CN117746994B - Fungus stick maturity judging method based on data analysis - Google Patents
Fungus stick maturity judging method based on data analysis Download PDFInfo
- Publication number
- CN117746994B CN117746994B CN202410185781.XA CN202410185781A CN117746994B CN 117746994 B CN117746994 B CN 117746994B CN 202410185781 A CN202410185781 A CN 202410185781A CN 117746994 B CN117746994 B CN 117746994B
- Authority
- CN
- China
- Prior art keywords
- parameter data
- growth parameter
- fungus
- fungus stick
- obtaining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 241000233866 Fungi Species 0.000 title claims abstract description 241
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000007405 data analysis Methods 0.000 title claims abstract description 21
- 238000009499 grossing Methods 0.000 claims abstract description 72
- 238000009826 distribution Methods 0.000 claims abstract description 18
- 238000005457 optimization Methods 0.000 claims description 23
- 230000001580 bacterial effect Effects 0.000 claims description 16
- 241000894006 Bacteria Species 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 description 17
- 206010010356 Congenital anomaly Diseases 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012258 culturing Methods 0.000 description 1
- 239000003102 growth factor Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
- Apparatus Associated With Microorganisms And Enzymes (AREA)
Abstract
The invention relates to the technical field of growth parameter data prediction, in particular to a fungus stick maturity judging method based on data analysis. According to the method, the response coefficient of each fungus stick is obtained according to the overall distribution of fitting errors of each fungus stick at all historical moments and the overall trend difference degree distribution characteristics between each fungus stick and other fungus sticks in a neighborhood range; obtaining the noisiness of each growth parameter data at each historical time according to the difference characteristics between each growth parameter data and other growth parameter data, and further obtaining the constraint weight of each growth parameter data at each historical time; adjusting the initial exponential smoothing weight to obtain the optimized exponential smoothing weight of each growth parameter data at each historical time; and predicting the growth parameter data, and judging the maturity of the fungus stick according to the prediction result. According to the invention, by obtaining the proper exponential smoothing weight of the growth parameter data, the accuracy of the predicted value is improved, and the maturity is timely judged.
Description
Technical Field
The invention relates to the technical field of growth parameter data prediction, in particular to a fungus stick maturity judging method based on data analysis.
Background
In cultivating edible fungus products, the fungus culturing time of most fungus is far longer than the fungus producing time, after the physiological maturity of mycelia in the fungus sticks is reached, the mycelia are required to be moved into a grain shed for fruiting, so that whether the mycelia in the fungus sticks reach the physiological maturity moment or not is accurately judged, fruiting is timely carried out, the fungus sticks are ensured to be collected in the optimal time period, and the quality and the yield are ensured, so that the fruiting efficiency is remarkably improved.
In the prior art, related parameters in a fungus stick are generally collected through a sensor, and growth related parameter data are predicted through an input index smoothing algorithm, so that related maturity is judged according to the predicted data; however, due to the limitation of the sensor, the acquired data has interference of certain noise, if only the time sequence difference is considered, the data points have no proper exponential smoothing weight in the prediction process, and further the predicted value is inaccurate, so that the maturity of the fungus stick cannot be timely judged.
Disclosure of Invention
In order to solve the technical problems that the prediction result is inaccurate and the maturity cannot be timely judged due to the fact that proper exponential smoothing weights of data points cannot be determined, the invention aims to provide a fungus stick maturity judging method based on data analysis, and the adopted technical scheme is as follows:
The invention provides a fungus stick maturity judging method based on data analysis, which comprises the following steps:
Acquiring growth parameter data of each fungus stick at each historical moment;
According to the position of the historical moment on the time sequence, the difference of the corresponding growth parameter data between each fungus stick at each historical moment is adjusted, and the local trend difference degree is obtained; obtaining the overall trend difference degree of the corresponding growth parameter data among each fungus stick according to the local trend difference degree at all the historical moments; performing curve fitting on the corresponding growth parameter data in each fungus stick on a time sequence to obtain fitting errors between the corresponding growth parameter data and fitting results at each historical moment; obtaining a response coefficient of each fungus stick according to the overall distribution of the fitting errors of each growth parameter of each fungus stick at all historical moments and the distribution characteristics of the overall trend difference degree between each fungus stick and other fungus sticks in a neighborhood range;
Obtaining the noisiness of each growth parameter data of each fungus stick at each historical time according to the difference characteristics of the local trend difference degree between each growth parameter data and other growth parameter data between each fungus stick and other fungus sticks in the neighborhood range at each historical time;
Obtaining constraint weight of each growth parameter data of each fungus stick at each historical time according to the response coefficient of each fungus stick and the noisiness of each growth parameter data; acquiring initial exponential smoothing weights, and adjusting the initial exponential smoothing weights according to the constraint weights to acquire optimized exponential smoothing weights of each growth parameter data of each fungus stick at each historical time; predicting the growth parameter data according to the optimization index smoothing weight;
and judging the maturity of the fungus stick according to the prediction result.
Further, the method for obtaining the local trend difference degree comprises the following steps:
Calculating the relative distance of the corresponding growth parameter data between each fungus stick and other fungus sticks in the neighborhood range at each historical time based on a DTW algorithm, and taking the relative distance as the initial path cost value of the corresponding growth parameter data between each fungus stick at each historical time;
calculating the difference between each historical moment and the current moment, and normalizing the difference to be used as a time weighting factor;
Calculating the product of the initial path cost value and the time weighting factor to obtain the local trend difference degree of the corresponding growth parameter data between each fungus stick at each historical moment;
and the initial journey cost value and the time weighting factor are in positive correlation with the local trend difference degree.
