A method for environmental monitoring during the development of Bacillus bio-organic fertilizer

Abstract

The present invention relates to the field of environmental monitoring technology, and in particular to an environmental monitoring method for the development process of Bacillus bio-organic fertilizer. The method comprises: obtaining the initial smoothing coefficient of each subsequence to be analyzed according to the differential sequence and curvature characterization value corresponding to the target subsequence and the differential sequence and curvature characterization value corresponding to the subsequence to be analyzed, and obtaining the optimal smoothing coefficient of the environmental parameter sequence according to the initial smoothing coefficient of each subsequence to be analyzed; smoothing the environmental parameter sequence using the optimal smoothing coefficient, and recording the sequence obtained after smoothing as the target smoothed sequence, and predicting the smoothed sequence using a quadratic exponential smoothing algorithm with the smoothing coefficient being the optimal smoothing coefficient, obtaining the predicted data at the future moment, and obtaining the monitoring result at the current monitoring moment according to the predicted data. The present invention can improve the timeliness and reliability of monitoring the environment in a fermentation container.

Description

Technical Field

[0001] The invention relates to the technical field of environmental monitoring, and in particular to an environmental monitoring method in the process of producing a bacillus bio-organic fertilizer. Background Art

[0002] ‌Since Bacillus bio-organic fertilizer has the functions of improving soil environment, promoting crop growth, biological control and resisting diseases and pests, etc.‌, Bacillus bio-organic fertilizer is currently mainly used in the agricultural field, and the development or production of Bacillus bio-organic fertilizer is usually carried out in a fermentation container; however, if in the development or production process of Bacillus bio-organic fertilizer, if the development environment or the production environment does not meet the requirements, and the development environment or the production environment refers to the environment in the fermentation container, then the quality of the subsequently developed or produced Bacillus bio-organic fertilizer will not meet the requirements, and it will also cause a waste of resources. Therefore, in the development or production process of Bacillus bio-organic fertilizer, it is very important to conduct timely and reliable monitoring of the environment in the fermentation container.

[0003] In the prior art, in order to achieve the purpose of timely and reliable monitoring of the environment in the fermentation container, an exponential smoothing method is generally used to predict the environmental data collected by the sensor, wherein the environmental data is the environmental data in the fermentation container, and then the predicted data is analyzed, and whether the environment at the monitored time meets the requirements is determined based on the analysis result, that is, timely monitoring of the environment in the fermentation container is achieved based on the predicted data, and the smoothing coefficient when the prior art uses the exponential smoothing method for prediction is generally an empirical value; however, during the development or production process, some external factors may affect or interfere with the induction of the sensor, resulting in the environmental data collected by the sensor having low accuracy or causing the collected data to deviate from the real environmental data, that is, causing the authenticity or accuracy of the collected environmental data to be affected or interfered by external factors, and when the problem of low accuracy or deviation from the real data occurs, the accuracy of the subsequent predicted data obtained based on the collected data will also be low, resulting in the inability to timely and reliably monitor the environment in the fermentation container, so how to improve the timeliness and reliability of monitoring the environment in the fermentation container is an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above problems, the present invention provides an environmental monitoring method for the production process of Bacillus bio-organic fertilizer, and the technical scheme adopted is as follows:

[0005] An embodiment of the present invention provides an environmental monitoring method for a Bacillus bio-organic fertilizer production process, comprising the following steps:

[0006] In the current monitoring time period, an environmental parameter sequence of a fermentation container is obtained, wherein the fermentation container is a container in which the Bacillus bio-organic fertilizer is located during the preparation process;

[0007] According to the preset sliding window and the environmental parameter sequence, a set of subsequences to be analyzed and a target fitting curve corresponding to the environmental parameter sequence are obtained, and according to the target fitting curve and a differential sequence corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed, a target subsequence of the set of subsequences to be analyzed is obtained;

[0008] Obtaining the curvature characterization value corresponding to the subsequence, and obtaining the initial smoothing coefficient of each subsequence to be analyzed according to the differential sequence and curvature characterization value corresponding to the target subsequence and the differential sequence and curvature characterization value corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed; obtaining the optimal smoothing coefficient of the environmental parameter sequence according to the initial smoothing coefficient of each subsequence to be analyzed;

[0009] The environmental parameter sequence is smoothed using the optimal smoothing coefficient, and the sequence obtained after smoothing is recorded as the target smoothed sequence. The smoothed sequence is predicted using a quadratic exponential smoothing algorithm whose smoothing coefficient is the optimal smoothing coefficient to obtain predicted data at future moments, and the monitoring results at the current monitoring moment are obtained based on the predicted data.

[0010] Beneficial effects: The present invention first obtains the environmental parameter sequence of the fermentation container; then, according to the preset sliding window and the environmental parameter sequence, obtains the set of subsequences to be analyzed and the target fitting curve corresponding to the environmental parameter sequence, and obtains the target subsequence of the subsequence set to be analyzed according to the target fitting curve and the differential sequence corresponding to each subsequence to be analyzed in the subsequence set to be analyzed, and then obtains the initial smoothing coefficient of each subsequence to be analyzed according to the differential sequence and curvature characterization value corresponding to the target subsequence of the subsequence set to be analyzed and the differential sequence and curvature characterization value corresponding to each subsequence to be analyzed in the subsequence set to be analyzed, and obtains the optimal smoothing coefficient of the environmental parameter sequence according to the initial smoothing coefficient of each subsequence to be analyzed; finally, the environmental parameter sequence is smoothed using the optimal smoothing coefficient, and the sequence obtained after smoothing is recorded as the target smoothed sequence, and the smoothed sequence is predicted using a quadratic exponential smoothing algorithm with the smoothing coefficient being the optimal smoothing coefficient to obtain the predicted data at a future moment, and the monitoring result at the current monitoring moment is obtained according to the predicted data. The present invention reduces the interference of external factors on the accuracy and reliability of subsequent data prediction through the optimal smoothing coefficient, that is, by determining an optimal smoothing coefficient to improve the accuracy and reliability of subsequent data prediction, thereby also improving the timeliness and reliability of monitoring the environment in the fermentation container. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 The present invention is a flow chart of an environmental monitoring method for a process of producing a Bacillus bio-organic fertilizer. DETAILED DESCRIPTION

[0013] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the embodiments of the present invention.

[0014] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0015] This embodiment provides an environmental monitoring method for the production process of Bacillus bio-organic fertilizer, which is described in detail as follows:

[0016] like Figure 1 As shown, the environmental monitoring method for the preparation process of the Bacillus bio-organic fertilizer comprises the following steps:

[0017] Step S001, in the current monitoring time period, obtaining the environmental parameter sequence of the fermentation container, wherein the fermentation container is the container in which the Bacillus bio-organic fertilizer is located during the preparation process.

