An adaptive adjustment method for storage environment of granular fertilizer

By collecting external and internal environmental data of the storage facility and combining it with the intrinsic demand data of granular fertilizer, the control period is determined. An overlapping conflict compensation mechanism for gas release sensitive periods is introduced, and a time-delay correlation mapping network for gas release of multiple fertilizer varieties is established. This solves the problem of mismatch between control actions and states in existing technologies and achieves more efficient storage environment control.

CN122431466BActive Publication Date: 2026-11-17LIAONING XINLIANXIN ECOLOGICAL TECHNOLOGY CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202610571327.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-04-28
Publication Date
2026-11-17
Estimated Expiration
2046-04-28

AI Technical Summary

Technical Problem

Existing methods are unable to respond to the dynamic phase changes of external environmental parameters and the time-varying characteristics of multi-source gas composition inside the storage when dealing with the complex working conditions of mobile storage carriers and mixed storage of multiple varieties of granular fertilizers, resulting in a systematic mismatch between the adjustment actions and the internal and external states.

Method used

By collecting external and internal environmental data of the storage facility and combining it with the intrinsic demand data of granular fertilizer, the control period is determined. An overlapping conflict compensation mechanism for gas release sensitive periods is introduced to establish a time-delay correlation mapping network for gas release of multiple fertilizer varieties, and the control duration and parameters are dynamically corrected.

Benefits of technology

It improves the precision of warehousing environment regulation and the collaborative adaptive capability of process control systems to complex environmental fluctuations and the temporal evolution of multi-source gases, overcoming the shortcomings of timing and threshold control strategies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122431466B_ABST
    Figure CN122431466B_ABST
Patent Text Reader

Abstract

This invention relates to the field of program control systems, and more particularly to an adaptive adjustment method for the storage environment of granular fertilizers. The method includes the following steps: S1: Collecting data on changes in the external and internal environments of the storage facility during storage operations, and real-time collecting data on the intrinsic demand of the granular fertilizer within the storage facility; S2: Determining the control period based on the data on changes in the external and internal environments and the intrinsic demand data; S3: Collecting mixed gas data of the granular fertilizer within the control period, and determining the intrinsic gas data of different types of granular fertilizer based on the type of granular fertilizer within the storage facility. This invention achieves precise adaptive control of gas release in the storage of multiple fertilizer varieties by simultaneously collecting environmental and fertilizer demand data and constructing time offset characteristics and conflict compensation mechanisms.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of program control systems, and more particularly to an adaptive adjustment method for the storage environment of granular fertilizer. Background Technology

[0002] Granular fertilizers are solid agricultural inputs made from raw materials such as urea, calcium ammonium nitrate, diammonium phosphate, and potassium chloride through a granulation process. Their storage refers to the static storage and dynamic transfer process within facilities such as factory warehouses, transit warehouses, and transport vehicles from production line to field application. Existing adaptive adjustment methods for the storage environment of granular fertilizers mainly include timed ventilation control, fixed threshold start-stop control, and feedback adjustment based on single-point temperature and humidity sensors. These methods have evolved from manual experience-based operation and semi-automatic mechanical ventilation to basic automated monitoring, achieving positive progress in reducing reliance on manual labor and improving the standardization of storage management.

[0003] However, existing methods have significant shortcomings in dealing with the complex operating conditions of mobile storage carriers and the co-storage of multiple varieties of granular fertilizers. Timed and threshold control strategies cannot respond to the dynamic phase changes of external environmental parameters and the time-varying characteristics of multi-source gas composition within the storage facility, and also lack the ability to coordinate the timing and duration of regulation across cycles. At the process control level, the temporal differences in gas release behavior among different fertilizer varieties and the time misalignment of regulation windows are not incorporated into the decision-making logic, leading to a systematic mismatch between regulatory actions and the actual needs of internal and external conditions.

[0004] Therefore, there is an urgent need to study an adaptive regulation method for the mobile storage process of granular fertilizers, which incorporates dynamic planning of regulation timing and temporal correlation of multi-fertilizer gas release behavior into the control framework, so as to improve the coordinated response capability of the process control system to environmental fluctuations and material state changes. Summary of the Invention

[0005] To overcome the drawbacks of mismatch between timing and state of regulation, this invention provides an adaptive regulation method for the storage environment of granular fertilizers.

[0006] The technical implementation scheme of the present invention is: an adaptive adjustment method for the storage environment of granular fertilizer, comprising the following steps: S1: Collect data on changes in the external and internal environments of the warehouse during the storage process, and collect real-time data on the demand for granular fertilizer inside the warehouse. S2: Determine the control period based on data on changes in the external environment of the warehouse, data on changes in the internal environment of the warehouse, and data on the demand of the warehouse itself; S3: Collect mixed gas data of granular fertilizer during the control period, and determine the bulk gas data of different types of granular fertilizer according to the type of granular fertilizer in the storage. S4: Adaptive adjustment of the storage environment based on mixed gas data and bulk gas data.

[0007] Preferably, determining the control period based on changes in the external environment of the warehouse, changes in the internal environment of the warehouse, and the demand data includes: Determine the warehouse's self-adjustment cycle based on data on changes in the internal environment and the warehouse's own demand data; The phase transition period of the external environment is determined based on data on changes in the external and internal environments of the warehouse. The control period is determined based on the warehouse's own adjustment cycle and the phase change cycle of the external environment.

[0008] Preferably, determining the phase transition period of the external environment based on changes in the external environment and changes in the internal environment of the warehouse includes: Extract each external environmental parameter from the data on changes in the external environment of the warehouse, and extract each internal environmental parameter from the data on changes in the internal environment of the warehouse. When the changing trend of external environmental parameters has a positive impact on the internal environment of the warehouse or the demand for granular fertilizer, the corresponding dimension of the external environmental parameters is determined to be in a matching state; when the changing trend of external environmental parameters has a negative impact or no impact on the internal environment of the warehouse or the demand for granular fertilizer, the corresponding dimension of the external environmental parameters is determined to be in a mismatch state. The matching results of all external environment parameter dimensions are summarized to form the matching state vector at the current moment; The matching state vectors are recorded continuously at fixed time intervals to form a matching state vector sequence.

