprocessed and subjected to characteristic band extraction processing to obtain real-time characteristic spectrum data; The real-time characteristic spectrum data is input into a total organic carbon content detection model to obtain the real-time total organic carbon content of the compost to be evaluated; the total organic carbon content detection model is constructed based on the historical characteristic spectrum data and historical total organic carbon content of multiple composts using a gradient boosting regression algorithm; Obtaining a real-time organic matter reduction rate according to the real-time total organic carbon content and the initial total organic carbon content; The real-time characteristic spectrum data is input into a total nitrogen content detection model to obtain the real-time total nitrogen content of the compost to be evaluated; the total nitrogen content detection model is constructed based on the historical characteristic spectrum data and historical total nitrogen content of multiple composts using a gradient boosting regression algorithm; Obtaining a real-time carbon-nitrogen ratio reduction rate according to the real-time total organic carbon content, the initial total organic carbon content, the real-time total nitrogen content and the initial total nitrogen content; The real-time organic matter reduction rate and the real-time carbon-nitrogen ratio reduction rate are input into a compost maturity determination model to obtain a compost maturity detection result; the compost maturity determination model is obtained by training a support vector machine model using the historical organic matter reduction rates and historical carbon-nitrogen ratio reduction rates of multiple composts; The compost maturity detection result is whether the compost is mature.

2. The method for determining compost maturity according to claim 1, characterized in that: The real-time spectrum information is sequentially preprocessed and subjected to characteristic band extraction processing to obtain real-time characteristic spectrum data, including: Processing the real-time spectrum information using a polynomial smoothing algorithm to obtain real-time smoothed spectrum data; Performing multivariate scattering correction on the real-time smoothed spectral data to obtain real-time denoised spectral data; Performing first-order differential processing on the real-time denoised spectral data to obtain real-time differential spectral data; The real-time differential spectrum data is subjected to characteristic band extraction processing to obtain real-time characteristic spectrum data.

3. The method for determining compost maturity according to claim 1, characterized in that: Before obtaining the initial total organic carbon content, initial total nitrogen content and real-time spectral information of the compost to be evaluated, it also includes: Prepare multiple compost samples and label the compost samples according to whether they are mature or not; Determine the historical total organic carbon content, initial total organic carbon content, historical total nitrogen content, and initial total nitrogen content of each compost sample; Obtain historical spectral information for each compost sample; The historical spectrum information is sequentially preprocessed and subjected to characteristic band extraction processing to obtain historical characteristic spectrum data; The first training set is constructed using historical characteristic spectral data and historical total organic carbon content as data pairs; The second training set is constructed using historical characteristic spectrum data and historical total nitrogen content as data pairs; Determine the historical organic matter reduction rate for each compost sample based on the historical total organic carbon content and the initial total organic carbon content of the compost sample; Determine the historical carbon-nitrogen ratio reduction rate for each compost sample based on the historical total organic carbon content, initial total organic carbon content, historical total nitrogen content, and initial total nitrogen content of the compost sample; Determine the historical organic matter reduction rate and the historical carbon-nitrogen ratio reduction rate for the same compost sample as input data pairs; The third training set is constructed using the input data pairs and labels as data pairs.

4. The method for determining compost maturity according to claim 3, characterized in that: After constructing the first training set with historical characteristic spectral data and historical total organic carbon content as data pairs, it also includes: Taking historical characteristic spectral data as input and historical total organic carbon content as output, a total organic carbon content detection model was constructed using the gradient boosting regression algorithm.

5. The method for determining compost maturity according to claim 3, characterized in that: After constructing the second training set with historical characteristic spectrum data and historical total nitrogen content as data pairs, it also includes: Taking historical characteristic spectral data as input and historical total nitrogen content as output, a total nitrogen content detection model was constructed using the gradient boosting regression algorithm.

6. The method for determining compost maturity according to claim 3, characterized in that: After constructing the third training set with input data pairs and labels as data pairs, it also includes: With the input data pairs as input and the labels as output, the support vector machine model is trained to obtain the compost maturity determination model.

7. The method for determining compost maturity according to claim 3, characterized in that: The historical organic matter reduction rate is: Among them, C i,R is the historical organic matter reduction rate of the i-th compost sample, C0 is the initial total organic carbon content of the i-th compost sample, C i is the historical total organic carbon content of the i-th compost sample.

8. The method for determining compost maturity according to claim 7, characterized in that: The historical carbon-nitrogen ratio reduction rate is: Among them, R i,R is the historical carbon-nitrogen ratio reduction rate of the i-th composting sample, R0 is the initial carbon-nitrogen ratio of the i-th composting sample, R i is the historical carbon-nitrogen ratio of the i-th compost sample, N i is the historical total nitrogen content of the i-th compost sample, and N0 is the initial total nitrogen content of the i-th compost sample.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for determining the maturity of compost according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for determining the degree of maturity of compost according to any one of claims 1 to 8 is implemented.

Images