[0018] This embodiment mainly reduces the interference of the collected environmental parameters by adjusting the smoothing coefficient in real time. That is, this embodiment mainly analyzes the collected environmental parameter sequence to determine the optimal smoothing coefficient of the environmental parameter sequence. The optimal smoothing coefficient can improve the accuracy and reliability of data prediction, thereby improving the timeliness and reliability of monitoring the environment in the fermentation container.

[0019] In addition, for ease of understanding, the present embodiment will be analyzed with the development process or the production process of Velez subtilis TA-1 biological organic fertilizer as an example, that is, the present embodiment mainly monitors the environment where Velez subtilis TA-1 biological organic fertilizer is in development or production, and the environment where organic fertilizer is in development or production refers to fermentation vessel, therefore the fermentation vessel that occurs in the follow-up of the present embodiment refers to the container that Velez subtilis TA-1 biological organic fertilizer uses when developing, and the environmental parameters collected in the follow-up are also the environmental parameters in this fermentation vessel.

[0020] This embodiment first obtains the environmental parameters in the fermentation container at each monitoring moment in the current monitoring time period, and records the time series constructed by the same type of environmental parameters obtained in the current monitoring time period as the environmental parameter sequence of the fermentation container, that is, multiple types of environmental parameter sequences can be obtained; and the types of environmental parameters collected in this embodiment include but are not limited to temperature, humidity, light intensity, pH value, gas composition, etc., but the types of all parameters in the same environmental parameter sequence must be consistent, and this embodiment will subsequently take any environmental parameter sequence as an example for analysis, such as this embodiment will take temperature data as an example for analysis, that is, the environmental parameter sequence analyzed subsequently in this embodiment is a temperature sequence.

[0021] In specific applications, the implementer needs to set the time interval between the current monitoring time period and the adjacent monitoring moments according to the actual situation. For example, in this embodiment, the ten minutes before the current moment can be used as the current monitoring time period, and the current monitoring time period includes the current monitoring moment. In this embodiment, the time interval between adjacent monitoring moments can be set to 1 second.

[0022] Therefore, this embodiment can obtain the environmental parameter sequence through the above process.

[0023] Step S002, according to the preset sliding window and the environmental parameter sequence, obtain the set of subsequences to be analyzed and the target fitting curve corresponding to the environmental parameter sequence, and obtain the target subsequence of the subsequence set to be analyzed according to the target fitting curve and the differential sequence corresponding to each subsequence to be analyzed in the subsequence set to be analyzed; obtain the curvature representation value corresponding to the subsequence, and obtain the initial smoothing coefficient of each subsequence to be analyzed according to the differential sequence and curvature representation value corresponding to the target subsequence and the differential sequence and curvature representation value corresponding to each subsequence to be analyzed in the subsequence set to be analyzed; obtain the optimal smoothing coefficient of the environmental parameter sequence according to the initial smoothing coefficient of each subsequence to be analyzed;.

[0024] And when collecting the environmental parameters in the fermentation container, if there is no interference from external factors, generally speaking, the curve drawn based on the collected environmental parameters generally shows regular changes. If there is interference from external factors, the collected temperature data curve is a relatively smooth curve. However, generally speaking, most of the data collection will be interfered or affected by external factors, that is, external factors will affect or interfere with the sensing of the sensor. For example, nuclear magnetic resonance technology is required in the development process, and the equipment related to nuclear magnetic resonance technology used may generate electromagnetic interference to affect the sensing of the sensor. When the sensing of the sensor is affected by external factors, the environmental data collected by the sensor will have low accuracy or the collected data may deviate from the actual environmental data, that is, the authenticity or accuracy of the collected environmental data will be affected by external factors or When the accuracy is low or there is a deviation from the real data, the accuracy and reliability of the predicted data subsequently obtained based on the collected data will also be low. When the accuracy and reliability of the predicted data are low, the environment in the fermentation container cannot be monitored in a timely and reliable manner. In order to reduce the problem of the inability to monitor the environment in the fermentation container in a relatively timely and reliable manner caused by the interference of external factors, that is, to improve the timeliness and reliability of monitoring the environment in the fermentation container, this embodiment will subsequently determine an optimal smoothing coefficient to reduce the accuracy and reliability of subsequent data prediction caused by external factors, that is, to improve the accuracy and reliability of subsequent data prediction by determining an optimal smoothing coefficient, thereby achieving the purpose of improving the timeliness and reliability of monitoring the environment in the fermentation container.

[0025] Since the local interference conditions of the environmental parameter sequence are different, this embodiment needs to first perform local analysis on the environmental parameter sequence, and compare the data with the same trend through local analysis, that is, to divide the subsequences with the same trend into the same set, that is, to first obtain the subsequence sets to be analyzed corresponding to the environmental parameter sequence according to the preset sliding window and the environmental parameter sequence. The specific process is as follows:

[0026] First, a preset sliding window is obtained, and the preset sliding window is slid on the environmental parameter sequence with a preset moving step length to obtain each window on the environmental parameter sequence, and then the sequence formed by the data in each window on the environmental parameter sequence is recorded as a subsequence on the environmental parameter sequence, and in this embodiment, the preset moving step length is the same as the length of the preset sliding window. In addition, in a specific application, the implementer needs to set the size of the preset sliding window according to the actual situation. If the number of environmental parameters in the environmental parameter sequence is N, then the value of n is required to belong to An integer in , such as the value of n can be 5.

[0027] Then, in this embodiment, the set constructed by constructing all subsequences on the environmental parameter sequence is recorded as the initial subsequence set, and the a-th subsequence in the initial subsequence set is the a-th subsequence on the environmental parameter sequence; then, all abnormal subsequences in the initial subsequence set are obtained, and in this embodiment, the specific process of obtaining the abnormal subsequence is as follows:

[0028] For the ath subsequence in the initial subsequence set, if the correlation coefficients between the ath subsequence and the a-1th subsequence and the a+1th subsequence in the initial subsequence set are not greater than the preset correlation coefficient threshold, the ath subsequence is recorded as an abnormal subsequence, and a is not equal to 1 and A, and A is the total number of subsequences in the initial subsequence set; for the 1st subsequence and the Ath subsequence in the initial subsequence set, if the correlation coefficient between the 1st subsequence and the 2nd subsequence in the initial subsequence set is not greater than the preset correlation coefficient threshold, the 1st subsequence is directly recorded as an abnormal subsequence, and if the correlation coefficient between the Ath subsequence and the A-1th subsequence in the initial subsequence set is not greater than the preset correlation coefficient threshold, the Ath subsequence is recorded as an abnormal subsequence.