Further, the method for obtaining the overall trend difference degree comprises the following steps:
and accumulating the local trend difference degrees under all the historical moments to obtain the overall trend difference degrees of the corresponding growth parameter data among each fungus stick.
Further, the method for acquiring the fitting error comprises the following steps:
and performing curve fitting on the corresponding growth parameter data of each fungus stick at all the historical moments by adopting a least square method, and calculating the difference value between each growth parameter data and the corresponding fitting result of each fungus stick at each historical moment to serve as a fitting error.
Further, the method for obtaining the response coefficient comprises the following steps:
Obtaining a response coefficient according to an obtaining formula of the response coefficient, wherein the obtaining formula of the response coefficient is as follows:
; wherein/> Represents the/>Response coefficients of the individual sticks; /(I)Represents the/>Individual fungus sticks and/>The bacterial stick is at the first placeOverall trend variability in the individual growth parameter data; /(I)Representing the number of bacteria sticks in the neighborhood range; /(I)Represents the/>Individual sticks and others/>The bacterial stick is at the first placeAn average of overall trend variability in the individual growth parameter data; /(I)Expressed in/>At the moment,/>First/>, of individual fungus sticksFitting errors of the growth parameter data and the fitting result; /(I)Indicating at all historic times, the/>In the individual fungus stick/>Accumulating fitting errors corresponding to the growth parameters, and calculating to obtain standard deviation; /(I)A number representing a growth parameter; /(I)The number of times of historic time is represented; /(I)Representing the normalization function.
Further, the method for obtaining the noisiness comprises the following steps:
Obtaining the noisiness according to an acquisition formula of the noisiness, wherein the acquisition formula of the noisiness is as follows:
; wherein/> Represents the/>At the moment,/>First/>, of individual fungus sticksNoisiness of the individual growth parameter data; /(I)Represents the/>At the moment,/>Individual fungus sticks and/>Corresponding/>, between individual fungus sticksLocal trend variability of the individual growth parameter data; /(I)Represents the/>At the moment,/>Individual fungus sticks and/>Corresponding/>, between individual fungus sticksLocal trend variability of the individual growth parameter data; /(I)A number representing a growth parameter; /(I)The number of bacteria sticks in the neighborhood range is indicated.
Further, the constraint weight acquisition method includes:
obtaining constraint weights according to the constraint weight obtaining formula, wherein the constraint weight obtaining formula is as follows:
; wherein/> Represents the/>At the moment,/>The/>Constraint weights of the individual growth parameter data; /(I)Represents the/>Response coefficients of the individual sticks; /(I)Represents the/>At the moment,/>The first of the fungus sticksNoisiness of the individual growth parameter data; /(I)Representing a preset adjustment factor.
Further, the method for obtaining the optimization index smoothing weight comprises the following steps:
obtaining an optimization index smoothing weight according to an acquisition formula of the optimization index smoothing weight, wherein the acquisition formula of the optimization index smoothing weight is as follows:
; wherein/> Represents the/>At the moment,/>The/>Optimizing exponential smoothing weights of the individual growth parameter data; /(I)Indicating a preset super parameter in an exponential smoothing algorithm; /(I)A serial number indicating a history time; /(I)Represents the/>At the moment,/>The/>Constraint weights of the individual growth parameter data; /(I)Representing a logistic function.
Further, the predicting the growth parameter data according to the optimization exponential smoothing weights includes:
and carrying out an exponential smoothing algorithm based on the optimized exponential smoothing weight of each growth parameter data at the historical moment to obtain the growth parameter data at the next moment.
Further, the preset adjustment factor is 0.5.
The invention has the following beneficial effects:
according to the method, the difference of corresponding growth parameter data between each fungus stick at each historical moment is adjusted according to the position of the historical moment on the time sequence, so that the local trend difference degree is obtained; the overall trend difference degree between each fungus stick is further obtained, different weights are given to the data at different historical moments, the data difference at different historical moments can be balanced, the influence of the congenital difference in the fungus sticks is reduced, and the growth state and trend of each fungus stick are accurately estimated; performing curve fitting on the corresponding growth parameter data in each fungus stick in time sequence to obtain fitting errors between the growth parameter data and fitting results at each historical time, finding abnormal data in the fungus stick, and judging the possibility that the corresponding growth parameters are interfered by noise; obtaining the response coefficient of each fungus stick, obtaining the overall ratio condition of abnormal data points in the fungus stick, and judging the subsequent adjustment direction of the index smoothing weight; in order to obtain a more accurate prediction result, carrying out local analysis on each growth parameter data, and obtaining the noisiness of each growth parameter data of each fungus stick at each historical time according to the difference characteristics of local trend difference degrees between each fungus stick and other fungus sticks in a neighborhood range; further, constraint weight of each growth parameter data of each fungus stick at each historical time is obtained, noise interference is reduced, and quality and reliability of data are improved; the initial exponential smoothing weight is adjusted to obtain the optimal exponential smoothing weight of each growth parameter data of each fungus stick at each historical time, so that the method is better suitable for the change of the data and reduces the influence of noise on a prediction result; and predicting the growth parameter data, judging the maturity of the fungus stick according to the prediction result, and timely finding out abnormal data to improve the management efficiency. According to the invention, by obtaining the proper exponential smoothing weight of the growth parameter data, the accuracy of the predicted value is improved, and the maturity is timely judged.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for determining the maturity of a fungus stick based on data analysis according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of a fungus stick maturity judging method based on data analysis according to the invention, and the detailed description is given below of the specific implementation, structure, characteristics and effects thereof. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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 belongs.