[0009] Preferably, the step of continuously recording the matching state vector at fixed time intervals to form a matching state vector sequence includes: The number of dimensions of the matching state at each time step in the matching state vector sequence is counted to obtain the curve of the number of dimensions of the matching state changing over time. Mark the moment when the curve changes from a continuous decline to a continuous rise as the starting moment of the beneficial period of the external environment; Mark the moment when the curve changes from continuous rise to continuous fall as the end of the beneficial period of the external environment; Calculate the time length between the start time and the end time of the beneficial external environment interval, and determine the time length as the duration of a single beneficial external environment phase transition; Record the duration of all single beneficial phase transitions in the external environment within the preset observation period, calculate the average duration of all single beneficial phase transitions in the external environment, and determine the average duration as the external environment phase transition period.

[0010] Preferably, determining the control period based on the warehouse's own adjustment cycle and the external environment's phase transition cycle includes: Extract the adjustment symmetry point of the warehouse's self-adjustment cycle. The adjustment symmetry point is the moment corresponding to the maximum deviation of the internal environmental parameters of the warehouse from the preset target value within the warehouse's self-adjustment cycle. The adjustment symmetry point represents the moment of the most urgent action center of the warehouse's internal adjustment needs. The equilibrium point of the phase transition cycle of the external environment is extracted. The equilibrium point is the moment when the absolute value of the rate of change of the external environment parameters is the smallest within the beneficial range of the external environment. The equilibrium point represents the central moment when the external environment has the most stable positive impact on the warehouse interior. Calculate the first time offset between the adjustment symmetry point and the equilibrium point. The first time offset is the signed difference obtained by subtracting the equilibrium point time from the time of adjustment symmetry point. To obtain the gas release sensitive period of granular fertilizer.

[0011] Preferably, the gas release sensitive period for obtaining granular fertilizer includes: The gas release sensitive period is the duration during which the concentration of characteristic gases released by granular fertilizer reaches its peak during storage. Calculate the second time offset between the equilibrium point and the center of the gas release sensitive period; When the second time offset is less than the preset overlap threshold, it is determined that the equilibrium point and the gas release sensitive period overlap and conflict, and a conflict offset compensation amount is generated. The conflict offset compensation amount is the minimum time length required to shift the equilibrium point along the time axis to the outside of the boundary of the gas release sensitive period. The first time offset is superimposed with the conflict offset compensation to obtain the total time offset. Specifically, the candidate control time is obtained by adding the first time offset to the equilibrium point time. If the candidate control time is within the gas release sensitive period, the total time offset is the conflict offset compensation; otherwise, the total time offset is the first time offset. Using the equilibrium point as a reference, the total time offset is added to the equilibrium point to obtain the start time of the control period; the length of the control period is equal to the original control action execution time.

[0012] Preferably, the adaptive adjustment of the storage environment based on the mixed gas data and the bulk gas data includes: Extract the variation amplitude value of the bulk gas data corresponding to each type of granular fertilizer within the same control period. The variation amplitude value is the absolute value of the concentration difference of the bulk gas data at the start and end times of the control period. The granular fertilizer with the smallest change value is marked as the first associated fertilizer, and the granular fertilizer with the largest change value is marked as the second associated fertilizer. A trend mapping relationship between the first associated fertilizer and the second associated fertilizer is established. The trend mapping relationship is the time delay corresponding to the maximum value of the time delay correlation coefficient between the gas data of the first associated fertilizer and the gas data of the second associated fertilizer over time. Then, pair the granular fertilizer with the second smallest change value with the granular fertilizer with the second largest change value to establish a trend mapping relationship between the two; and so on, until all granular fertilizers are paired. If the number of fertilizer types is odd, the middle one will not participate in the pairing, thus obtaining the mapping relationship of all pairs. During subsequent control periods, the gas data of the current control period will be collected in real time.

[0013] Preferably, the step of collecting the bulk gas data in real time during the current control period during subsequent control periods includes: Update the mapping relationship of changing trends among all granular fertilizers based on the current bulk gas data; Obtain the list of gas components of the first target gas for which control actions need to be performed during the current control period; Extract the granular fertilizer varieties corresponding to each gas component in the list of gas components of the first regulation target, determine all related granular fertilizer varieties that are associated with the corresponding granular fertilizer varieties based on the mapping relationship of the change trend, and generate a list of related regulation predictions. For subsequent control periods, obtain a list of the gas components of the second control target that require control actions during the subsequent control periods; Calculate the time misalignment between the current control period and subsequent control periods.

[0014] Preferably, the calculation of the time misalignment between the current control period and subsequent control periods includes: The time misalignment is the difference between the start time of the subsequent control period and the start time of the current control period; When the time misalignment is positive, the subsequent control window is determined to be delayed; if the time misalignment is less than the length of the current control period, the two windows overlap, and the coexistence influence coefficient is calculated to correct the current control intensity as in the case of a negative value; if the time misalignment is greater than the length of the current control period, a window period is generated, the concentration value of each gas component in the gas component list of the second control target is extracted at the current moment, the degree of approximation between the concentration value and the upper limit of the corresponding body gas data range is calculated, and the execution time of the control action in the current control period is extended according to the degree of approximation. When the time misalignment is negative, it is determined that the subsequent control window is advanced. The concentration value of each gas component in the second control target gas component list at the current time is extracted, the concentration value of each gas component in the first control target gas component list at the current time is extracted, the ratio of the concentration value of the second control target gas component to the concentration value of the first control target gas component is calculated, and the ratio is used as the coexistence influence coefficient to correct the control action execution parameters in the current control period.

[0015] Preferably, the step of extending the execution duration of the control action within the current control period based on the degree of approximation includes: the formula for calculating the extension amount is as follows: ; in, This represents the extension of the execution time of the regulatory action within the current regulatory period. This is the baseline value for the originally planned execution duration of the control action within the current control period. This represents the measured concentration of the gas component at the current moment. This represents the upper limit of the normal concentration range of this gas component in the bulk gas data. This is the hazardous concentration threshold for this gas component. This is a dimensionless correction factor.