[0029] Therefore, this embodiment can obtain all abnormal subsequences in the initial subsequence set through the above process, and after obtaining the abnormal subsequences, the target fitting curve is obtained according to the abnormal subsequences, and the specific acquisition process of the target fitting curve is:

[0030] In the environmental parameter sequence, all environmental parameters belonging to the abnormal subsequence are eliminated, and the sequence constructed by all the remaining environmental parameters after the elimination is recorded as the sequence to be fitted, and the parameters in the sequence to be fitted are also arranged in the order of acquisition time; after obtaining the sequence to be fitted, curve fitting is performed on the sequence to be fitted, and the curve obtained by fitting is recorded as the target fitting curve, which is an important data for subsequently determining the optimal smoothing coefficient of the environmental parameter sequence, and the specific acquisition process of obtaining the target fitting curve is: first, a two-dimensional mapping space is constructed, and the abscissa of the two-dimensional mapping space is time, and the ordinate is the environmental parameter value; then, each environmental parameter in the sequence to be fitted and the acquisition time of each environmental parameter are mapped to the two-dimensional mapping space, and the data point corresponding to each environmental parameter in the sequence to be fitted is obtained, and the ordinate value of the data point corresponding to the environmental parameter is the corresponding environmental parameter, and the abscissa value is the acquisition time of the corresponding environmental parameter; then, polynomial regression fitting is performed on the data points corresponding to all environmental parameters in the sequence to be fitted using the least squares method, and the curve obtained by fitting is recorded as the target fitting curve; and the process of obtaining the fitting curve by polynomial regression fitting using the least squares method is a well-known technology, so this embodiment will not be described in detail.

[0031] After obtaining the target fitting curve, a set of subsequences to be divided is obtained according to the abnormal subsequences, and the set of subsequences to be divided refers to a set formed by all subsequences remaining after all abnormal subsequences in the initial subsequence set are removed; after obtaining the set of subsequences to be divided, a correlation analysis is performed on the subsequences in the set of subsequences to be divided, and each set of subsequences to be analyzed is determined based on the analysis results. The specific process is:

[0032] First, determine whether the correlation coefficient between the second subsequence in the subsequence set to be divided and the first subsequence in the subsequence set to be divided is greater than the preset correlation coefficient threshold. If so, continue to determine whether the correlation coefficient between the third subsequence in the subsequence set to be divided and the second subsequence is greater than the preset correlation coefficient threshold. If not, record the sequence set constructed by the first subsequence and the second subsequence as the first subsequence set to be analyzed corresponding to the environmental parameter sequence; and continue to determine the correlation coefficient between the fourth subsequence in the subsequence set to be divided and the third subsequence. If so, continue to judge whether the correlation coefficient between the 5th subsequence and the 4th subsequence in the subsequence set to be divided is greater than the preset correlation coefficient threshold. If not, record the sequence set constructed by the 3rd subsequence and the 4th subsequence as the second subsequence set to be analyzed corresponding to the environmental parameter sequence, and so on, until all subsequences in the subsequence set to be divided are traversed, and all subsequence sets to be analyzed corresponding to the environmental parameter sequence are counted, and all subsequences in the subsequence set to be analyzed are recorded as subsequences to be analyzed.

[0033] In addition, the correlation coefficient that appears in this embodiment refers to the Spearman correlation coefficient, and the calculation process of the Spearman correlation coefficient between two sequences is a well-known technology, so this embodiment will not be described in detail; and in specific applications, the implementer needs to set the preset correlation coefficient threshold according to the actual situation, such as setting the preset correlation coefficient threshold to 0.6 in this embodiment.

[0034] Therefore, this embodiment can obtain all the subsequence sets to be analyzed and the target fitting curve through the above process; and after obtaining the subsequence sets to be analyzed and the target fitting curve, the differential sequence corresponding to each subsequence to be analyzed in each subsequence set to be analyzed is obtained, and in this embodiment, the g-th value in the differential sequence corresponding to a subsequence to be analyzed refers to the absolute value of the difference between the g+1-th parameter in the subsequence to be analyzed and the g-th parameter in the subsequence to be analyzed; then, according to the target fitting curve and the differential sequence corresponding to each subsequence to be analyzed in each subsequence set to be analyzed, the target subsequence of each subsequence set to be analyzed is obtained, and whether the target subsequence is the key to determine the optimal smoothing coefficient; in addition, for ease of understanding, this embodiment will be described later by taking the acquisition process of the target subsequence corresponding to any subsequence set W to be analyzed as an example, that is, the acquisition process of the target subsequence of the subsequence set W to be analyzed is:

[0035] First, the fitting similarity characterization value of each subsequence to be analyzed in the subsequence set W to be analyzed is obtained, and then in the subsequence set W to be analyzed, the subsequence to be analyzed corresponding to the largest fitting similarity characterization value is selected as the target subsequence of the subsequence set W to be analyzed, and compared with other subsequences in the subsequence set W to be analyzed, the selected target subsequence is less affected or interfered by external factors during acquisition. Therefore, in this embodiment, the initial smoothing coefficient of each subsequence in the subsequence set W to be analyzed is obtained by analyzing the target subsequence, and the initial smoothing coefficient is the key to the subsequent determination of the optimal smoothing coefficient; then in this embodiment, the process of obtaining the fitting similarity characterization value of each subsequence to be analyzed in the subsequence set W to be analyzed is:

[0036] For any subsequence B to be analyzed in the subsequence set W to be analyzed, the fitting values ​​of each environmental parameter in the subsequence B to be analyzed are obtained, and then the fitting similarity characterization value of the subsequence B to be analyzed is obtained according to the fitting values ​​of each environmental parameter in the subsequence B to be analyzed. The specific process of obtaining the fitting similarity characterization value of the subsequence B to be analyzed according to the fitting values ​​of each environmental parameter in the subsequence B to be analyzed is as follows: first, a sequence constructed by the fitting values ​​of each environmental parameter in the subsequence B to be analyzed is recorded as the fitting value sequence of the subsequence B to be analyzed, and the b1th fitting value in the fitting value sequence is the fitting value of the b1th environmental parameter in the subsequence B to be analyzed; then, a first difference sequence is obtained according to the fitting value sequence of the subsequence B to be analyzed and the cth first difference in the first difference sequence. is the absolute value of the difference between the cth environmental parameter in the subsequence B to be analyzed and the cth fitting value in the fitting value sequence; then the mean of the difference sequence corresponding to the subsequence B to be analyzed is obtained, and recorded as the first difference mean; then the mean of the difference sequence corresponding to the fitting value sequence of the subsequence B to be analyzed is obtained, and recorded as the second difference mean; then the mean of the first difference sequence is obtained, and the sum of the mean of the first difference sequence and the preset first constant is recorded as the first characterization value; then the absolute value of the difference between the first difference mean and the second difference mean is obtained, and recorded as the mean difference; then the sum of the mean difference and the preset first constant is obtained, and recorded as the second characterization value; then the mean after adding the reciprocal of the first characterization value and the reciprocal of the second characterization value is recorded as the fitting similarity characterization value corresponding to the subsequence B to be analyzed.

[0037] In this embodiment, the method for obtaining the fitting values ​​of each environmental parameter in the subsequence B to be analyzed is: for the b-th environmental parameter in the subsequence B to be analyzed, on the target fitting curve, obtain the data point whose abscissa value is the same as the collection time of the b-th environmental parameter, and record it as the fitting data point of the b-th environmental parameter, and then record the ordinate value of the fitting data point of the b-th environmental parameter as the fitting value of the b-th environmental parameter.