The invention provides a bacterial stick maturity judging method based on data analysis, which is specifically described below with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for determining a maturity of a fungus stick based on data analysis according to an embodiment of the present invention is shown, where the method includes:
Step S1: and acquiring growth parameter data of each fungus stick at each historical time.
In the embodiment of the invention, in order to judge the maturity of the fungus stick, the related growth parameters inside the fungus stick need to be analyzed; firstly, monitoring relevant growth parameter data of fungus sticks through a plurality of types of sensors, and acquiring the growth parameter data of each fungus stick at each historical moment, wherein the growth parameter data in a local range monitored by the sensors comprises: height, color, temperature and humidity, bacterial stick self-growth time, etc.
It should be noted that, the growth parameter data are all data points rather than data curves, so as to facilitate the subsequent data processing, avoid the difference between the units and the magnitude of the values of the growth parameter data, and normalize all the obtained growth parameter data. In one embodiment of the invention, the Z-score algorithm may be used to normalize the acquired data; in other embodiments of the present invention, the normalization method may be constructed by other basic mathematical operations, and the specific normalization method is a technical means well known to those skilled in the art, and is not limited and described herein.
Step S2: according to the position of the historical moment on the time sequence, the difference of the corresponding growth parameter data between each fungus stick at each historical moment is adjusted, and the local trend difference degree is obtained; obtaining the overall trend difference of the corresponding growth parameter data among each fungus stick according to the local trend difference at all the historical moments; performing curve fitting on the corresponding growth parameter data in each fungus stick on a time sequence to obtain fitting errors between the corresponding growth parameter data and fitting results at each historical moment; and obtaining the response coefficient of each fungus stick according to the overall distribution of the fitting error of each growth parameter of each fungus stick at all historical moments and the distribution characteristics of the overall trend difference degree between each fungus stick and other fungus sticks in the neighborhood range.
In the same area, the relevant growing environments of the fungus sticks are similar, other relevant information is the same except for the difference of hyphae contained in the fungus sticks, and the difference of the internal hyphae is larger and larger along with the growth of the fungus sticks, so that the influence of the growth time of the fungus sticks on the difference between the fungus sticks is considered, the difference of corresponding growth parameter data between each fungus stick at each historical moment is adjusted according to the position of the historical moment on the time sequence, and the local trend difference degree is obtained; the time sequence information is considered, so that the variation difference of the whole data points of each fungus stick can be reflected better. Meanwhile, due to the consideration of the time span, abnormal data points can be better processed, misjudgment caused by congenital differences is avoided, the overall trend difference degree among all bacteria sticks is obtained according to the local trend difference degrees at all historical moments, and the growth condition of each bacteria stick is more accurately reflected.
Preferably, in one embodiment of the present invention, the method for obtaining the local trend difference degree includes:
calculating the relative distance of corresponding growth parameter data between each fungus stick and other fungus sticks in the neighborhood range at each historical time based on a DTW algorithm, and taking the relative distance as the initial path cost value of the corresponding growth parameter data between each fungus stick at each historical time; calculating the difference between each historical moment and the current moment, and normalizing the difference to be used as a time weighting factor; and calculating the product of the initial path cost value and the time weighting factor, and obtaining the local trend difference degree of the corresponding growth parameter data between each fungus stick at each historical moment. The larger the initial path cost value, the larger the time weighting factor, and the larger the local trend difference degree among bacteria sticks. In one embodiment of the invention, the formula for the local trend discrepancy gauge is expressed as:
；
wherein, Represents the/>At the moment,/>Individual fungus sticks and/>Corresponding/>, between individual fungus sticksLocal trend variability of the individual growth parameter data; /(I)Represents the/>Individual fungus sticks and/>Between the individual fungus sticks at the/>Corresponding to the/>, at each momentInitial path cost values of the individual growth parameter data; /(I)Represents the/>Bacterial stick No./>The time difference between the growth parameter data at each moment and the current moment; /(I)Representing natural constants.
In the formula of the local trend difference, the larger the relative distance of corresponding growth parameter data between fungus sticks is, the larger the initial path cost value is, and the larger the local trend difference is; is obtained by an exponential function based on natural constant The greater the interval between each historical moment and the current time, the more self-characteristics the fungus sticks can exhibit, the greater the difference between the fungus sticks at that moment.
Preferably, in one embodiment of the present invention, the method for acquiring the overall trend difference degree includes:
and accumulating the local trend difference degrees at all the historical moments to obtain the overall trend difference degrees of the corresponding growth parameter data among each fungus stick. In one embodiment of the invention, the overall trend discrepancy gauge is formulated as:
；
wherein, Represents the/>Individual fungus sticks and/>The bacterial stick is at the first placeOverall trend variability when DTW is performed in the individual growth parameter data; /(I)Represents the/>Individual fungus sticks and/>Between the individual fungus sticks at the/>Corresponding to the/>, at each momentInitial path cost values of the individual growth parameter data; /(I)Represents the/>Bacterial stick No./>The time difference between the growth parameter data at each moment and the current moment; /(I)Representing natural constants.
In the formula of the overall trend difference degree, the larger the local trend difference degree between the fungus sticks and other fungus sticks is, the larger the overall trend difference degree of the fungus sticks is, and the larger the difference between the fungus sticks is.
In one embodiment of the present invention, each fungus stick is adjacent to the same areaThe individual bacteria sticks form a neighborhood range; in other embodiments of the present invention, the size of the neighborhood range may be specifically set according to specific situations, which is not limited and described herein.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation and normalization method may be constructed by other basic mathematical operations, and specific means are technical means well known to those skilled in the art, and will not be described herein. The specific DTW algorithm is a technical means well known to those skilled in the art, and will not be described herein.
