CN-121981846-A - Morchella growth environment dynamic optimization method based on Internet of things and data fusion

CN121981846ACN 121981846 ACN121981846 ACN 121981846ACN-121981846-A

Abstract

The invention relates to the technical field of agricultural informatization and refinement management, in particular to a dynamic optimization method for a Morchella growing environment based on the Internet of things and data fusion. A dynamic optimization method for a Morchella esculenta growth environment based on the Internet of things and data fusion comprises the following steps of S1, collecting control parameters and demand parameters of the Morchella esculenta growth environment, determining a moderate interval and a stress interval according to the control parameters and the demand parameters, and S2, obtaining a moderate interval correlation pair and a stress interval similarity pair based on the moderate interval and the stress interval, and obtaining a moderate interval line segment set and a stress interval line segment set according to the moderate interval correlation pair and the stress interval similarity pair. The method and the device can accurately identify the critical points and the mutation points by digging the Morchella growth rhythm, and make a dynamic regulation and control plan according to the critical points and the mutation points, and effectively distinguish biological signals and interference by multi-stage data fusion, thereby improving the environment regulation and control precision and the system self-adaption capability and overcoming the defects of the traditional static control.

Inventors

MA LINGFA
LI SANXIANG
YANG MIN

Assignees

天水师范大学
天水市博浩致源农业科技发展有限公司

Dates

Publication Date: 20260505
Application Date: 20260205

Claims (10)