[0038] In addition, the specific expression for obtaining the fitting similarity characterization value corresponding to the subsequence B to be analyzed is:

[0039]

[0040] in, is the fitting similarity representation value corresponding to the subsequence B to be analyzed, is the total number of parameters in the subsequence B to be analyzed or the total number of fitted values ​​in the fitted value sequence of the subsequence B to be analyzed, is the bth environmental parameter in the subsequence B to be analyzed, is the bth fitting value in the fitting value sequence of the subsequence B to be analyzed, is the bth first difference in the first difference sequence, is the total number of parameters in the differential sequence corresponding to the subsequence B to be analyzed, is the jth difference value in the difference sequence corresponding to the subsequence B to be analyzed, is the total number of parameters in the difference sequence corresponding to the fitted value sequence, is the ith difference value in the difference sequence corresponding to the fitting value sequence, c1 is a preset first constant, and in this embodiment, the value of c1 is 0.01, and c1 is added to prevent the denominator from being 0.

[0041] And when as well as The smaller the value of The larger the value of The larger the value of is, the more similar the subsequence B to be analyzed is to the fitted value sequence, or the more similar the change trend between the subsequence B to be analyzed is to the fitted value sequence. as well as The larger the value of The smaller the value of The smaller the value of , the less similar the subsequence B to be analyzed is to the fitted value sequence, or the less similar the change trend between the subsequence B to be analyzed is to the fitted value sequence.

[0042] Therefore, this embodiment can obtain the target subsequence of each subsequence set to be analyzed through the above process of obtaining the target subsequence of the subsequence set to be analyzed W, and after obtaining the target subsequence corresponding to the subsequence set to be analyzed, the curvature representation value corresponding to each subsequence in the subsequence set to be analyzed is then obtained, and the curvature representation value is an important parameter for subsequently determining the initial smoothing coefficient of each subsequence to be analyzed. Therefore, the process of obtaining the curvature representation value corresponding to the subsequence is as follows:

[0043] For the subsequence B to be analyzed in the subsequence set W to be analyzed: first, according to each environmental parameter in the subsequence B to be analyzed and the acquisition time of each environmental parameter, a curve corresponding to the subsequence B to be analyzed is drawn and recorded as the curve to be analyzed of the subsequence B to be analyzed, then the curvature of each data point on the curve to be analyzed is obtained, and the mean curvature of all data points on the curve to be analyzed is used as the curvature representation value corresponding to the subsequence B to be analyzed; and the method for obtaining the curve to be analyzed of the subsequence B to be analyzed is: mapping each environmental parameter in the subsequence B to be analyzed and the acquisition time of each environmental parameter is mapped to a two-dimensional mapping space to obtain the data point corresponding to each environmental parameter in the subsequence B to be analyzed, and then connecting the data points corresponding to the environmental parameters in the subsequence B to be analyzed in order from small to large in the horizontal coordinate value, and then the connected curve is the curve to be analyzed of the subsequence B to be analyzed; and the method for obtaining the curvature of the data point is a well-known technology, so it will not be described in detail in this embodiment.

[0044] Therefore, through the above process, the curvature characterization value corresponding to each subsequence in the set of subsequences to be analyzed can be obtained, including the curvature characterization value corresponding to the target subsequence. After obtaining the corresponding curvature characterization value, this embodiment obtains the initial smoothing coefficient of each subsequence to be analyzed in the set of subsequences to be analyzed according to the differential sequence and curvature characterization value corresponding to the target subsequence in the set of subsequences to be analyzed and the differential sequence and curvature characterization value corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed. The initial smoothing coefficient is the key to subsequently determining the target smoothing coefficient of the set of subsequences to be analyzed, and the target smoothing coefficient is the key to finally determining the optimal smoothing coefficient. Therefore, the process of obtaining the initial smoothing coefficient of the subsequence to be analyzed is as follows:

[0045] First, the change trend characterization value and the volatility characterization value corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed are obtained, and the change trend characterization value and the volatility characterization value corresponding to the subsequence to be analyzed can reflect the trend change and fluctuation of the corresponding subsequence to be analyzed; and in this embodiment, the change trend characterization value and the volatility characterization value corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed are obtained as follows: for subsequence B to be analyzed, the reciprocal of the value obtained by adding the mean of the differential sequence corresponding to subsequence B to be analyzed and a preset first constant is obtained, and recorded as the first eigenvalue, and then the value obtained by subtracting the first eigenvalue from the preset second constant is recorded as the change trend characterization value corresponding to subsequence B to be analyzed, the reciprocal of the curvature characterization value corresponding to subsequence B to be analyzed is obtained, and recorded as the second eigenvalue, and then the value obtained by subtracting the second eigenvalue from the preset second constant is recorded as the volatility characterization value corresponding to subsequence B to be analyzed.

[0046] And because the target subsequence of the subsequence set to be analyzed belongs to the corresponding subsequence set to be analyzed, the change trend characterization value and volatility characterization value corresponding to the target subsequence can also be obtained; then, the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set to be analyzed is obtained according to the change trend characterization value and volatility characterization value corresponding to the target subsequence of the subsequence set to be analyzed; and in this embodiment, the process of obtaining the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set to be analyzed according to the change trend characterization value and volatility characterization value corresponding to the target subsequence of the subsequence set to be analyzed is: for any subsequence set W to be analyzed, the mean of the change trend characterization value and volatility characterization value corresponding to the target subsequence of the subsequence set W to be analyzed is recorded as the mean of the characterization value of the subsequence set to be analyzed, and the value obtained by subtracting the mean of the characterization value of the subsequence set W to be analyzed from the preset second constant is recorded as the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set W to be analyzed. In addition, the specific process of obtaining the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set W to be analyzed is:

[0047]

[0048] in, is the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set W to be analyzed, c2 is a preset second constant, and in this embodiment, the value of the preset second constant is 1, is the change trend representation value corresponding to the target subsequence of the subsequence set W to be analyzed, is the volatility characterization value corresponding to the target subsequence of the subsequence set W to be analyzed; and when and The smaller the value of is, the less the target subsequence is affected by external factors or the less interference it is when it is collected. When the target subsequence is affected by external factors or the less interference it is when it is collected, the larger the feature smoothing coefficient corresponding to the target subsequence is required to be. and The larger the value of , the greater the impact or interference degree of the target subsequence during collection by external factors. When the target subsequence is affected by external factors or the degree of interference during collection, the smaller the feature smoothing coefficient corresponding to the target subsequence is required to be. And because the smaller the smoothing coefficient, the higher the weight of the prediction model will be given to the historical data during prediction, so this embodiment requires a smaller smoothing coefficient when the impact or interference degree of external factors is greater.