Because the sensor is susceptible to noise when collecting data, the sensor is inaccurate only in consideration of the change condition of the time sequence characteristics; when the whole is accumulated, the degree that some growth parameter data may be affected by the outside is different, for example, the height, the size and the like of the fungus sticks are hardly affected by the outside, and the growth parameters such as temperature and humidity are affected by the outside more, for the situation, the difference among the fungus sticks is relatively higher in some growth parameters, in order to accurately analyze the abnormal situation existing in the growth parameter data and the possibility of being affected by the outside factors, curve fitting is performed on the corresponding growth parameter data in each fungus stick on time sequence, and the fitting error between the corresponding growth parameter data and the fitting result at each historical moment is obtained.
Preferably, in one embodiment of the present invention, the method for obtaining the fitting error includes:
In order to understand the distribution and variation trend of each growth parameter, the abnormal conditions in the fungus sticks are analyzed, curve fitting is carried out on the growth parameter data corresponding to each fungus stick at all historical moments by adopting a least square method, and the difference between each growth parameter data and the corresponding fitting result of each fungus stick at each historical moment is calculated and used as a fitting error. It should be noted that, the specific least square method is a technical means well known to those skilled in the art, and will not be described herein.
Considering that the larger the fitting error between each growth parameter data and the fitting curve, the more likely the growth parameter data is an abnormal data point; in order to judge whether the abnormal data points are due to the influence of self growth factors or external noise, analyzing the overall distribution and the change trend of the data under all growth parameters of the fungus stick, wherein the more uniform the change among the growth parameters is, the more likely the abnormal state is caused by the growth of the fungus stick is; the time sequence of the change between the fungus sticks is similar in trend under normal conditions, and noise can interfere with the feature, so that based on the information, the change of the data points under the growth parameters of each fungus stick can be analyzed. And obtaining the response coefficient of each fungus stick according to the overall distribution of fitting errors of each fungus stick at all historical moments and the distribution characteristics of overall trend difference degrees between each fungus stick and other fungus sticks in the neighborhood range.
Preferably, in one embodiment of the present invention, the method for acquiring a response coefficient includes:
Obtaining a response coefficient according to an obtaining formula of the response coefficient, wherein the obtaining formula of the response coefficient is as follows:
；
wherein, Represents the/>Response coefficients of the individual sticks; /(I)Represents the/>Individual fungus sticks and/>The bacterial stick is at the first placeOverall trend variability in the individual growth parameter data; /(I)Representing the number of bacteria sticks in the neighborhood range; /(I)Represents the/>Individual fungus sticks and othersThe bacterial stick is at the first placeAn average of overall trend variability in the individual growth parameter data; /(I)Expressed in/>At the moment, the firstFirst/>, of individual fungus sticksFitting errors of the growth parameter data and the fitting result; /(I)Indicating at all historic times, the/>In the individual fungus stick/>Accumulating fitting errors corresponding to the growth parameters, and calculating to obtain standard deviation; /(I)A number representing a growth parameter; /(I)The number of times of historic time is represented; /(I)Representing the normalization function.
In the acquisition formula of the response coefficient,Represents the/>Individual fungus sticks and others/>The variance of the overall trend difference degree of each bacterial stick under all growth parameters, the larger the variance is, the larger the variation of the growth parameter data change trend of the bacterial stick is represented to be different from the other bacterial sticks, the higher the possibility that the growth parameter data interfered by noise exists in the corresponding bacterial stick is, and the larger the weight adjusting range of each growth parameter data is when the index is smooth; /(I)Represents the/>The first/>, of the individual fungus sticks at all historic timesAccumulating the fitting errors of the growth parameter data and the fitting results; the higher the fitting error accumulation sum is, the more abnormal data points of the current fungus stick in the growth parameters are indicated, and the larger the weight adjusting range is when the corresponding indexes are smooth; /(I)Represents the/>Calculated/>, of individual fungus sticksThe smaller the standard deviation between the fitting error summations of the individual growth parameters, the larger the weight adjustment range when the stick is exponentially smoothed, which indicates that the abnormal data points of the stick are mainly represented by the differences of the stick itself, but not by noise. The more abnormal data points exist inside the fungus stick, and the more the proportion of the fungus stick is affected by noiseResponse coefficient of individual fungus stick, namely/>The larger the corresponding weight adjustment range when performing exponential smoothing is, the smaller the weight can be used to attenuate the noise interference.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation and normalization method may be constructed by other basic mathematical operations, and specific means are technical means well known to those skilled in the art, and will not be described herein.
Step S3: and under each historical time, obtaining the noisiness of each growth parameter data of each fungus stick under each historical time according to the difference characteristics of the local trend difference degree between each growth parameter data and other growth parameter data between each fungus stick and other fungus sticks in the neighborhood range.
According to the response coefficient of each fungus stick, analyzing the difference between the whole data point of each fungus stick and the rest fungus sticks, and determining the whole duty ratio and the noise receiving degree of the abnormal data point in each fungus stick; however, in order to process the growth parameter data more accurately, the mutual relationship between the fungus sticks needs to be known by analyzing the local trend difference degree between the fungus sticks, so that the growth dynamics of the fungus sticks can be better understood, and the more the difference of the local trend difference degree between the two growth parameter data is larger, the more abnormal conditions are likely to exist, and the more unstable the growth state is; and carrying out local analysis on each growth parameter data, and obtaining the noisiness of each growth parameter data of each fungus stick at each historical time according to the difference characteristics of the local trend difference degree between each growth parameter data and other growth parameter data between each fungus stick and other fungus sticks in the neighborhood range.