[0016] The beneficial effects are as follows: This invention constructs a time offset feature quantity with the adjustment symmetry point and the external environment balance point as the core by synchronously collecting internal and external environmental data and granular fertilizer demand data during the storage process, and introduces an overlapping conflict compensation mechanism for gas release sensitive periods to form a method for precise quantitative control periods; by establishing a time-delay correlation mapping network for gas release of multiple fertilizer varieties, and dynamically correcting the execution duration or execution parameters of the current control action according to the direction of the subsequent control window delay or advance, it effectively overcomes the defects of the timing and state system mismatch of the timing and state of the time-based and threshold control strategies in mobile storage and multi-fertilizer mixed storage scenarios; finally, while improving the accuracy of storage environment regulation, it significantly enhances the collaborative adaptive capability of the process control system to complex environmental fluctuations and multi-source gas time-series evolution. Attached Figure Description

[0017] Figure 1 This is a flowchart of the adaptive adjustment method for the storage environment of granular fertilizer according to the present invention; Figure 2 This is a flowchart of the time misalignment determination and control parameter adjustment of the present invention. Detailed Implementation

[0018] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the present invention.

[0019] An adaptive adjustment method for the storage environment of granular fertilizer, such as Figures 1-2 As shown, it includes the following steps: S1: Collect data on changes in the external and internal environments of the warehouse during the storage process, and collect real-time data on the demand for granular fertilizer inside the warehouse. It should be noted that the movement of granular fertilizer during its transport in vehicles, railway carriages, or mobile silos constitutes the storage process. Meteorological sensors installed on the outer wall of the carrier continuously acquire data on changes in the external environment, including air temperature, relative humidity, solar irradiance, and wind speed. Temperature, humidity, and pressure sensors installed on the inner wall of the carrier and in the gaps between the fertilizer piles collect data on the internal environment, specifically internal air temperature, relative humidity, absolute humidity, and air pressure. This simultaneous acquisition of internal and external environmental data provides a dual-state reference for control decisions, preventing adjustments from being triggered solely by a single internal parameter, thus missing opportunities to utilize external drying or cooling windows. The granular fertilizer's own demand data is monitored using matrix thermocouples and humidity probes embedded within the pile. These serve to determine the upper limits of temperature, humidity, and gas concentration necessary for maintaining the fertilizer's physicochemical stability during the current storage stage. This demand data transforms the adjustment threshold from a fixed empirical value into a dynamic constraint linked to the fertilizer type and storage age.

[0020] S2: Determine the control period based on data on changes in the external environment of the warehouse, data on changes in the internal environment of the warehouse, and data on the demand for the warehouse itself, including: Determine the warehouse's self-adjustment cycle based on data on changes in the internal environment and the warehouse's own demand data; It should be noted that the determination of the warehouse's self-adjustment cycle relies on continuous comparative analysis of internal environmental change data and intrinsic demand data. When internal environmental parameters exceed the safety limits defined by intrinsic demand data and the cumulative value of the deviation over time reaches a preset trigger condition, the preset trigger condition is set as follows: Accelerated aging experiments are conducted on each type of granular fertilizer, measuring its physicochemical property changes at different degrees and durations exceeding the safety limits. The product of the excess magnitude and duration corresponding to the onset of irreversible changes in physicochemical properties is determined as the critical cumulative value. The preset trigger condition is taken within the range of 50% to 80% of this critical cumulative value. The cumulative value is a fixed value; for the temperature parameter, the cumulative value is the integral of the portion exceeding the upper limit of temperature over time; for the humidity parameter, the cumulative value is the integral of the portion exceeding the upper limit of relative humidity over time; when the cumulative value of any parameter reaches the preset trigger condition, it is determined that a control action needs to be performed; the trigger time is recorded, and the interval between two consecutive trigger times is defined as the storage self-regulation cycle; this cycle reflects the time span taken for the internal heat and humidity or gas concentration state of granular fertilizer under storage conditions to change from the safe range to the critical state where regulation intervention is necessary, thus clarifying the rhythmic outline of the internal regulation needs of the process control system and establishing a time-series reference guided by endogenous drive.

[0021] The phase transition cycle of the external environment is determined based on data on changes in the external and internal environments of the warehouse, including: Extract each external environmental parameter from the data on changes in the external environment of the warehouse, and extract each internal environmental parameter from the data on changes in the internal environment of the warehouse. When the changing trend of external environmental parameters has a positive impact on the internal environment of the warehouse or the demand for granular fertilizer, the corresponding dimension of the external environmental parameters is determined to be in a matching state; when the changing trend of external environmental parameters has a negative impact or no impact on the internal environment of the warehouse or the demand for granular fertilizer, the corresponding dimension of the external environmental parameters is determined to be in a mismatch state. It should be noted that the determination of positive and negative impacts is achieved through the following quantitative rules: For each external environmental parameter, a pre-defined calibration table is set to indicate its impact direction on the internal storage environment and fertilizer demand. For example, for the temperature parameter, when the external temperature is lower than the internal temperature and the temperature difference is greater than 2°C, it is determined to be a positive impact (beneficial for heat dissipation); when the external temperature is higher than the internal temperature, it is determined to be a negative impact. For the humidity parameter, when the external absolute humidity is lower than the internal absolute humidity, it is determined to be a positive impact (beneficial for drying), and vice versa. When multiple external environmental parameters change simultaneously and their impact directions conflict (such as positive temperature and negative humidity), the contribution weight of each parameter to the overall state of the internal environment is calculated separately, and the weight is determined by the partial correlation coefficient between the parameter and the state of the internal environment. If the number of positive impact dimensions is greater than the number of negative impact dimensions, the overall state is determined to be a matching state; otherwise, it is a mismatch state. The partial correlation coefficient is obtained through regression of historical data and is updated periodically.

[0022] The matching results of all external environment parameter dimensions are summarized to form the matching state vector at the current moment; The matching state vectors are continuously recorded at fixed time intervals to form a matching state vector sequence. The number of dimensions of the matching state at each time step in the matching state vector sequence is counted to obtain the curve of the number of dimensions of the matching state changing over time. Mark the moment when the curve changes from a continuous decline to a continuous rise as the starting moment of the beneficial period of the external environment; Mark the moment when the curve changes from continuous rise to continuous fall as the end of the beneficial period of the external environment; Calculate the time length between the start time and the end time of the beneficial external environment interval, and determine the time length as the duration of a single beneficial external environment phase transition; Record the duration of all single beneficial phase transitions in the external environment within the preset observation period, calculate the average duration of all single beneficial phase transitions in the external environment, and determine the average duration as the external environment phase transition period.