[0049] After obtaining the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set to be analyzed, the initial smoothing coefficient of each subsequence to be analyzed in the subsequence set to be analyzed is obtained according to the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set to be analyzed, the change trend representation value and the volatility representation value corresponding to the target subsequence, and the change trend representation value and the volatility representation value corresponding to each subsequence to be analyzed in the subsequence set to be analyzed; and for ease of understanding, this embodiment will be described later by taking the process of obtaining the initial smoothing coefficient of the subsequence B to be analyzed in the subsequence set W to be analyzed as an example, then the process of obtaining the initial smoothing coefficient of the subsequence B to be analyzed is:

[0050] First, the change trend representation value, volatility representation value and characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set W to be analyzed are recorded as the basic change trend representation value, the basic volatility representation value and the basic characteristic smoothing coefficient respectively; then the absolute value of the difference between the change trend representation value corresponding to the subsequence B to be analyzed and the basic change trend representation value is obtained, and recorded as the first absolute value of the difference, and the ratio of the first absolute value of the difference to the basic change trend representation value is obtained, and recorded as the first ratio; then the absolute value of the difference between the volatility representation value corresponding to the subsequence B to be analyzed and the basic volatility representation value is obtained, and recorded as the second absolute value of the difference, and the ratio of the second absolute value of the difference to the basic volatility representation value is obtained, and recorded as the second ratio; then the mean of the first ratio and the second ratio is obtained, and recorded as the ratio mean, and then the product of the ratio mean and the basic characteristic smoothing coefficient is obtained, and recorded as the first product, and then the value obtained by subtracting the first product from the basic characteristic smoothing coefficient is recorded as the initial smoothing coefficient of the subsequence B to be analyzed.

[0051] In addition, in this embodiment, the specific expression for obtaining the initial smoothing coefficient of the subsequence B to be analyzed is:

[0052]

[0053] in, is the initial smoothing coefficient of the subsequence B to be analyzed, is the change trend representation value corresponding to the subsequence B to be analyzed, is the volatility representation value corresponding to the subsequence B to be analyzed; and when and The larger the value of The smaller the value, the greater the influence or interference of external factors on the subsequence to be analyzed during collection. The larger the value of .

[0054] Therefore, this embodiment can obtain the initial smoothing coefficient of each subsequence to be analyzed in each subsequence set to be analyzed through the above process. After obtaining the initial smoothing coefficient of each subsequence to be analyzed, the optimal smoothing coefficient of the environmental parameter sequence is obtained according to the initial smoothing coefficient of each subsequence to be analyzed. The specific process of obtaining the optimal smoothing coefficient of the environmental parameter sequence is:

[0055] First, the target smoothing coefficient of each subsequence set to be analyzed is obtained. For ease of understanding, this embodiment takes the process of obtaining the target smoothing coefficient of the subsequence set W to be analyzed as an example for description, that is, the process of obtaining the target smoothing coefficient of the subsequence set W to be analyzed is:

[0056] First, a new sequence formed by all subsequences in the subsequence set W to be analyzed is recorded as the original subsequence; then, among the initial smoothing coefficients of all subsequences to be analyzed in the subsequence set W to be analyzed, the largest initial smoothing coefficient and the smallest initial smoothing coefficient are selected, and the coefficient interval formed by the selected largest initial smoothing coefficient and the smallest initial smoothing coefficient is recorded as the smoothing coefficient interval to be selected corresponding to the subsequence set W to be analyzed; then, the smoothing coefficient interval to be selected is segmented using the preset first coefficient value, and the sequence constructed by the smoothing coefficients at all segmented positions is recorded as the smoothing coefficient sequence to be determined; for example, if the preset first coefficient value is 0.05, and the minimum value in the smoothing coefficient interval to be selected is 0.05 and the maximum value is 0.25, then the smoothing coefficient sequence to be determined is {0.05, 0.10, 0.15, 0.20, 0.25}.

[0057] Then, a smoothed data sequence corresponding to each smoothing coefficient in the smoothing coefficient sequence to be determined is obtained, and the smoothed data sequence corresponding to the f-th smoothing coefficient in the smoothing coefficient sequence to be determined refers to the sequence obtained after smoothing the original subsequence with the f-th smoothing coefficient; then, the mean square error characterization value corresponding to each smoothing coefficient in the smoothing coefficient sequence to be determined is obtained, and the mean square error characterization value corresponding to the f-th smoothing coefficient refers to the mean square error value between the original subsequence and the smoothed data sequence corresponding to the f-th smoothing coefficient; then, in the smoothing coefficient sequence to be determined, the smoothing coefficient to be determined corresponding to the smallest mean square error characterization value is selected as the smoothing coefficient to be analyzed; then, the sum of the smoothing coefficient to be analyzed and the preset second coefficient value is taken as the first smoothing coefficient, and the difference between the smoothing coefficient to be analyzed and the preset second coefficient value is taken as the first characteristic coefficient, and the mean square error characterization value corresponding to the first smoothing coefficient and the mean square error characterization value corresponding to the first characteristic coefficient are obtained.

[0058] Then, it is determined whether the mean square error characterization value corresponding to the smoothing coefficient to be analyzed is the smallest among the mean square error characterization values ​​corresponding to the first smoothing coefficient, the smoothing coefficient to be analyzed and the first characteristic coefficient. If so, the smoothing coefficient to be analyzed is directly used as the target smoothing coefficient of the subsequence set W to be analyzed.

[0059] If it is determined that among the mean square error characterization values ​​corresponding to the first smoothing coefficient, the smoothing coefficient to be analyzed and the first characteristic coefficient, the mean square error characterization corresponding to the first smoothing coefficient is the smallest, then the sum of the first smoothing coefficient and the preset second coefficient value is used as the second smoothing coefficient, and the mean square error characterization value corresponding to the second smoothing coefficient is obtained, and then it is further determined whether the mean square error characterization value corresponding to the first smoothing coefficient is the smallest among the mean square error characterization values ​​corresponding to the first smoothing coefficient and the second smoothing coefficient. If so, the first smoothing coefficient is directly used as the target smoothing coefficient of the subsequence set W to be analyzed. Otherwise, the sum of the second smoothing coefficient and the preset second coefficient value is used as the third smoothing coefficient, and the mean square error characterization value corresponding to the third smoothing coefficient is obtained. Then, it is further determined whether the mean square error characterization value corresponding to the second smoothing coefficient is the smallest among the mean square error characterization values ​​corresponding to the second smoothing coefficient and the third smoothing coefficient. By analogy, the target smoothing coefficient of the subsequence set W to be analyzed is obtained.