Preferably, in one embodiment of the present invention, the method for obtaining noisiness includes:
Obtaining the noisiness according to an acquisition formula of the noisiness, wherein the acquisition formula of the noisiness is as follows:
；
wherein, Represents the/>At the moment,/>First/>, of individual fungus sticksNoisiness of the individual growth parameter data; /(I)Represents the/>At the moment,/>Individual fungus sticks and/>Corresponding/>, between individual fungus sticksLocal trend variability of the individual growth parameter data; Represents the/> At the moment,/>Individual fungus sticks and/>Corresponding/>, between individual fungus sticksLocal trend variability of the individual growth parameter data; /(I)A number representing a growth parameter; /(I)Representing the number of bacteria sticks in the neighborhood range; /(I)Representing the normalization function.
In the noisiness acquisition formula, the larger the path cost value between bacteria sticks is, the higher the degree of abnormality of the data points under the parameters at the moment is, and the more likely the data points are noise; expressed in/> At the moment,/>Individual fungus sticks and/>First/>, of individual fungus sticksDiscrete differences in local trend variability between individual growth parameter data and the remaining growth parameter data, the higher the difference, indicate at that moment, the/>The growth parameter data are not the difference in the growth process caused by the congenital difference of hypha in the fungus stick, but local abnormal points caused by noise interference to a great extent, and the higher the noisiness of the growth parameter data of the fungus stick at the moment, the more the noise needs to be restrained.
It should be noted that, in other embodiments of the present invention, the positive-negative correlation and normalization method may be constructed by other basic mathematical operations, and specific means are technical means well known to those skilled in the art, and will not be described herein.
Step S4: obtaining the constraint weight of each growth parameter data of each fungus stick at each historical moment according to the response coefficient of each fungus stick and the noisiness of each growth parameter data; acquiring initial exponential smoothing weights, and adjusting the initial exponential smoothing weights according to constraint weights to acquire optimized exponential smoothing weights of each growth parameter data of each fungus stick at each historical moment; and predicting the growth parameter data according to the optimization index smoothing weight.
Variations in response coefficients and noisiness can affect the assignment of data point weights and thus the assessment of overall growth. The weight of each data point when reflecting the growth condition of the fungus stick is more accurately determined by combining analysis response coefficient and noisiness, so that data analysis, prediction and the like are better carried out. The sensitivity degree of the fungus stick to the growth parameters can be known by the response coefficient, the response coefficient is larger, the change of the growth parameters is more sensitive, the more abnormal data points exist, and the growth parameters need to be restrained; the higher the noisiness, the more constraints are needed to avoid interference of noise; and (3) distributing corresponding constraint weights to each growth parameter data, so that the behavior and the performance of each fungus stick in the growth process can be better understood, and simultaneously, the data containing noise can be better processed and analyzed, so that more accurate results can be obtained. And obtaining the constraint weight of each growth parameter data of each fungus stick at each historical time according to the response coefficient of each fungus stick and the noisiness of each growth parameter data at each historical time. It should be noted that, in one embodiment of the present invention, the range of the adaptive adjustment weight of each fungus stick isWhen/>The larger the size, the higher the degree of noise influence existing in the whole interior, and correspondingly, when the weight optimization is carried out on each growth parameter data, the smaller/>, is selectedThereby achieving a higher weight constraint for higher noisiness values.
Preferably, in one embodiment of the present invention, the method for acquiring the constraint weight includes:
obtaining constraint weights according to the constraint weight obtaining formula, wherein the constraint weight obtaining formula is as follows:
；
wherein, Represents the/>At the moment,/>The/>Constraint weights of the individual growth parameter data; /(I)Represent the firstResponse coefficients of the individual sticks; /(I)Represents the/>At the moment,/>The/>Noisiness of the individual growth parameter data; /(I)Representing a preset adjustment factor.
In the constraint weight acquisition formula, when the response coefficientWhen increasing, the overall internal noise influence is relatively high,/>The portion will decrease, which means that the sensitivity or response intensity of the stick to external stimuli increases. At the same time, the method comprises the steps of,Part of the strain stick is increased, which indicates that the strain stick has stronger overall response coefficient and generates higher weight constraint on a value with higher noise degree; noise level/>, when the growth parameter dataWhen increasing,/>The part will increase, which means that the influence of the noisy growth parameter data in the overall growth situation assessment will be enhanced, requiring a higher weight constraint.
It should be noted that, in one embodiment of the present invention, although noise has a certain influence on data, the response coefficient of the bacteria stick is a more important factor, and in order to balance the relationship between the overall response coefficient and the noise degree in weight calculation, an adjustment factor is preset0.5; In other embodiments of the present invention, the size of the preset adjustment factor may be specifically set according to specific situations, which are not limited and described herein.
The optimization index smoothing weight can be better adapted to the change of data, the influence of noise on a prediction result is reduced, the calculation efficiency is improved, the prediction process is more efficient, and the prediction value is more accurate. And acquiring an initial exponential smoothing weight, and adjusting the initial exponential smoothing weight according to the constraint weight to acquire an optimized exponential smoothing weight of each growth parameter data of each fungus stick at each historical moment.
Preferably, in one embodiment of the present invention, the method for obtaining the optimized exponential smoothing weight includes:
obtaining an optimization index smoothing weight according to an acquisition formula of the optimization index smoothing weight, wherein the acquisition formula of the optimization index smoothing weight is as follows:
；
wherein, Represents the/>At the moment,/>The/>Optimizing exponential smoothing weights of the individual growth parameter data; /(I)Indicating a preset super parameter in an exponential smoothing algorithm; /(I)A serial number indicating a history time; /(I)Represents the/>At the moment,/>The/>Constraint weights of the individual growth parameter data; /(I)Representing a logistic function.