1. The dynamic optimization method for the Morchella esculenta growth environment based on the Internet of things and data fusion is characterized by comprising the following steps: S1, collecting control parameters and demand parameters of a Morchella esculenta growth environment, and determining a moderate interval and a stress interval according to the control parameters and the demand parameters; S2, obtaining a moderate interval correlation pair and a stress interval similarity pair based on the moderate interval and the stress interval, and obtaining a moderate interval line segment set and a stress interval line segment set according to the moderate interval correlation pair and the stress interval similarity pair; s3, clustering the moderate interval line segment set and the stress interval line segment set to obtain a growth state continuous critical point and a growth stage transition mutation point; S4, dynamically optimizing the Morchella esculenta growth environment based on the growth state continuous critical point and the growth stage transition mutation point.
2. The dynamic optimization method for the Morchella esculenta growth environment based on the Internet of things and data fusion as set forth in claim 1, wherein the collecting the control parameters and the demand parameters of the Morchella esculenta growth environment comprises: dividing the control parameters into a first control parameter and a second control parameter according to the control response time, wherein the first control parameter comprises control start time and control end time, and the second control parameter comprises control start actual action time and control end actual action time; the demand parameters are divided into first demand parameters and second demand parameters according to the demand response time, wherein the first demand parameters are demand start time, and the second demand parameters are demand end time.
3. The dynamic optimization method of the Morchella esculenta growth environment based on the Internet of things and data fusion of claim 1, wherein the determining the moderate interval and the stress interval according to the control parameter and the demand parameter comprises: determining a target period of time for executing environment regulation based on the demand start time and the demand end time in the demand parameters; generating a control instruction according to requirements of a Morchella growing stage in a target period, and recording a control start time and a control end time; recording the actual action time of starting control and the actual action time of ending control by monitoring the actual change of environmental parameters; taking the actual action time from the control start time to the control start time as a moderate interval, wherein the moderate interval is a time period for ensuring that the environmental parameters reach and maintain the optimal range of the nutrition growth of Morchella mycelium stably; and taking the actual acting time from the control end time to the control end as a stress interval, wherein the stress interval is a time period for actively stopping the current comfortable environment and controllably deviating the parameter from the optimal range.
4. The dynamic optimization method of the Morchella esculenta growth environment based on the Internet of things and data fusion of claim 1, wherein the obtaining the moderate interval association pair and the stress interval similarity pair based on the moderate interval and the stress interval comprises: Selecting a key parameter set based on indirect features and medium-dependent features of Morchella vital activities; Based on the key parameter set, selecting a cooperative parameter pair according to the cooperative steady-state requirement of soil and air environmental factors in the Morchella esculenta nutrition growth stage; calculating mutual information values of the cooperative parameter pairs in a historical moderate interval, and taking the moderate interval which is higher than the cooperative parameter pair corresponding to a preset mutual information threshold value as a moderate interval association pair; Based on the key parameter set, selecting a similar parameter pair according to the triggering requirement of the Morchella reproductive growth stage on the formation of a composite stress signal from the multiple environmental factors; Extracting a time sequence of parameter pairs of similar parameter pairs in the stress interval to form a stress interval feature vector; Calculating the dynamic time warping distance between feature vectors of different stress intervals, and taking the stress interval which is lower than a preset warping distance threshold as a stress interval similarity pair.
5. The dynamic optimization method of the Morchella esculenta growth environment based on the Internet of things and data fusion of claim 1, wherein the obtaining the moderate interval line segment set and the stress interval line segment set according to the moderate interval correlation pair and the stress interval similarity pair comprises: extracting the moderate intervals contained in all the moderate interval association pairs, and merging the moderate intervals into a moderate interval complete set; Mapping each moderate interval in the moderate interval total set into a line segment on a time axis to obtain a moderate interval line segment set, wherein a line segment starting point and a line segment ending point respectively correspond to the starting time and the ending time of the moderate interval, mapping each stress interval in the stress interval total set into a line segment on the time axis to obtain a stress interval line segment set, and the line segment starting point and the line segment ending point respectively correspond to the starting time and the ending time of the stress interval.
6. The dynamic optimization method for Morchella esculenta growth environment based on the Internet of things and data fusion of claim 1, wherein the clustering of the moderate interval line segment set and the stress interval line segment set comprises: The time density clustering operation comprises the steps of calculating the time distance between every two line segment center points in the corresponding line segment set for one input line segment set, corresponding time neighborhood radius parameters and minimum cluster scale parameters, counting the number of adjacent line segments in the time neighborhood radius by taking each line segment center point in the corresponding line segment set as a base point, marking the line segments with the number reaching the minimum cluster scale as core line segments, merging the time distances between the two line segment center points in the core line segments to be the same time cluster, distributing the line segments which are not core line segments in the corresponding line segment set to any core line segment cluster of which the line segment center points fall in the time neighborhood radius, and outputting the time cluster of the corresponding line segment set.
7. The dynamic optimization method of Morchella esculenta growth environment based on the Internet of things and data fusion of claim 1, wherein the clustering of the moderate interval line segment set and the stress interval line segment set to obtain the growth state continuous critical point and the growth stage transition mutation point comprises: invoking time density clustering operation on the moderate interval line segment set, and applying moderate time neighborhood radius and moderate minimum cluster scale parameters to obtain a moderate interval time cluster; invoking time density clustering operation on the stress interval line segment set, and applying stress time neighborhood radius and stress minimum cluster scale parameters to obtain a stress interval time cluster; Based on the moderate interval time clustering cluster and the stress interval time clustering cluster, calculating the average time position of all line segment center points in each cluster to be used as the center point of the corresponding cluster, identifying the moderate interval cluster center point as a growth state continuous critical point, and identifying the stress interval cluster center point as a growth stage transition mutation point.
8. The dynamic optimization method of the Morchella esculenta growth environment based on the Internet of things and data fusion as set forth in claim 1, wherein the dynamic optimization of the Morchella esculenta growth environment based on the growth state continuous critical point and the growth stage transition mutation point comprises: generating a target regulation and control time sequence plan in a moderate environment maintenance stage by taking a continuous critical point of a growth state as a reference, wherein the target regulation and control time sequence plan comprises a steady transition curve from an environmental parameter to an optimal range before the critical point, and a maintenance duration and a tolerance range of the environmental parameter in a target steady-state interval after the critical point; Generating a target regulation and control time sequence plan of a stress environment application stage by taking a growth stage conversion mutation point as a reference, wherein the target regulation and control time sequence plan comprises a compound adversity signal mode which is started at the mutation point and accords with the requirements of Morchella reproduction conversion, and the compound adversity signal mode is formed by cooperatively changing N environmental parameters according to preset amplitude, time sequence and speed; And carrying out joint analysis and fusion optimization on the dynamic optimization result and the real-time data monitored by the Internet of things.
9. The dynamic optimization method of the Morchella esculenta growth environment based on the combination of the Internet of things and the data as set forth in claim 1, wherein the joint analysis and the combination optimization of the dynamic optimization result and the monitoring real-time data of the Internet of things comprise: collecting instantaneous values of all key parameters in a key parameter set monitored by an internet of things sensor network in real time; Calculating the instantaneous deviation between the instantaneous value of each key parameter in the key parameter set and the corresponding time target value in the target regulation time sequence plan; counting standard deviation of historical instantaneous deviation in a time window of a preset length for each key parameter in the key parameter set, and marking the corresponding instantaneous deviation as candidate abnormal deviation if the absolute value of the current instantaneous deviation of the corresponding key parameter is larger than the product of the historical standard deviation of the corresponding key parameter and a preset abnormal discrimination coefficient; And if the ratio of the number of the coupling sensors, in which the reading change direction of the coupling sensors is consistent with the direction of the candidate abnormal deviation, to the number of all the coupling sensors exceeds a preset consistency ratio threshold value, updating the state of the candidate abnormal deviation into an effective environment deviation event, otherwise updating the state of the candidate abnormal deviation into suspected isolated noise.
10. The dynamic optimization method of Morchella esculenta growth environment based on the Internet of things and data fusion of claim 9, wherein the updating the candidate abnormal deviation state into the effective environment deviation event comprises: For each instantaneous deviation of which the state is an effective environment deviation event, executing biological response necessity judgment based on a historical case, and searching all historical environment deviation event records in a historical database by taking a current effective environment deviation event mode as a query condition, wherein the historical environment deviation event records comprise the environment deviation event mode and observed morchella growth state change results after the event occurs; Calculating the ratio of the number of records of successful triggering reproduction conversion to the total number of records in all the searched historical environment deviation event records, and judging the category of the current effective environment deviation event as an effective stress signal if the ratio of successful triggering reproduction conversion exceeds a preset success threshold; calculating the ratio of the number of records which cause growth inhibition or invalid response to the total number of records in all the retrieved historical environment deviation event records to obtain the growth inhibition or invalid response ratio, and judging the category of the current effective environment deviation event as harmful interference if the growth inhibition or invalid response ratio exceeds a preset risk threshold; and for the instantaneous deviation of which the category is judged to be harmful interference or the state is suspected isolated noise, filtering the influence of the corresponding instantaneous deviation when the control instruction is generated, and strictly following the execution of the original target regulation and control time sequence plan.