[0060] However, if it is determined that among the mean square error characterization values ​​corresponding to the first smoothing coefficient, the smoothing coefficient to be analyzed and the first characteristic coefficient, the mean square error characterization corresponding to the first characteristic coefficient is the smallest, then the difference between the first characteristic coefficient and the preset second coefficient value is taken as the second characteristic coefficient, and the mean square error characterization value corresponding to the second characteristic coefficient is obtained, and then it is further determined whether among the mean square error characterization values ​​corresponding to the first characteristic coefficient and the second characteristic coefficient, the mean square error characterization value corresponding to the first characteristic coefficient is the smallest. If so, the first characteristic coefficient is directly used as the target smoothing coefficient of the subsequence set W to be analyzed. Otherwise, the difference between the second characteristic coefficient and the preset second coefficient value is taken as the third characteristic coefficient, and the mean square error characterization value corresponding to the third characteristic coefficient is obtained. Then, it is further determined whether among the mean square error characterization values ​​corresponding to the second characteristic coefficient and the third characteristic coefficient, the mean square error characterization value corresponding to the second characteristic coefficient is the smallest. And so on, the target smoothing coefficient of the subsequence set W to be analyzed is obtained.

[0061] For example, if the first smoothing coefficient is 0.56, the smoothing coefficient to be analyzed is 0.55, and the first characteristic coefficient is 0.54, and among the mean square error characterization values ​​corresponding to the first smoothing coefficient, the smoothing coefficient to be analyzed and the first characteristic coefficient, the mean square error characterization value corresponding to the first characteristic coefficient is the smallest, and if the preset second coefficient value at this time is 0.01, then it is necessary to obtain the mean square error characterization value when the smoothing coefficient is 0.53, and compare the mean square error characterization value when the smoothing coefficient is 0.53 with the mean square error characterization value corresponding to the first characteristic coefficient, if the mean square error characterization value corresponding to the first characteristic coefficient is still smaller, then the first characteristic coefficient is the target smoothing coefficient of the subsequence set W to be analyzed. Coefficient, if not, compare the mean square error characterization value when the smoothing coefficient is 0.52 and the mean square error characterization value corresponding to the smoothing coefficient is 0.53, and so on, to determine the target smoothing coefficient of the subsequence set W to be analyzed; and the mean square error characterization value when the smoothing coefficient is 0.53 also refers to the mean square error value between the original subsequence and the smoothed data sequence corresponding to the smoothing coefficient of 0.53, and the smoothed data sequence corresponding to the smoothing coefficient of 0.53 refers to the sequence obtained after smoothing the original subsequence with a smoothing coefficient of 0.53; that is, the mean square error value and the acquisition process of the smoothed data sequence in this embodiment are both well-known technologies, so this embodiment will not be described in detail.

[0062] In a specific application, the implementer needs to set the preset first coefficient value and the preset second coefficient value according to the actual situation. For example, in this embodiment, the preset first coefficient value and the preset second coefficient value can be set to 0.05 and 0.01 respectively.

[0063] Therefore, this embodiment can obtain the target smoothing coefficient of each subsequence set to be analyzed through the above process, and then use the mean value of the target smoothing coefficients of all subsequence sets to be analyzed as the optimal smoothing coefficient of the environmental parameter sequence.

[0064] Step S003, using the optimal smoothing coefficient to smooth the environmental parameter sequence, and recording the sequence obtained after smoothing as the target smoothed sequence, and using the quadratic exponential smoothing algorithm with the smoothing coefficient as the optimal smoothing coefficient to predict the smoothed sequence, to obtain the predicted data at the future moment, and according to the predicted data, to obtain the monitoring result at the current monitoring moment.

[0065] After obtaining the optimal smoothing coefficient of the environmental parameter sequence, the environmental parameter sequence is immediately smoothed using the optimal smoothing coefficient, and the sequence obtained after smoothing is recorded as the target smoothed sequence, and then the smoothed sequence is predicted using the quadratic exponential smoothing algorithm with the smoothing coefficient as the optimal smoothing coefficient, to obtain the predicted data at the future moment, wherein the predicted data at the future moment refers to the predicted data at the next monitoring moment, and then the monitoring result at the current monitoring moment is obtained according to the predicted data; and according to the predicted data, the process of obtaining the monitoring result at the current monitoring moment is: if it is judged that the predicted data is not within the preset environmental parameter interval, then it is judged that the environment in the fermentation container at the current monitoring moment does not meet the development requirements of the bacillus bio-organic fertilizer; and the preset environmental parameter interval is related to the analyzed environmental parameter type, such as the environmental parameter sequence analyzed in the present embodiment is temperature, then the preset environmental parameter interval can be set to 15 to 45 degrees Celsius in the present embodiment, and if the environmental parameter sequence is temperature, when the predicted data is within the preset environmental parameter interval, it can only be judged that the temperature in the fermentation container at the current monitoring moment meets the development requirements of the bacillus bio-organic fertilizer.

[0066] In addition, using a smoothing coefficient to smooth a sequence and using a quadratic exponential smoothing algorithm with an optimal smoothing coefficient to predict a smoothed sequence to obtain prediction data at a future moment are both well-known techniques, and therefore will not be described in detail in this embodiment.

[0067] In summary, the present embodiment first obtains the environmental parameter sequence of the fermentation container; then, according to the preset sliding window and the environmental parameter sequence, obtains the set of subsequences to be analyzed and the target fitting curve corresponding to the environmental parameter sequence, and obtains the target subsequence of the subsequence set to be analyzed according to the target fitting curve and the differential sequence corresponding to each subsequence to be analyzed in the subsequence set to be analyzed, and then obtains the initial smoothing coefficient of each subsequence to be analyzed according to the differential sequence and curvature characterization value corresponding to the target subsequence of the subsequence set to be analyzed and the differential sequence and curvature characterization value corresponding to each subsequence to be analyzed in the subsequence set to be analyzed, and obtains the optimal smoothing coefficient of the environmental parameter sequence according to the initial smoothing coefficient of each subsequence to be analyzed; finally, the environmental parameter sequence is smoothed using the optimal smoothing coefficient, and the sequence obtained after smoothing is recorded as the target smoothed sequence, and the smoothed sequence is predicted using a quadratic exponential smoothing algorithm with the optimal smoothing coefficient as the smoothing coefficient to obtain the predicted data at the future moment, and the monitoring result at the current monitoring moment is obtained according to the predicted data. In addition, this embodiment reduces the interference of external factors on the accuracy and reliability of subsequent data prediction through the optimal smoothing coefficient, that is, by determining an optimal smoothing coefficient to improve the accuracy and reliability of subsequent data prediction, thereby also improving the timeliness and reliability of monitoring the environment in the fermentation container.