[0023] It should be noted that a matching relationship is formed when the evolution trend of external environmental parameters is consistent with the direction of internal storage demand; when they deviate or are irrelevant, they are classified as mismatched. The purpose of performing matching judgment dimensionally is to identify components that have a positive benefit to the storage of granular fertilizer from multi-source environmental parameters. The set of matching situations of all parameter dimensions at the same point in time constitutes the matching state vector, and the vector sequence recorded continuously at a predetermined time step clarifies the extension trajectory of the comprehensive beneficial attributes of the external environment on the time axis. The curve drawn based on the number of dimensions of the matching state shows the rhythm of the rise and fall of the degree of benefit. The inflection point of the curve from continuous decay to continuous rise indicates the boundary of the external environment entering the usable range, while the inflection point from continuous rise to continuous decay indicates the beginning of the external environment withdrawing from the usable range. The time length between the two inflection points is defined as a single external The duration of beneficial phase transitions in the external environment is determined by establishing the arithmetic mean of the durations of all beneficial phase transitions within a preset observation window as the external environment phase transition cycle. The length of the preset observation period is determined by the following method: collecting historical data corresponding to at least one complete external environment phase transition cycle, and using an integer multiple of the cycle length as the observation period; when there is no historical data, using 24 hours as the initial observation period, and after accumulating 3 beneficial phase transition events, recalculating and updating the observation period to the total duration spanned by the currently recorded events; this is used to mitigate the impact of occasional meteorological disturbances on the accuracy of cycle identification; the external environment phase transition cycle reveals the cyclical pattern of external natural conditions exerting a beneficial effect on the storage interior. For example, in the Gobi Desert, from summer evening to midnight, a decrease in absolute humidity and a drop in temperature to within the fertilizer tolerance range constitute a beneficial phase transition.

[0024] The control period is determined based on the warehouse's own adjustment cycle and the phase transition cycle of the external environment, including: Extract the adjustment symmetry point of the warehouse's self-adjustment cycle. The adjustment symmetry point is the moment corresponding to the maximum deviation of the internal environmental parameters of the warehouse from the preset target value within the warehouse's self-adjustment cycle. The adjustment symmetry point represents the moment of the most urgent action center of the warehouse's internal adjustment needs. The equilibrium point of the phase transition cycle of the external environment is extracted. The equilibrium point is the moment when the absolute value of the rate of change of the external environment parameters is the smallest within the beneficial range of the external environment. The equilibrium point represents the central moment when the external environment has the most stable positive impact on the warehouse interior. Calculate the first time offset between the adjustment symmetry point and the equilibrium point. The first time offset is the signed difference obtained by subtracting the equilibrium point time from the time of adjustment symmetry point. Identify the sensitive period for gas release from granular fertilizers; The gas release sensitive period is the duration during which the concentration of characteristic gases released by granular fertilizer reaches its peak during storage; this period usually corresponds to the stage when the fertilizer's hygroscopic or decomposition reaction is most intense. Calculate the second time offset between the equilibrium point and the center of the gas release sensitive period; When the second time offset is less than the preset overlap threshold, it is determined that there is an overlap and conflict between the equilibrium point and the gas release sensitive period. A conflict offset compensation amount is generated. The conflict offset compensation amount is the minimum time length required to translate the equilibrium point along the time axis to the outside of the boundary of the gas release sensitive period. The magnitude of the conflict offset compensation amount is the minimum time length required, and its sign is the sign corresponding to the translation direction (positive for translation in the positive direction of the time axis, negative for translation in the negative direction). The first time offset is superimposed with the conflict offset compensation to obtain the total time offset. Specifically, the candidate control time is obtained by adding the first time offset to the equilibrium point time. If the candidate control time is within the gas release sensitive period, the total time offset is the conflict offset compensation; otherwise, the total time offset is the first time offset. Using the equilibrium point as a reference, the total time offset is added to the equilibrium point to obtain the start time of the control period; the length of the control period is equal to the original control action execution time.

[0025] It should be noted that the adjustment symmetry point of the storage's self-adjustment cycle is the moment corresponding to the maximum deviation of internal environmental parameters from the preset target value. This moment reflects the most urgent action center of internal adjustment needs. The preset target value is the upper limit of temperature, upper limit of humidity, or safe boundary of gas concentration specified in the granular fertilizer's own demand data, or the recommended storage parameters provided by the fertilizer supplier. When there is no clear data, the historical average value of the storage's internal environmental parameters is taken. The equilibrium point of the external environment phase change cycle is taken from the moment when the absolute value of the rate of change of environmental parameters within the beneficial range is the smallest, used to identify the time pivot where the external gain effect exerts the most stable influence on the storage interior. The first time offset characterizes the time difference between the adjustment symmetry point and the equilibrium point. An increase in the absolute spacing on the axis indicates a deeper disconnect between the most critical point of internal regulation and the point of stable external gain, while a decrease suggests that the two are gradually converging towards a synchronized state. The introduction of a gas release-sensitive period is because the tolerance margin of granular fertilizer to external environmental intervention is narrowest during the peak concentration of characteristic gases. The specific method for obtaining this period is to retrospectively record the gas concentration time series during historical storage and extract the continuous span of concentration exceeding a preset proportional threshold. The preset proportional threshold is determined by: collecting historical concentration time series data of the characteristic gases during the storage process of the granular fertilizer, plotting a curve of concentration change over time, and identifying the concentration value corresponding to the inflection point where the rate of increase in concentration first exceeds the rate of decrease in the curve. Using the reference point as a reference, the ratio of the concentration at this reference point to the peak concentration is taken as the proportional threshold. When there is no historical data, the full width at half maximum (FWHM) method is used to define the sensitive period, i.e., the continuous interval where the concentration exceeds half of the peak value. The second time offset is defined as the deviation of the equilibrium point time relative to the center time of the gas release sensitive period. An increase in the second time offset indicates that the equilibrium point is moving away from the peak gas release interval, and the probability of overlap and conflict decreases. Conversely, a decrease indicates that the equilibrium point is approaching the peak release region. The preset overlap threshold is constructed by statistically analyzing the minimum interval between the equilibrium point and the center of the gas release sensitive period in multiple rounds of beneficial phase change events that has not triggered a conflict, and selecting quantile values ​​as critical thresholds. For example, the 75th quantile is selected as the critical threshold, and statistics are compiled up to... Ten fewer beneficial phase transition events; once the second time offset is lower than the preset overlap threshold, an overlap conflict is confirmed. The conflict offset compensation is set as the minimum time shift required to shift the equilibrium point to the outside of the gas release sensitive period boundary. When the gas release sensitive period lasts too long, causing the equilibrium point after shifting according to the above method to fall outside the beneficial range of the external environment, the following boundary processing is performed: calculate the distance between the shifted equilibrium point and the boundary of the beneficial range of the external environment. If the distance exceeds 50% of the total length of the beneficial range, abandon the shift, maintain the original equilibrium point position, and actively strengthen ventilation or adsorption measures to reduce the gas concentration peak within the current control cycle; at the same time, generate early warning information to prompt operators to intervene manually.Furthermore, if the translation failure boundary condition is triggered in three consecutive control events, the current sensitive period judgment parameter (preset proportional threshold) is deemed improperly set. The preset proportional threshold is automatically adjusted downwards by 10 percentage points, and the gas release sensitive period is recalibrated. The translation direction is selected as either positive (delayed) or negative (advanced) on the time axis, with the smaller translation distance preferred. Specifically, the smaller of the time lengths required to move the equilibrium point to the outside of the left or right boundary of the sensitive period is chosen. If the translation distances on both sides are equal, the default is to translate in the negative (advanced) direction on the time axis. This selection principle aims to avoid the peak gas release period with minimal control disturbance. The total time offset is determined according to the following rules: Using the equilibrium point as a reference, the candidate control time is obtained by adding the first time offset. If the candidate control time falls within the gas release sensitive period, the total time offset is the conflict offset compensation amount; otherwise, the total time offset is the first time offset. When the candidate control time falls exactly on the boundary of the sensitive period, it is considered to be within the sensitive period. When the total time offset is positive, the control action must be postponed; when it is negative, it must be advanced. The total time offset is the correction amount required for the internal regulation rhythm and the external gain rhythm to match in the time dimension. For example, in the Gobi Desert transportation scenario, when the center of the external beneficial window lags behind the internal regulation symmetry point by two hours, the total time offset is two hours.

[0026] S3: Collect mixed gas data of granular fertilizer during the control period, and determine the bulk gas data of different types of granular fertilizer according to the type of granular fertilizer in the storage. It should be noted that the total time offset indicates the time span of the misalignment between internal adjustment demands and favorable external conditions. The control period determined by this offset is the transitional state in which the warehouse is awaiting the execution of adjustment instructions. During the control period, mixed gas data is collected to capture the accumulation and evolution of gas components in granular fertilizer under closed or semi-closed conditions. The identification information on the fertilizer packaging is read by a barcode scanner or RFID reader at the warehouse entrance, or the type of granular fertilizer is obtained by the operator manually inputting the current fertilizer type and proportion on the human-machine interface. The mixed gas data is obtained by a gas sensor array deployed on the inner wall of the carrier, covering parameters such as ammonia concentration, carbon dioxide concentration, and volatile organic compound concentration. The gas data of different types of granular fertilizers are determined by searching a pre-established fertilizer gas benchmark database, which contains the characteristic gas categories and concentration limits that normally escape from each fertilizer type in a standardized storage environment. The defined gas data is used to set a benchmark for identifying mixed gas anomalies.

[0027] S4: Adaptive adjustment of the storage environment based on mixed gas data and bulk gas data, including: Extract the variation amplitude value of the bulk gas data corresponding to each type of granular fertilizer within the same control period. The variation amplitude value is the absolute value of the concentration difference of the bulk gas data at the start and end times of the control period. The granular fertilizer with the smallest change value is marked as the first associated fertilizer, and the granular fertilizer with the largest change value is marked as the second associated fertilizer. A trend mapping relationship between the first associated fertilizer and the second associated fertilizer is established. The trend mapping relationship is the time delay corresponding to the maximum value of the time delay correlation coefficient between the gas data of the first associated fertilizer and the gas data of the second associated fertilizer over time. Then, pair the granular fertilizer with the second smallest change value with the granular fertilizer with the second largest change value to establish a trend mapping relationship between the two; and so on, until all granular fertilizers are paired. If the number of fertilizer types is odd, the middle one will not participate in the pairing, thus obtaining the mapping relationship of all pairs. The time delay corresponding to the maximum value of the time delay correlation coefficient is specifically calculated using a cross-correlation function. ,in For delay quantity, and The data are time-series data of gas concentrations for two different fertilizers. This represents the average value of the gas concentration sequence for the first type of fertilizer. This represents the average value of the second type of fertilizer gas concentration sequence. For the second type of fertilizer at any time The gas concentration value, In order to delay The Pearson correlation coefficient is given, with a value range of [-1, 1].

[0028] It should be noted that the start and end points of the control period provide a unified time boundary for gas concentration measurement, ensuring the comparability of the variation amplitude values ​​and the accuracy of the time-series mapping relationship. Within the same control period, the intensity of gas release from different fertilizer varieties varies significantly. The variation amplitude values ​​of their respective intrinsic gas data are extracted to measure this degree of differentiation. The variation amplitude value is determined by taking the absolute value of the difference between the concentration at the end of the control period and the concentration at the start. The variety with the smallest variation amplitude is marked as the first associated fertilizer, and the largest is marked as the second associated fertilizer, used to establish the two extreme reference coordinates of the release behavior spectrum. The first associated fertilizer and the second associated fertilizer... The mapping relationship of changing trends among fertilizers is constructed by calculating the time-delay correlation coefficient of the gas concentration time series curves of the two fertilizers and extracting the time delay corresponding to the peak value. It aims to reveal the predictive or follow-up indication effectiveness of the fertilizer with the most gradual release rate for the fertilizer with the most intense release. Then, the fertilizers with the smallest and second largest changes are associated, and so on, until all fertilizers are paired. If there is an odd number of fertilizer types, the middle one is not paired and no mapping relationship is established. This forms a complete gas release time series correlation network among fertilizer types, providing a structured decision reference for the process control system to predict the pollution diffusion sequence under the condition of mixed storage of multiple fertilizers.

[0029] During subsequent control periods, real-time collection of the gas data for the current control period will be conducted, including: Update the mapping relationship of changing trends among all granular fertilizers based on the current bulk gas data; Obtain the list of gas components of the first target gas for which control actions need to be performed during the current control period; Extract the granular fertilizer varieties corresponding to each gas component in the list of gas components of the first regulation target, determine all related granular fertilizer varieties that are associated with the corresponding granular fertilizer varieties based on the mapping relationship of the change trend, and generate a list of related regulation predictions. For subsequent control periods, obtain a list of the gas components of the second control target that require control actions during the subsequent control periods; It should be noted that the update operation of the trend mapping relationship involves placing the real-time collected time-series data of the bulk gas concentration during the current control period into the modeling data pool to replace the old records and re-perform the time-delay correlation coefficient calculation and time-delay extreme value extraction. The purpose of this update is to eliminate the correlation deviation caused by environmental condition drift or fertilizer age extension, and to maintain the ability of the mapping network to follow the actual evolution law. The first control target gas component list specifically refers to the compilation of gas components whose concentrations are close to the safety limit and therefore require control intervention during the current control period, as confirmed by comparison between the mixed gas and the bulk gas. Specifically, it includes gas components whose current concentration exceeds the upper limit of the normal concentration. A certain proportion (this proportion is preset according to fertilizer safety requirements), or gaseous components whose concentration rise rate exceeds a preset threshold, are included in the first list of controlled targets. The formation path of the associated control prediction list is to first extract the fertilizer variety to which each gaseous component in the first list of controlled targets belongs, then retrieve all fertilizer varieties with time-series coupling relationships with that variety through the trend mapping relationship, and finally aggregate the gaseous components corresponding to the associated varieties to generate a prediction list. The purpose of this prediction list is to pre-identify the associated gaseous components whose concentration is very likely to fluctuate due to the linkage effect after the implementation of the current control action. The technical basis for re-acquiring the second list of controlled gaseous components for subsequent control periods is that the composition of gaseous components that urgently need to be controlled in the next control period will increase or decrease due to the secondary disturbance of the current control, so they are extracted independently to ensure the coordination of cross-cycle decision-making. Specifically, the established time-series mapping relationship (time delay) is used. Predict the concentration of related commodities at the start of the next regulatory period. If the predicted value exceeds the upper limit of the normal concentration... If a certain proportion (the same preset proportion as the first list) is included in the second list of control targets, then if there is no mapping relationship, then a time series forecasting method (such as linear extrapolation) is used for forecasting.

[0030] Calculate the time misalignment between the current control period and subsequent control periods, including: The time misalignment is the difference between the start time of the subsequent control period and the start time of the current control period; When the time misalignment is positive, the subsequent control window is delayed. If the time misalignment is less than the current control period length, the two windows overlap, and the coexistence influence coefficient is calculated to correct the current control intensity as in the negative case. If the time misalignment is greater than the current control period length, a window period is generated. The concentration value of each gas component in the second control target gas component list is extracted at the current moment, and the approximation degree between the concentration value and the upper limit of the corresponding bulk gas data range is calculated. Based on the approximation degree, the execution duration of the control action within the current control period is extended. The extension calculation formula is as follows: ; in, This represents the extension of the execution time of the regulatory action within the current regulatory period. This is the baseline value for the originally planned execution duration of the control action within the current control period. This represents the measured concentration of the gas component at the current moment. This represents the upper limit of the normal concentration range of this gas component in the bulk gas data. This is the hazardous concentration threshold for this gas component. This is a dimensionless correction factor; It should be noted that the formula for calculating the extension amount requires... That is, the dangerous concentration threshold is higher than the upper limit of the normal concentration, and only when... The formula takes effect only when the time is right; otherwise, it is not applicable. At this time, the current concentration has not exceeded the normal upper limit, and there is no need to extend the duration of the control action. The dimensionless correction coefficient The initial value is determined by fitting historical data: the logarithmic function form in the extension formula is based on the following approximate derivation: assuming the gas concentration decays exponentially under regulation, the time required to reach the safety threshold is logarithmically related to the initial excess concentration; this model is suitable for scenarios where the concentration excess is not extreme; actual extension amount The measurement method is as follows: After executing the control action for the original timing within the current control period, continue to monitor the gas concentration and record the concentration from the start of the control to below [a certain value]. The total duration, minus the original duration. The difference obtained is If the concentration rebounds again after the regulation period ends, exceeding [a certain threshold]... The time when the second descent after the rebound is completed shall be taken as the standard; the measured value... For subsequent λ updates, the measured duration required to extend the duration during several control events is collected and calculated according to the formula (taking...). =1) The calculated theoretical duration, using the average of the two ratios as the initial... Value; after each subsequent adjustment, the value is updated using an exponentially weighted moving average method based on the deviation between the actual and predicted effects. value: ,in This is a smoothing factor, for example, between 0.7 and 0.9. This is the actual extended duration. To be in accordance with the current The calculated theoretical extension time; Determination method: For each type of granular fertilizer, a static storage experiment was conducted under standard storage conditions (temperature 25±2℃, relative humidity 50±5%). The concentration of characteristic gases was continuously monitored, and the 95th percentile of the concentration values ​​in the first 72 hours was taken as the upper limit of the normal concentration range. ; The method for determining the hazardous concentration threshold is as follows: Accelerated degradation experiments are conducted on the same type of fertilizer, gradually increasing the concentration until irreversible changes occur in the fertilizer's physicochemical properties (such as pH value and total nutrient content). The gas concentration corresponding to the critical point of change is taken as the hazardous concentration threshold. For example, when the pH value changes by more than ±0.5, or the total nutrient content decreases by more than 5%, it is determined that an irreversible change has begun; if there are no destructive experimental conditions, the safety limit of the gas component in the national or industry standard shall be used to determine the value.

[0031] When the time misalignment is negative, it is determined that the subsequent control window is advanced. The concentration value of each gas component in the second control target gas component list at the current time is extracted, the concentration value of each gas component in the first control target gas component list at the current time is extracted, the ratio of the concentration value of the second control target gas component to the concentration value of the first control target gas component is calculated, and the ratio is used as the coexistence influence coefficient to correct the control action execution parameters in the current control period.

[0032] It should be noted that the reference Figure 2The time misalignment is calculated by subtracting the start time of the current control period from the start time of the subsequent control period, representing the relative displacement of two adjacent control windows on the time axis. When the time misalignment is positive, the start time of the subsequent window is later than the start time of the current window. If the time misalignment is less than the length of the current control period, the two windows overlap, and the coexistence influence coefficient should be calculated to correct the current control intensity as in the case of a negative value. If the time misalignment is greater than the length of the current control period, a window period occurs. In this case, the concentration value of the second controlled gas component at the current moment is extracted, and the degree of approximation between the concentration value and the upper limit of the corresponding bulk gas data range is calculated. Based on the degree of approximation, the control action within the current control period is extended. The duration of the operation; the degree of approximation is specifically the proportion of the current concentration exceeding the upper limit of normal to the difference between the danger threshold and the upper limit of normal. Its function is to measure the urgency of the gas concentration exceeding the danger threshold in the subsequent window if no early intervention is taken. By strengthening the adjustment intensity of the current cycle, the cumulative effect of the subsequent window delay is partially offset. The extension amount is calculated using a logarithmic function, so that the extension magnitude increases with the degree of approximation but the rate of increase slows down, avoiding over-adjustment. When the time misalignment is negative, it indicates that there is a time overlap between the subsequent control window and the current control window (i.e., the start time of the subsequent window is earlier than the end time of the current window), that is, the two control periods will be executed simultaneously or continuously, forming a repeated control area. The ratio of the concentration of the second controlled gas component to the concentration of the first controlled gas component needs to be calculated to quantify the superposition effect. An increase in the ratio indicates a rise in the relative concentration of the gas to be controlled in subsequent windows, strengthening the cascading effect on the current control; a decrease in the ratio indicates a weakening cascading effect. This ratio is used as a coexistence influence coefficient to correct the execution parameters of the current control action. The coexistence influence coefficient reflects the degree of mutual interference when two gases coexist. The ratio refers to the ratio of the concentration of the same gas component in the second list to its concentration in the first list, for each gas component appearing simultaneously in both the first and second controlled gas lists. If multiple gas components appear simultaneously in both lists... If the two lists do not contain the same gas composition, the coexistence influence coefficient is set to 1. The specific correction method is as follows: multiply the original control action execution parameters (including but not limited to ventilation rate, dehumidification power, heating power and air exchange duration) in the current control period by the coexistence influence coefficient to obtain the corrected execution parameters. If the coexistence influence coefficient is greater than 1, the control intensity is increased; if it is less than 1, the control intensity is weakened; if it is equal to 1, it remains unchanged. For example, the coexistence influence coefficient is limited to between 0.5 and 2.0. When it exceeds the limit, the boundary value is taken. When the coexistence influence coefficient is 1.3, the ventilation rate of the current control is adjusted to 1.3 times the original rate.

[0033] The method also includes a fallback mechanism for control failure: control failure is determined when any of the following occurs—(a) for example, during two consecutive control periods, the gas concentration continues to rise and exceeds [a certain value] after the control action is performed. (b) For example, if the maximum value of the time-delay correlation coefficient is less than 0.3, it indicates that the trend mapping relationship is unreliable; (c) For example, if the ratio of the actual extension to the theoretical extension exceeds the range of 0.5 to 2.0; When the control fails, the following fallback operation is automatically performed: stop the current adaptive adjustment logic, switch to the fixed frequency forced ventilation mode, for example, ventilate for 10 minutes every 30 minutes; at the same time, send a fault alarm to the management personnel through the audible and visual alarm or remote communication module, the alarm content including the failure type, relevant parameters and suggested inspection items; after manual reset, the adaptive adjustment mode is restored.

[0034] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for adaptive adjustment of the storage environment for granular fertilizer, characterized in that, Includes the following steps: S1: Collect data on changes in the external and internal environments of the warehouse during the storage process, and collect real-time data on the demand for granular fertilizer inside the warehouse. The intrinsic demand data characterizes the upper limits of temperature, humidity, and gas concentration safety boundaries necessary for fertilizers to maintain physicochemical stability during the current storage stage. S2: Determine the control period based on data on changes in the external environment, internal environment, and the warehouse's own demand. This includes: determining the warehouse's self-adjustment cycle based on data on changes in the internal environment and the warehouse's own demand; determining the external environment phase transition cycle based on data on changes in the external and internal environments; and determining the control period based on the warehouse's self-adjustment cycle and the external environment phase transition cycle. This includes: extracting the adjustment symmetry point of the warehouse's self-adjustment cycle, where the adjustment symmetry point is the moment corresponding to the maximum deviation of the warehouse's internal environmental parameters from the preset target value within the warehouse's self-adjustment cycle, and the adjustment symmetry point represents the most urgent action center moment for the warehouse's internal adjustment needs; extracting the equilibrium point of the external environment phase transition cycle, where the equilibrium point is the moment with the smallest absolute value of the rate of change of external environmental parameters within the beneficial range of the external environment, and the equilibrium point represents the center moment when the external environment has the most stable positive impact on the warehouse's interior; and calculating the first time offset between the adjustment symmetry point and the equilibrium point, where the first time offset is the adjustment symmetry point. The signed difference between the equilibrium point time and the equilibrium point time is used to obtain the gas release sensitive period of the granular fertilizer. The gas release sensitive period is the duration range during which the concentration of characteristic gases released by the granular fertilizer reaches its peak during storage. The second time offset between the equilibrium point time and the center time of the gas release sensitive period is calculated. When the second time offset is less than a preset overlap threshold, it is determined that there is an overlap conflict between the equilibrium point and the gas release sensitive period, and a conflict offset compensation amount is generated. The conflict offset compensation amount is the minimum time length required to shift the equilibrium point along the time axis to the outside of the boundary of the gas release sensitive period. The candidate control time is obtained by adding the first time offset to the equilibrium point time. If the candidate control time is located within the gas release sensitive period, the total time offset is the conflict offset compensation amount; otherwise, the total time offset is the first time offset. The total time offset is added to the equilibrium point time to obtain the start time of the control period. The length of the control period is equal to the original control action execution time. S3: Collect mixed gas data of granular fertilizer during the control period, and determine the bulk gas data of different types of granular fertilizer according to the type of granular fertilizer in the storage; the bulk gas data is used to set a reference scale for the identification of mixed gas anomalies. S4: Adaptive adjustment of the storage environment based on mixed gas data and bulk gas data.

2. The adaptive adjustment method for the storage environment of granular fertilizer as described in claim 1, characterized in that, The process of determining the phase transition period of the external environment based on changes in the external and internal environments of the warehouse includes: Extract each external environmental parameter from the data on changes in the external environment of the warehouse, and extract each internal environmental parameter from the data on changes in the internal environment of the warehouse. When the changing trend of external environmental parameters has a positive impact on the internal environment of the warehouse or the demand for granular fertilizer, the corresponding dimension of the external environmental parameters is determined to be in a matching state; when the changing trend of external environmental parameters has a negative impact or no impact on the internal environment of the warehouse or the demand for granular fertilizer, the corresponding dimension of the external environmental parameters is determined to be in a mismatch state. The matching results of all external environment parameter dimensions are summarized to form the matching state vector at the current moment; The matching state vectors are recorded continuously at fixed time intervals to form a matching state vector sequence.

3. The adaptive adjustment method for the storage environment of granular fertilizer as described in claim 2, characterized in that, The step of continuously recording the matching state vector at fixed time intervals to form a matching state vector sequence includes: The number of dimensions of the matching state at each time step in the matching state vector sequence is counted to obtain the curve of the number of dimensions of the matching state changing over time. Mark the moment when the curve changes from a continuous decline to a continuous rise as the starting moment of the beneficial period of the external environment; Mark the moment when the curve changes from continuous rise to continuous fall as the end of the beneficial period of the external environment; Calculate the time length between the start time and the end time of the beneficial external environment interval, and determine the time length as the duration of a single beneficial external environment phase transition; Record the duration of all single beneficial phase transitions in the external environment within the preset observation period, calculate the average duration of all single beneficial phase transitions in the external environment, and determine the average duration as the external environment phase transition period.

4. The adaptive adjustment method for the storage environment of granular fertilizer as described in claim 1, characterized in that, The adaptive adjustment of the storage environment based on mixed gas data and bulk gas data includes: Extract the variation amplitude value of the bulk gas data corresponding to each type of granular fertilizer within the same control period. The variation amplitude value is the absolute value of the concentration difference of the bulk gas data at the start and end times of the control period. The granular fertilizer with the smallest change value is marked as the first associated fertilizer, and the granular fertilizer with the largest change value is marked as the second associated fertilizer. A trend mapping relationship between the first associated fertilizer and the second associated fertilizer is established. The trend mapping relationship is the time delay corresponding to the maximum value of the time delay correlation coefficient between the gas data of the first associated fertilizer and the gas data of the second associated fertilizer over time. Then, pair the granular fertilizer with the second smallest change value with the granular fertilizer with the second largest change value to establish a trend mapping relationship between the two; and so on, until all granular fertilizers are paired. If the number of fertilizer types is odd, the middle one will not participate in the pairing, thus obtaining the mapping relationship of all pairs. During subsequent control periods, the gas data of the current control period will be collected in real time.

5. The adaptive adjustment method for the storage environment of granular fertilizer as described in claim 4, characterized in that, The process of collecting real-time gas data during the current control period during subsequent control periods includes: Update the mapping relationship of changing trends among all granular fertilizers based on the current bulk gas data; Obtain the list of gas components of the first target gas for which control actions need to be performed during the current control period; Extract the granular fertilizer varieties corresponding to each gas component in the list of gas components of the first regulation target, determine all related granular fertilizer varieties that are associated with the corresponding granular fertilizer varieties based on the mapping relationship of the change trend, and generate a list of related regulation predictions. For subsequent control periods, obtain a list of the gas components of the second control target that require control actions during the subsequent control periods; Calculate the time misalignment between the current control period and subsequent control periods.

6. The adaptive adjustment method for the storage environment of granular fertilizer as described in claim 5, characterized in that, The calculation of the time misalignment between the current control period and subsequent control periods includes: The time misalignment is the difference between the start time of the subsequent control period and the start time of the current control period; When the time misalignment is positive, the subsequent control window is determined to be delayed; if the time misalignment is less than the length of the current control period, the two windows overlap, and the coexistence influence coefficient is calculated to correct the current control intensity as in the case of a negative value; if the time misalignment is greater than the length of the current control period, a window period is generated, the concentration value of each gas component in the gas component list of the second control target is extracted at the current moment, the degree of approximation between the concentration value and the upper limit of the corresponding body gas data range is calculated, and the execution time of the control action in the current control period is extended according to the degree of approximation. When the time misalignment is negative, it is determined that the subsequent control window is advanced. The concentration value of each gas component in the second control target gas component list at the current time is extracted, the concentration value of each gas component in the first control target gas component list at the current time is extracted, the ratio of the concentration value of the second control target gas component to the concentration value of the first control target gas component is calculated, and the ratio is used as the coexistence influence coefficient to correct the control action execution parameters in the current control period.

7. The adaptive adjustment method for the storage environment of granular fertilizer as described in claim 6, characterized in that, The method of extending the execution duration of the control action within the current control period based on the degree of approximation includes: the formula for calculating the extension amount is as follows: ; in, This represents the extension of the execution time of the regulatory action within the current regulatory period. This is the baseline value for the originally planned execution duration of the control action within the current control period. This represents the measured concentration of the gas component at the current moment. This represents the upper limit of the normal concentration range of this gas component in the bulk gas data. This is the hazardous concentration threshold for this gas component. This is a dimensionless correction factor.

Citation Information

Patent Citations

  • Microbial action monitoring and analyzing system

    CN120927758A

  • Garden automatic irrigation regulation and control method and system

    CN121400333A