In an acquisition formula for optimizing the exponential smoothing weight, the farther each data point is from the current moment, the smaller the obtained weight is; the closer to the current moment, the larger the constraint weight, the larger the possibility of being noisy, the larger the optimization index smoothing weight, byThe function scales the input values such that the sum of all optimization exponential smoothing weights is 1.
It should be noted that, in an embodiment of the present invention, the preset super parameter in the exponential smoothing algorithm is preset by an operator according to specific situations, and the value range is between 0.4 and 0.6, which is not limited and described herein.
The optimized exponential smoothing weight is used for prediction, so that the actual condition of the growth of the fungus stick can be reflected more accurately, and the accuracy of prediction is improved; and predicting the growth parameter data according to the optimization index smoothing weight, and finding abnormal conditions or problems in time so as to take measures in time to intervene and process.
Preferably, in one embodiment of the present invention, predicting the growth parameter data according to the optimization exponential smoothing weights comprises:
And carrying out an index smoothing algorithm based on the optimized index smoothing weight of each growth parameter data at the historical time to obtain the growth parameter data at the next time, and predicting the growth trend of the fungus stick and the problems possibly encountered in advance, so that preventive maintenance measures are adopted in advance, and the problems are avoided or the loss is reduced.
Step S5: and judging the maturity of the fungus stick according to the prediction result.
In order to better understand the growth condition and the characteristics of the fungus sticks, the maturity of the fungus sticks is judged according to the prediction result, so that the production process is better regulated, the culture conditions and management measures are optimized, and the production efficiency is improved. In one embodiment of the present invention, the criterion for determining the maturity of the growth parameters is preset by an operator according to specific situations, and is not limited and described herein. The mature fungus sticks are prepared in advance by comparing the prediction result with a preset maturity judgment standard, so that the fungus sticks can be transplanted to a fruiting shed in time, and the efficiency is improved.
In summary, the overall trend difference degree of the corresponding growth parameter data among each fungus stick is obtained, curve fitting is carried out on the corresponding growth parameter data in each fungus stick on a time sequence, and fitting errors between the corresponding growth parameter data and fitting results at each historical moment are obtained; obtaining a response coefficient of each fungus stick according to the overall distribution of fitting errors of each fungus stick at all historical moments and the overall trend difference degree distribution characteristics between each fungus stick and other fungus sticks in the neighborhood range; obtaining the noisiness of each growth parameter data at each historical time according to the difference characteristics between each growth parameter data and other growth parameter data, and further obtaining the constraint weight of each growth parameter data at each historical time; adjusting the initial exponential smoothing weight to obtain the optimized exponential smoothing weight of each growth parameter data at each historical time; and predicting the growth parameter data, and judging the maturity of the fungus stick according to the prediction result. According to the invention, by obtaining the proper exponential smoothing weight of the growth parameter data, the accuracy of the predicted value is improved, and the maturity is timely judged.
An embodiment of a fungus stick growth parameter data prediction method comprises the following steps:
In the prior art, related parameters inside a fungus stick are generally collected through a sensor, and growth related data of the fungus stick are predicted through an input index smoothing algorithm, but due to limitation of the sensor, the collected data can have interference of certain noise, if only time sequence difference is considered, higher index smoothing weight is caused in a data point with low confidence in the prediction process, and further the technical problem of inaccurate predicted value is caused, and in order to solve the technical problem, the embodiment provides a fungus stick growth parameter data prediction method, which comprises the following steps:
Step S1: and acquiring growth parameter data of each fungus stick at each historical time.
Step S2: according to the position of the historical moment on the time sequence, the difference of the corresponding growth parameter data between each fungus stick at each historical moment is adjusted, and the local trend difference degree is obtained; obtaining the overall trend difference of the corresponding growth parameter data among each fungus stick according to the local trend difference at all the historical moments; performing curve fitting on the corresponding growth parameter data in each fungus stick on a time sequence to obtain fitting errors between the corresponding growth parameter data and fitting results at each historical moment; and obtaining the response coefficient of each fungus stick according to the overall distribution of the fitting error of each growth parameter of each fungus stick at all historical moments and the distribution characteristics of the overall trend difference degree between each fungus stick and other fungus sticks in the neighborhood range.
Step S3: and under each historical time, obtaining the noisiness of each growth parameter data of each fungus stick under each historical time according to the difference characteristic of the local trend difference degree between each growth parameter data and other growth parameter data between each fungus stick and other fungus sticks in the neighborhood range.
Step S4: obtaining the constraint weight of each growth parameter data of each fungus stick at each historical moment according to the response coefficient of each fungus stick and the noisiness of each growth parameter data; acquiring initial exponential smoothing weights, and adjusting the initial exponential smoothing weights according to constraint weights to acquire optimized exponential smoothing weights of each growth parameter data of each fungus stick at each historical moment; and predicting the growth parameter data according to the optimization index smoothing weight.
Because the specific implementation process of steps S1-S4 is already described in detail in the above method for determining the maturity of a fungus stick based on data analysis, no further description is given.