[0068] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. An environmental monitoring method for a bacillus bio-organic fertilizer production process, characterized in that: The method comprises the following steps: In the current monitoring time period, an environmental parameter sequence of a fermentation container is obtained, wherein the fermentation container is a container in which the Bacillus bio-organic fertilizer is located during the preparation process; According to the preset sliding window and the environmental parameter sequence, a set of subsequences to be analyzed and a target fitting curve corresponding to the environmental parameter sequence are obtained, and according to the target fitting curve and a differential sequence corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed, a target subsequence of the set of subsequences to be analyzed is obtained; Obtaining the curvature characterization value corresponding to the subsequence, and obtaining the initial smoothing coefficient of each subsequence to be analyzed according to the differential sequence and curvature characterization value corresponding to the target subsequence and the differential sequence and curvature characterization value corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed; obtaining the optimal smoothing coefficient of the environmental parameter sequence according to the initial smoothing coefficient of each subsequence to be analyzed; The environmental parameter sequence is smoothed using the optimal smoothing coefficient, and the sequence obtained after smoothing is recorded as a target smoothed sequence, and the smoothed sequence is predicted using a quadratic exponential smoothing algorithm whose smoothing coefficient is the optimal smoothing coefficient to obtain predicted data at a future moment, and the monitoring result at the current monitoring moment is obtained according to the predicted data; The method for obtaining the subsequence set to be analyzed and the target fitting curve includes: Sliding the preset sliding window on the environmental parameter sequence with a preset moving step length to obtain each window on the environmental parameter sequence, and recording the sequence formed by the data in each window on the environmental parameter sequence as a subsequence on the environmental parameter sequence; The set constructed by all subsequences on the environmental parameter sequence is recorded as the initial subsequence set, and the a-th subsequence in the initial subsequence set is the a-th subsequence on the environmental parameter sequence; Acquire all abnormal subsequences in the initial subsequence set, and remove all abnormal subsequences from the initial subsequence set, recording the subsequence set as the subsequence set to be divided; Determine whether the correlation coefficient between the second subsequence in the set of subsequences to be divided and the first subsequence in the set of subsequences to be divided is greater than a preset correlation coefficient threshold; if so, determine whether the correlation coefficient between the third subsequence in the set of subsequences to be divided and the second subsequence is greater than a preset correlation coefficient threshold; if not, record the sequence set constructed by the first subsequence and the second subsequence as the first subsequence set to be analyzed corresponding to the environmental parameter sequence; and continue to determine whether the correlation coefficient between the fourth subsequence in the set of subsequences to be divided and the third subsequence is greater than a preset correlation coefficient threshold. coefficient threshold, if yes, then determine whether the correlation coefficient between the 5th subsequence in the subsequence set to be divided and the 4th subsequence is greater than the preset correlation coefficient threshold, if not, then record the sequence set constructed by the 3rd subsequence and the 4th subsequence as the 2nd subsequence set to be analyzed corresponding to the environmental parameter sequence, and so on, until all subsequences in the subsequence set to be divided are traversed, and all subsequence sets to be analyzed corresponding to the environmental parameter sequence are statistically obtained, and all subsequences in the subsequence set to be analyzed are recorded as subsequences to be analyzed, and the correlation coefficient refers to the Spearman correlation coefficient; In the environmental parameter sequence, all environmental parameters belonging to the abnormal subsequence are removed, and a sequence constructed by all remaining environmental parameters after the removal is recorded as a sequence to be fitted, a curve fitting is performed on the sequence to be fitted, and a curve obtained by fitting is recorded as a target fitting curve; The method for obtaining the abnormal subsequence includes: For the ath subsequence in the initial subsequence set, if the correlation coefficients between the ath subsequence and the a-1th subsequence and the a+1th subsequence in the initial subsequence set are not greater than a preset correlation coefficient threshold, the ath subsequence is recorded as an abnormal subsequence, where a is not equal to 1 and A, and A is the total number of subsequences in the initial subsequence set; For the 1st subsequence and the Ath subsequence in the initial subsequence set, if the correlation coefficient between the 1st subsequence and the 2nd subsequence in the initial subsequence set is not greater than a preset correlation coefficient threshold, the 1st subsequence is recorded as an abnormal subsequence; if the correlation coefficient between the Ath subsequence and the A-1th subsequence in the initial subsequence set is not greater than the preset correlation coefficient threshold, the Ath subsequence is recorded as an abnormal subsequence; The method for obtaining a target subsequence of the set of subsequences to be analyzed comprises: For the b-th environmental parameter in any subsequence B to be analyzed in any subsequence set W to be analyzed, on the target fitting curve, a data point whose abscissa value is the same as the collection time of the b-th environmental parameter is obtained, and recorded as the fitting data point of the b-th environmental parameter, and the ordinate value of the fitting data point of the b-th environmental parameter is recorded as the fitting value of the b-th environmental parameter; Obtaining a fitting similarity characterization value of the subsequence to be analyzed according to the fitting values ​​of each environmental parameter in the subsequence to be analyzed; In any set of subsequences to be analyzed, the subsequence to be analyzed corresponding to the maximum fitting similarity representation value is used as the target subsequence of the corresponding set of subsequences to be analyzed; The method for obtaining the fitting similarity characterization value includes: For the subsequence B to be analyzed: a sequence constructed by fitting values ​​of each environmental parameter in the subsequence B to be analyzed is recorded as the fitting value sequence of the subsequence B to be analyzed; a first difference sequence is obtained according to the subsequence B to be analyzed and the fitting value sequence, wherein the cth first difference value in the first difference sequence is the absolute value of the difference between the cth environmental parameter in the subsequence B to be analyzed and the cth fitting value in the fitting value sequence; the mean of the difference sequence corresponding to the subsequence B to be analyzed is obtained and recorded as the first difference mean; the The mean of the difference sequence corresponding to the fitting value sequence is obtained, and recorded as the second difference mean; the mean of the first difference sequence is obtained, and the sum of the mean of the first difference sequence and a preset first constant is recorded as the first characterization value; the absolute value of the difference between the first difference mean and the second difference mean is recorded as the mean difference; the sum of the mean difference and the preset first constant is recorded as the second characterization value; the mean of the reciprocal of the first characterization value and the reciprocal of the second characterization value is added, and recorded as the fitting similarity characterization value corresponding to the subsequence B to be analyzed; The method for obtaining the curvature representation value corresponding to the subsequence to be analyzed includes: For the subsequence B to be analyzed: according to each environmental parameter in the subsequence B to be analyzed and the acquisition time of each environmental parameter, a curve corresponding to the subsequence B to be analyzed is plotted and recorded as the curve to be analyzed of the subsequence B to be analyzed; the curvature of each data point on the curve to be analyzed is obtained, and the mean curvature of all data points on the curve to be analyzed is taken as the curvature representation value corresponding to the subsequence B to be analyzed; The method for obtaining the initial smoothing coefficient of each subsequence to be analyzed includes: Obtaining a change trend representation value and a volatility representation value corresponding to each subsequence to be analyzed in the set of subsequences to be analyzed; Obtaining a characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set to be analyzed according to the change trend characterization value and the volatility characterization value corresponding to the target subsequence of the subsequence set to be analyzed, wherein the characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set