Description

Morchella growth environment dynamic optimization method based on Internet of things and data fusion Technical Field The invention relates to the technical field of agricultural informatization and refinement management, in particular to a dynamic optimization method for a Morchella growing environment based on the Internet of things and data fusion. Background The technology of the internet of things realizes real-time acquisition of multidimensional physical parameters in the Morchella cultivation environment by deploying a sensor network, and provides a data base for environment monitoring. The data fusion technology aims at carrying out cooperative processing and deep analysis on the multi-source heterogeneous monitoring information so as to form a more comprehensive perception on the growth state. In Morchella cultivation in facilities, the growing environment is a complex system formed by coupling multiple factors such as air temperature and humidity, soil temperature and humidity, carbon dioxide concentration, illumination intensity and the like, and the requirements of two stages of mycelium nutrition growth and fruiting body reproduction growth on the environment are obviously contradicted. The existing method realizes remote acquisition of environmental data and automatic control based on a fixed threshold by utilizing the technology of the Internet of things, achieves preliminary results in the aspects of reducing the labor intensity and maintaining the basic stability of the environment, and provides possibility for standardized production. However, the prior art relies mainly on predefined static rules or simple linkage strategies, the core drawback of which is the failure to achieve dynamic optimization based on unique biological characteristics of morchella. The method is characterized in that instantaneous noise of a sensor and a real biological stress signal are difficult to effectively discriminate, so that a control instruction oscillates, and a critical point for differentiating and converting hypha growth to primordia cannot be accurately identified and responded at a decision level, so that environment regulation is delayed or improper, and the problems of overgrowth of hypha and primordial ablation production are caused. The essence is that the current method lacks the data-driven modeling and fusion analysis capability for the growth nonlinearity and time sequence rule of Morchella. Therefore, research on an intelligent regulation and control method capable of deeply understanding Morchella esculenta growth logic is needed, and the limitation of static control is broken through by fusing time sequence environment data and biological state information, so that dynamic self-adaptive optimization of the growth environment along with the physiological demands of thalli is realized, and the stability and the output quality of cultivation are improved. Disclosure of Invention In order to overcome the defect of insufficient dynamic regulation and control precision of the growth environment, the invention provides a dynamic optimization method of the Morchella growth environment based on the integration of the Internet of things and data. The technical embodiment of the invention is that the morchella growing environment dynamic optimization method based on the Internet of things and data fusion comprises the following steps: S1, collecting control parameters and demand parameters of a Morchella esculenta growth environment, and determining a moderate interval and a stress interval according to the control parameters and the demand parameters; S2, obtaining a moderate interval correlation pair and a stress interval similarity pair based on the moderate interval and the stress interval, and obtaining a moderate interval line segment set and a stress interval line segment set according to the moderate interval correlation pair and the stress interval similarity pair; s3, clustering the moderate interval line segment set and the stress interval line segment set to obtain a growth state continuous critical point and a growth stage transition mutation point; S4, dynamically optimizing the Morchella esculenta growth environment based on the growth state continuous critical point and the growth stage transition mutation point. Preferably, the collecting the control parameters and the demand parameters of the Morchella esculenta growth environment includes: dividing the control parameters into a first control parameter and a second control parameter according to the control response time, wherein the first control parameter comprises control start time and control end time, and the second control parameter comprises control start actual action time and control end actual action time; the demand parameters are divided into first demand parameters and second demand parameters according to the demand response time, wherein the first demand parameters are demand start time, and the second demand paramet