The technical effects of this embodiment are:
According to the method, the integral trend difference degree of the corresponding growth parameter data among the fungus sticks is obtained, curve fitting is carried out on the corresponding growth parameter data in each fungus stick on a time sequence, and fitting errors between the corresponding growth parameter data and fitting results at each historical moment are obtained; obtaining a response coefficient of each fungus stick according to the overall distribution of fitting errors of each fungus stick at all historical moments and the overall trend difference degree distribution characteristics between each fungus stick and other fungus sticks in the neighborhood range; obtaining the noisiness of each growth parameter data at each historical time according to the difference characteristics between each growth parameter data and other growth parameter data, and further obtaining the constraint weight of each growth parameter data at each historical time; adjusting the initial exponential smoothing weight to obtain the optimized exponential smoothing weight of each growth parameter data at each historical time; and predicting the growth parameter data. According to the invention, the accuracy of the predicted value is improved by obtaining the proper exponential smoothing weight of the growth parameter data.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (9)
1. The utility model provides a fungus stick maturity judging method based on data analysis, which is characterized in that the method comprises the following steps:
Acquiring growth parameter data of each fungus stick at each historical moment;
According to the position of the historical moment on the time sequence, the difference of the corresponding growth parameter data between each fungus stick at each historical moment is adjusted, and the local trend difference degree is obtained; obtaining the overall trend difference degree of the corresponding growth parameter data among each fungus stick according to the local trend difference degree at all the historical moments; performing curve fitting on the corresponding growth parameter data in each fungus stick on a time sequence to obtain fitting errors between the corresponding growth parameter data and fitting results at each historical moment; obtaining a response coefficient of each fungus stick according to the overall distribution of the fitting errors of each growth parameter of each fungus stick at all historical moments and the distribution characteristics of the overall trend difference degree between each fungus stick and other fungus sticks in a neighborhood range;
Obtaining the noisiness of each growth parameter data of each fungus stick at each historical time according to the difference characteristics of the local trend difference degree between each growth parameter data and other growth parameter data between each fungus stick and other fungus sticks in the neighborhood range at each historical time;
Obtaining constraint weight of each growth parameter data of each fungus stick at each historical time according to the response coefficient of each fungus stick and the noisiness of each growth parameter data; acquiring initial exponential smoothing weights, and adjusting the initial exponential smoothing weights according to the constraint weights to acquire optimized exponential smoothing weights of each growth parameter data of each fungus stick at each historical time; predicting the growth parameter data according to the optimization index smoothing weight;
judging the maturity of the fungus stick according to the prediction result;
the response coefficient acquisition method comprises the following steps:
Obtaining a response coefficient according to an obtaining formula of the response coefficient, wherein the obtaining formula of the response coefficient is as follows:
; wherein/> Represents the/>Response coefficients of the individual sticks; /(I)Represents the/>Individual fungus sticks and/>The bacterial stick is at the first placeOverall trend variability in the individual growth parameter data; /(I)Representing the number of bacteria sticks in the neighborhood range; /(I)Represents the/>Individual sticks and others/>The bacterial stick is at the first placeAn average of overall trend variability in the individual growth parameter data; /(I)Expressed in/>At the moment,/>First/>, of individual fungus sticksFitting errors of the growth parameter data and the fitting result; /(I)Indicating at all historic times, the/>In the individual fungus stick/>Accumulating fitting errors corresponding to the growth parameters, and calculating to obtain standard deviation; /(I)A number representing a growth parameter; /(I)The number of times of historic time is represented; /(I)Representing the normalization function.
2. The method for determining the maturity of a fungus stick based on data analysis according to claim 1, wherein the method for obtaining the local trend difference comprises:
Calculating the relative distance of the corresponding growth parameter data between each fungus stick and other fungus sticks in the neighborhood range at each historical time based on a DTW algorithm, and taking the relative distance as the initial path cost value of the corresponding growth parameter data between each fungus stick at each historical time;
calculating the difference between each historical moment and the current moment, and normalizing the difference to be used as a time weighting factor;
Calculating the product of the initial path cost value and the time weighting factor to obtain the local trend difference degree of the corresponding growth parameter data between each fungus stick at each historical moment;
and the initial journey cost value and the time weighting factor are in positive correlation with the local trend difference degree.
3. The method for determining the maturity of a fungus stick based on data analysis according to claim 1, wherein the method for obtaining the overall trend difference comprises:
and accumulating the local trend difference degrees under all the historical moments to obtain the overall trend difference degrees of the corresponding growth parameter data among each fungus stick.
4. The method for determining the maturity of a fungus stick based on data analysis according to claim 1, wherein the method for obtaining the fitting error comprises the following steps:
and performing curve fitting on the corresponding growth parameter data of each fungus stick at all the historical moments by adopting a least square method, and calculating the difference value between each growth parameter data and the corresponding fitting result of each fungus stick at each historical moment to serve as a fitting error.
5. The method for determining the maturity of a fungus stick based on data analysis according to claim 1, wherein the method for obtaining the noisiness comprises the following steps:
Obtaining the noisiness according to an acquisition formula of the noisiness, wherein the acquisition formula of the noisiness is as follows:
; wherein/> Represents the/>At the moment,/>First/>, of individual fungus sticksNoisiness of the individual growth parameter data; /(I)Represents the/>At the moment,/>Individual fungus sticks and/>Corresponding first fungus sticksLocal trend variability of the individual growth parameter data; /(I)Represents the/>At the moment,/>Individual fungus sticks and/>Corresponding/>, between individual fungus sticksLocal trend variability of the individual growth parameter data; /(I)A number representing a growth parameter; /(I)The number of bacteria sticks in the neighborhood range is indicated.
6. The method for determining the maturity of a fungus stick based on data analysis according to claim 1, wherein the method for obtaining the constraint weight comprises:
obtaining constraint weights according to the constraint weight obtaining formula, wherein the constraint weight obtaining formula is as follows:
; wherein/> Represents the/>At the moment,/>The/>Constraint weights of the individual growth parameter data; /(I)Represents the/>Response coefficients of the individual sticks; /(I)Represents the/>At the moment,/>The/>Noisiness of the individual growth parameter data; /(I)Representing a preset adjustment factor.