to be analyzed refers to a value obtained by subtracting the mean of the characterization values ​​of the corresponding subsequence set to be analyzed from the preset second constant, and the mean of the characterization values ​​of the subsequence set to be analyzed refers to the mean of the change trend characterization value and the volatility characterization value corresponding to the target subsequence of the subsequence set to be analyzed; For the subsequence B to be analyzed: the change trend characterization value, volatility characterization value and characteristic smoothing coefficient corresponding to the target subsequence of the subsequence set W to be analyzed are respectively recorded as the basic change trend characterization value, the basic volatility characterization value and the basic characteristic smoothing coefficient; the absolute value of the difference between the change trend characterization value corresponding to the subsequence B to be analyzed and the basic change trend characterization value is recorded as the first difference absolute value; the ratio of the first difference absolute value to the basic change trend characterization value is recorded as the first ratio; the absolute value of the difference between the volatility characterization value corresponding to the subsequence B to be analyzed and the basic volatility characterization value is recorded as the second difference absolute value; the ratio of the second difference absolute value to the basic volatility characterization value is recorded as the second ratio; the mean of the first ratio and the second ratio is recorded as the ratio mean; the product of the ratio mean and the basic characteristic smoothing coefficient is recorded as the first product; the value obtained by subtracting the first product from the basic characteristic smoothing coefficient is recorded as the initial smoothing coefficient of the subsequence B to be analyzed; The method for obtaining the change trend characterization value and the volatility characterization value corresponding to the subsequence to be analyzed includes: For the subsequence B to be analyzed: the reciprocal of the value obtained by adding the mean of the differential sequence corresponding to the subsequence B to be analyzed to the preset first constant is recorded as the first eigenvalue, and the value obtained by subtracting the first eigenvalue from the preset second constant is recorded as the change trend representation value corresponding to the subsequence B to be analyzed; the reciprocal of the curvature representation value corresponding to the subsequence B to be analyzed is recorded as the second eigenvalue, and the value obtained by subtracting the second eigenvalue from the preset second constant is recorded as the volatility representation value corresponding to the subsequence B to be analyzed; The method for obtaining the optimal smoothing coefficient of the environmental parameter sequence includes: For the subsequence set W to be analyzed: The new sequence formed by all subsequences in the subsequence set W to be analyzed is recorded as the original subsequence; the largest initial smoothing coefficient and the smallest initial smoothing coefficient are selected from the initial smoothing coefficients of all subsequences to be analyzed in the subsequence set W to be analyzed, and the coefficient interval constructed by the selected largest initial smoothing coefficient and the smallest initial smoothing coefficient is recorded as the smoothing coefficient interval to be selected corresponding to the subsequence set W to be analyzed; the smoothing coefficient interval to be selected is segmented using a preset first coefficient value, and the sequence constructed by the smoothing coefficients at all segmented positions is recorded as the smoothing coefficient sequence to be determined, and the smoothing coefficient sequence to be determined is obtained. The smoothed data sequence corresponding to each smoothing coefficient in the coefficient sequence, the smoothed data sequence corresponding to the f-th smoothing coefficient in the smoothing coefficient sequence to be determined refers to the sequence obtained after smoothing the original subsequence with the f-th smoothing coefficient; obtaining the mean square error characterization value corresponding to each smoothing coefficient in the smoothing coefficient sequence to be determined, the mean square error characterization value corresponding to the f-th smoothing coefficient refers to the mean square error value between the original subsequence and the smoothed data sequence corresponding to the f-th smoothing coefficient; in the smoothing coefficient sequence to be determined, selecting the smoothing coefficient to be determined corresponding to the smallest mean square error characterization value as the smoothing coefficient to be analyzed; The sum of the smoothing coefficient to be analyzed and the preset second coefficient value is used as the first smoothing coefficient, and the difference between the smoothing coefficient to be analyzed and the preset second coefficient value is used as the first characteristic coefficient; obtaining a mean square error characterization value corresponding to the first smoothing coefficient and a mean square error characterization value corresponding to the first characteristic coefficient; If it is determined that among the mean square error characterization values ​​corresponding to the first smoothing coefficient, the smoothing coefficient to be analyzed and the first characteristic coefficient, the mean square error characterization value corresponding to the smoothing coefficient to be analyzed is the smallest, then the smoothing coefficient to be analyzed is directly used as the target smoothing coefficient of the subsequence set W to be analyzed; If it is determined that the mean square error representation corresponding to the first smoothing coefficient is the smallest among the mean square error representation values ​​corresponding to the first smoothing coefficient, the smoothing coefficient to be analyzed, and the first characteristic coefficient, then the sum of the first smoothing coefficient and the preset second coefficient value is used as the second smoothing coefficient, and it is further determined whether the mean square error representation value corresponding to the first smoothing coefficient is the smallest among the mean square error representation values ​​corresponding to the first smoothing coefficient and the second smoothing coefficient. If so, the first smoothing coefficient is directly used as the target smoothing coefficient of the subsequence set W to be analyzed. Otherwise, the sum of the second smoothing coefficient and the preset second coefficient value is used as the third smoothing coefficient, and the mean square error representation value corresponding to the third smoothing coefficient is obtained. It is further determined whether the mean square error representation value corresponding to the second smoothing coefficient is the smallest among the mean square error representation values ​​corresponding to the second smoothing coefficient and the third smoothing coefficient. By analogy, the target smoothing coefficient of the subsequence set W to be analyzed is obtained. If it is determined that the mean square error characterization corresponding to the first characteristic coefficient is the smallest among the mean square error characterization values ​​corresponding to the first smoothing coefficient, the smoothing coefficient to be analyzed, and the first characteristic coefficient, then the difference between the first characteristic coefficient and the preset second coefficient value is used as the second characteristic coefficient, and the mean square error characterization value corresponding to the second characteristic coefficient is obtained, and it is further determined whether the mean square error characterization value corresponding to the first characteristic coefficient is the smallest among the mean square error characterization values ​​corresponding to the first characteristic coefficient and the second characteristic coefficient. If so, the first characteristic coefficient is directly used as the target smoothing coefficient of the subsequence set W to be analyzed. Otherwise, the difference between the second characteristic coefficient and the preset second coefficient value is used as the third characteristic coefficient, and the mean square error characterization value corresponding to the third characteristic coefficient is obtained. It is further determined whether the mean square error characterization value corresponding to the second characteristic coefficient is the smallest among the mean square error characterization values ​​corresponding to the second characteristic coefficient and the third characteristic coefficient. By analogy, the target smoothing coefficient of the subsequence set W to be analyzed is obtained. The average of the target smoothing coefficients of all the subsequence sets to be analyzed is used as the optimal smoothing coefficient of the environmental parameter sequence.

2. the environmental monitoring method of a kind of bacillus bio-organic fertilizer development process as claimed in claim 1, is characterized in that, The method of obtaining the monitoring result at the current monitoring time according to the prediction data includes: If it is determined that the predicted data is not within the preset environmental parameter range, it is determined that the environment in the fermentation container at the current monitoring time does not meet the research and development requirements of the Bacillus bio-organic fertilizer.

Images