7. The method for determining the maturity of a fungus stick based on data analysis according to claim 1, wherein the method for obtaining the optimization index smoothing weight comprises the following steps:
obtaining an optimization index smoothing weight according to an acquisition formula of the optimization index smoothing weight, wherein the acquisition formula of the optimization index smoothing weight is as follows:
; wherein/> Represents the/>At the moment,/>The/>Optimizing exponential smoothing weights of the individual growth parameter data; /(I)Indicating a preset super parameter in an exponential smoothing algorithm; /(I)A serial number indicating a history time; /(I)Represents the/>At the moment,/>The/>Constraint weights of the individual growth parameter data; /(I)Representing a logistic function.
8. The method for determining the maturity of a bacterial stick based on data analysis according to claim 1, wherein said predicting the growth parameter data according to the optimization exponential smoothing weight comprises:
and carrying out an exponential smoothing algorithm based on the optimized exponential smoothing weight of each growth parameter data at the historical moment to obtain the growth parameter data at the next moment.
9. The method for determining the maturity of a fungus stick based on data analysis according to claim 6, wherein the preset adjustment factor is 0.5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410185781.XA CN117746994B (en) | 2024-02-20 | 2024-02-20 | Fungus stick maturity judging method based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410185781.XA CN117746994B (en) | 2024-02-20 | 2024-02-20 | Fungus stick maturity judging method based on data analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117746994A CN117746994A (en) | 2024-03-22 |
| CN117746994B true CN117746994B (en) | 2024-04-26 |
Family
ID=90257788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410185781.XA Active CN117746994B (en) | 2024-02-20 | 2024-02-20 | Fungus stick maturity judging method based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117746994B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003106685A1 (en) * | 2002-06-18 | 2003-12-24 | Universite Catholique De Louvain | Cell wall mutants for delivery of biologically active compounds |
| CN211353366U (en) * | 2019-11-15 | 2020-08-28 | 安远县飞龙农业投资开发有限公司 | Glossy ganoderma is planted and is cultivateed room device of moisturizing |
| CN115690777A (en) * | 2022-11-07 | 2023-02-03 | 北京市农林科学院智能装备技术研究中心 | Mushroom stick maturity identification method and system based on maturity identification device |
| CN117436353A (en) * | 2023-12-21 | 2024-01-23 | 泰安维创游乐设备有限公司 | Intelligent recreation facility fault prediction method based on big data |
-
2024
- 2024-02-20 CN CN202410185781.XA patent/CN117746994B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003106685A1 (en) * | 2002-06-18 | 2003-12-24 | Universite Catholique De Louvain | Cell wall mutants for delivery of biologically active compounds |
| CN211353366U (en) * | 2019-11-15 | 2020-08-28 | 安远县飞龙农业投资开发有限公司 | Glossy ganoderma is planted and is cultivateed room device of moisturizing |
| CN115690777A (en) * | 2022-11-07 | 2023-02-03 | 北京市农林科学院智能装备技术研究中心 | Mushroom stick maturity identification method and system based on maturity identification device |
| CN117436353A (en) * | 2023-12-21 | 2024-01-23 | 泰安维创游乐设备有限公司 | Intelligent recreation facility fault prediction method based on big data |
Non-Patent Citations (2)
| Title |
|---|
| Shiitake Mushroom Semantic Segmentation Method Based on Search Focus Network;Juan Du;IEEE;20230223;全文 * |
| 凉山地区小儿血培养检出菌及药物敏感性分析;王西蓉;杨尧;边程鹏;邵志英;;华西医学;20090625(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117746994A (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117196353B (en) | Environmental pollution assessment and monitoring method and system based on big data | |
| CN116756529B (en) | White piece production facility running state detecting system hangs | |
| CN116400126B (en) | Low-voltage power box with data processing system | |
| CN116073436B (en) | Capacity optimization control method for photovoltaic new energy power system | |
| CN115993856B (en) | Multi-region environmental condition management and control method and system for factory building | |
| CN117746994B (en) | Fungus stick maturity judging method based on data analysis | |
| CN113031670A (en) | Warehouse humidity automatic control system based on big data | |
| CN112418270A (en) | Blast furnace comprehensive state prediction method and system | |
| CN117172958B (en) | Bean sprout growth real-time data management method | |
| TWI697585B (en) | Hot dip galvanizing product defect estimation system and method thereof | |
| CN108664000A (en) | A kind of alumina producing evaporation process steady state detecting method for use and system | |
| CN117313017B (en) | Color leaf research and development data processing method and system | |
| CN114139446A (en) | On-line detection soft measurement method for components in special rectification process | |
| CN117992894B (en) | Agricultural greenhouse environment abnormal data monitoring method based on Internet of things | |
| CN117710008B (en) | Ecological product sales information management system suitable for karst region | |
| CN117786445B (en) | Intelligent processing method for operation data of automatic yarn reeling machine | |
| CN117829381B (en) | Agricultural greenhouse data optimization acquisition system based on Internet of things | |
| CN117951631B (en) | Intelligent constant-temperature constant-pressure cooling water circulation system | |
| CN118197635A (en) | Beef cattle health detection data processing method | |
| CN116774575B (en) | Viscosity control method and system for dextrin production process | |
| CN117851766B (en) | Temperature data monitoring method for DOTP crude ester continuous distillation process | |
| CN118013323B (en) | Driving motor state analysis method and system for large-caliber electric half ball valve | |
| CN117804639B (en) | Temperature calibration method and system for temperature control sensor of cementing machine | |
| CN117798744B (en) | Method for monitoring running state of numerical control machine tool | |
| CN117331363A (en) | Operation evaluation method for multi-structure refining controller |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |