A quantitative model of phytophthora and a method for analyzing the disease by fusing multiple environmental factors

Abstract

The application discloses a kind of pythium quantitative model and blight analysis method of fusing multiple environmental factors, belong to agricultural disease monitoring and intelligent early warning field, to solve the existing pythium blight quantitative evaluation factor dimension deficiency, leaf effective wetting time determination is not accurate and hour-level multidimensional data is difficult to be automatically processed, the present application obtains and standardizes air temperature, relative humidity, precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure and dew point temperature etc. Hour sequence is constructed Environmental factor fusion vector, multi-factor coupling wetting determination is carried out using evidence scoring mechanism, and the effective leaf wetting time of single day is determined by hysteresis state machine, which is calculated with the nonlinear pythium quantitative model of multiple environmental factor input including factor interaction, dynamic weight and persistence index Disease unit and output risk warning, realize the technical effect that wetting statistics is more stable, evaluation is more comprehensive and warning is more timely.

Description

Technical Field

[0001] This invention relates to the field of agricultural disease monitoring and intelligent early warning, and in particular to a quantitative model of disease outbreaks and a method for disease analysis that integrates multiple environmental factors. Background Technology

[0002] Phytophthora blight, a significant disease affecting many economic crops, is closely related to air temperature, relative humidity, precipitation, leaf moisture, light conditions, and soil environment. With the development of agricultural meteorological monitoring and IoT technology, existing Phytophthora blight risk assessment techniques have gradually evolved from qualitative judgments based on human experience to quantitative assessments using environmental data collected from weather stations or field sensors and then performing model calculations. Current methods typically use hourly or daily indicators such as temperature, relative humidity, precipitation, and duration of leaf moisture, employing threshold rules, cumulative indices, or simplified statistical models to output disease risk levels, providing a reference for disease early warning and control.

[0003] However, existing technologies still have the following shortcomings:

[0004] 1. Insufficient inclusion of environmental factors. Most models rely mainly on a few factors such as air temperature and humidity, precipitation or leaf wet duration, and rarely systematically introduce factors that have a significant impact on spore germination and spread and crop susceptibility, such as sunshine duration, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature. This results in incomplete assessment dimensions and insufficient ability to adapt to different regions and different weather processes.

[0005] 2. Insufficient accuracy in determining the effective wetting duration of leaves. Some methods only approximate the wetting state using precipitation or relative humidity thresholds, lacking multi-factor coupling determination of the formation mechanisms of condensation-type wetting and continuous wetting, and also lacking correction for wetting continuation under low light conditions. This easily leads to problems such as unstable determination of the start and end times of wetting and statistical distortion of continuous processes across days.

[0006] 3. The automation level of hourly multidimensional data processing is not high. Existing methods often rely on manual statistics, tabular calculations, or simple rules. When faced with multi-monitoring point data on a continuous hourly scale, it is difficult to achieve missing data repair, anomaly correction, trend feature extraction, and batch calculation. This results in low efficiency and is prone to introducing errors, thus affecting the accuracy and timeliness of Phytophthora blight risk assessment and early warning.

[0007] Therefore, there is a need for a quantitative model for epidemic fungi and a method for epidemic analysis that can address the shortcomings of existing technologies. Summary of the Invention

[0008] One objective of this invention is to propose a quantitative model and analysis method for Phytophthora blight that integrates multiple environmental factors. This addresses the problem that existing quantitative assessments of Phytophthora blight often rely solely on a few factors such as air temperature and humidity, precipitation, or leaf wetness duration, failing to adequately consider key environmental factors such as sunshine duration, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature. This leads to inaccurate determination and statistical analysis of effective leaf wetness duration, and the processing of hourly multidimensional data relies on manual statistics or simple table calculations, making automated batch processing difficult. Therefore, this invention proposes a multi-environmental factor fusion and analysis method based on continuous hourly-scale multidimensional environmental data. A hierarchical coupling judgment technology scheme is proposed: By aligning, repairing missing data, and unifying units for air temperature, relative humidity, precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature over time, an environmental factor fusion vector is constructed. An evidence scoring mechanism is used to perform multi-factor coupling judgment for direct precipitation wetting, condensation-type wetting, persistent wetting, and low-light calibration. A hysteretic state machine is used to determine the effective daily leaf surface wetting duration. This duration is then combined with various environmental factors into a nonlinear Phytophthora blight quantification model containing interaction terms, dynamic weights, and a persistence index to calculate disease units and output risk warnings. This invention offers the technical advantages of more comprehensive evaluation dimensions, more stable and accurate wetting judgment, automated batch calculation of multiple monitoring points at the hourly level, and improved timeliness of Phytophthora blight risk warnings.

[0009] This invention provides a quantitative model for fungal diseases that integrates multiple environmental factors and a method for disease analysis, including:

[0010] S1. Obtain multidimensional environmental data of the area to be analyzed on a continuous hourly scale, including air temperature, relative humidity, precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure and dew point temperature, perform time alignment, missing data repair and unit unification to obtain a standardized environmental sequence.

[0011] S2. Construct a multi-environmental factor fusion system based on standardized environmental sequences to obtain the environmental factor fusion vector corresponding to each hour;

[0012] S3. Determine the hourly humidity candidate value by performing multi-factor coupled humidity state determination based on the environmental factor fusion vector;

[0013] S4. Determine the hourly humidity status based on the hourly humidity candidate value, and obtain the effective leaf surface humidity duration on a daily basis.

[0014] S5. Input the effective leaf surface wet duration and corresponding environmental factors into the nonlinear Phytophthora quantification model to calculate the daily disease unit value;

[0015] S6. Generate disease risk analysis results and early warning results based on the daily disease unit value and its cumulative over time.

[0016] Optionally, S1 includes:

[0017] Collect or read environmental data from one or more monitoring points within the area to be analyzed on an hourly basis;

[0018] Data from different sources are matched according to a unified time identifier to form hourly data records;

[0019] For short-term missing data, interpolation is performed using adjacent hours or adjacent monitoring points on the same day; amplitude-limiting correction is applied to abnormal and sudden data.

[0020] By unifying precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature into preset dimensions, a standardized environmental sequence is obtained.

[0021] Optionally, S2 includes:

[0022] Air temperature, relative humidity, precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature are respectively mapped to dimensionless environmental factors;

[0023] The dimensionless environmental factors are combined according to the same hour time identifier to form an environmental factor fusion vector;

[0024] Factor values ​​that exceed the effective range for Phytophthora spore germination, spread, or crop susceptibility are truncated or weighted down.

[0025] The system calculates the temperature, humidity, air pressure, and light intensity changes in adjacent hours to form a multi-environmental factor fusion system that includes the current environmental state and its changing trends.

[0026] Optionally, S3 includes:

[0027] When the precipitation reaches the preset precipitation threshold, the corresponding hour will be marked as evidence of direct wetting by precipitation.

[0028] When the difference between air temperature and dew point temperature is less than the condensation threshold and the soil temperature meets the conditions for promoting condensation, the corresponding hour is marked as evidence of condensation-type wetting.

[0029] When changes in air pressure, insufficient sunshine duration, and relative humidity exceeding the humidity threshold together meet the conditions for continuous humidity, the corresponding hour is marked as evidence of continuous humidity.

[0030] When photosynthetically active radiation is below the weak light threshold, the intensity of wet evidence for the same hour is calibrated for weak light.

[0031] Hourly humidity candidate values ​​are generated based on the comprehensive results of various humidity evidences;

[0032] Furthermore, the multi-factor coupled wetting state determination employs an evidence scoring mechanism, including:

[0033] Base scores were set for evidence of direct precipitation wetting, evidence of condensation wetting, evidence of persistent wetting, and evidence of low light calibration.

[0034] The wet evidence score is obtained by weighting and summing the combinations of various wet evidence within the same hour.

[0035] When the wet evidence score reaches the wet confirmation threshold, a wet candidate value is output. When the wet evidence score is lower than the wet exclusion threshold, a non-wet candidate value is output. When the wet evidence score is between the wet confirmation threshold and the wet exclusion threshold, an intermediate evidence value is retained.

[0036] Optionally, S4 includes:

[0037] The hourly wetness candidate values ​​are input into the hysteresis state machine, and the hourly wetness state is determined based on the threshold for entering the wetness state and the threshold for exiting the wetness state.

[0038] The hours identified as being in a moist state are accumulated according to the calendar day to obtain the effective leaf surface moist duration for a single day;

[0039] When a continuous wet period spans two dates, the wet hours spanning those dates will be counted into the corresponding dates according to their actual time.

[0040] When low light calibration extends the wet state, the extended hours are included in the effective leaf wet duration for the corresponding date;

[0041] Furthermore, the hysteresis state machine includes a dry state, a wetness confirmation state, and a wetness maintenance state;

[0042] When the wet candidate value reaches the threshold for entering the wet state for one or more consecutive hours, the state switches from dry state to wet confirmation state.

[0043] When the wet candidate value continues to meet the maintenance condition, the state switches from wet confirmation state to wet maintenance state.

[0044] When the wetness candidate value is lower than the exit wetness state threshold and continues to reach the preset buffer time, the state is switched from wetness-maintaining state to dry state. The threshold for entering the wetness state is higher than the threshold for exiting the wetness state.

[0045] Optionally, S5 includes:

[0046] The effective leaf surface wet duration, average daily air temperature, average daily soil temperature, average daily relative humidity, daily cumulative precipitation, sunshine duration, average daily photosynthetically active radiation, average daily air pressure, and average daily dew point temperature are input into the nonlinear Phytophthora quantification model.

[0047] The environmental contribution was calculated using the temperature and humidity interaction term, the light inhibition term, the soil temperature adaptation term, the air pressure propagation term, and the time accumulation term, respectively.

[0048] The daily disease unit value is synthesized based on the dynamic weights of each environmental contribution;

[0049] Furthermore, the nonlinear phytotoxicity quantification model also includes a persistence index calculation process, which is calculated based on the number of consecutive wet hours, the number of times the effective leaf surface wetness continues on adjacent dates, the number of times the daily phytotoxicity unit value increases continuously, and the number of hours of weak light calibration.

[0050] The persistence index is used as an adjustment for the time-cumulative term to enhance or reduce the daily disease unit value, thereby characterizing the cumulative effect of a sustained suitable environment on Phytophthora spore germination, spread, and crop susceptibility.

[0051] Optionally, S6 includes:

[0052] The daily disease unit value is compared with the preset risk level threshold to generate low-risk, medium-risk and high-risk analysis results;

[0053] Identify periods of increased risk based on daily disease unit values ​​and their cumulative changes over consecutive dates;

[0054] The output includes an early warning data table containing date, monitoring point, effective leaf surface wet duration per day, dominant environmental factors, daily disease unit value, and risk level, which is used for Phytophthora blight control decision-making.

[0055] The beneficial effects of this invention are:

[0056] 1. More comprehensive assessment dimensions and stronger applicability: Based on commonly used factors such as air temperature and humidity and precipitation, this invention further integrates environmental factors that have a key impact on the germination and spread of Phytophthora spores and crop susceptibility, such as sunshine duration, photosynthetically active radiation, soil temperature, air pressure and dew point temperature. It also constructs a multi-environmental factor fusion system by combining trend change data, thereby improving the completeness of the quantitative assessment of Phytophthora risk and its adaptability to different meteorological processes.

[0057] 2. More accurate and stable determination of effective leaf surface wetting duration: This invention adopts a multi-factor coupled determination rule, including direct precipitation wetting, condensation-type wetting by combining temperature and dew point difference with soil temperature, and continuous wetting by the interaction of air pressure change, insufficient sunshine and high humidity, and introduces weak light calibration; at the same time, it suppresses hourly fluctuations and boundary jitters through evidence scoring mechanism and hysteresis state machine with entry threshold higher than exit threshold, making the determination of wetting start and end time more reliable and the statistics of continuous wetting across days more accurate.

[0058] 3. Enhanced automated batch processing and early warning output capabilities: This invention supports time alignment, missing data repair, and anomaly correction for multi-monitoring data on a continuous hourly scale, and automatically calculates the effective leaf wet duration and disease units per day; it characterizes the cumulative impact of a continuous suitable environment through a persistence index, and outputs risk levels and early warning data tables in combination with risk thresholds, thereby improving calculation efficiency, reducing human error, and enhancing the timeliness and accuracy of prevention and control decisions. Attached Figure Description

[0059] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0060] Figure 1 A flowchart illustrating a quantitative model for fungal diseases that integrates multiple environmental factors and a method for disease analysis;

[0061] Figure 2 This is a flowchart of step S3, multi-factor coupling determination, of the present invention.

[0062] Figure 3 This is a flowchart of the hysteresis state machine determination process in step S4 of the present invention. Detailed Implementation

[0063] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0064] refer to Figure 1 A quantitative model for fungal diseases and a method for analyzing diseases that integrates multiple environmental factors, including:

[0065] S1. Obtain multidimensional environmental data of the area to be analyzed on a continuous hourly scale, including air temperature, relative humidity, precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure and dew point temperature, perform time alignment, missing data repair and unit unification to obtain a standardized environmental sequence.

[0066] S2. Construct a multi-environmental factor fusion system based on standardized environmental sequences to obtain the environmental factor fusion vector corresponding to each hour;

[0067] S3. Determine the hourly humidity candidate value by performing multi-factor coupled humidity state determination based on the environmental factor fusion vector;

[0068] S4. Determine the hourly humidity status based on the hourly humidity candidate value, and obtain the effective leaf surface humidity duration on a daily basis.

[0069] S5. Input the effective leaf surface wet duration and corresponding environmental factors into the nonlinear Phytophthora quantification model to calculate the daily disease unit value;

[0070] S6. Generate disease risk analysis results and early warning results based on the daily disease unit value and its cumulative over time.

[0071] In this specific embodiment, S1 includes:

[0072] Environmental data is collected and standardized from monitoring points within the analysis area on an hourly basis, and the monitoring point numbers are denoted as follows. ,in and The total number of monitoring points is represented by the hourly time stamp. ,in and For continuous hours, for each monitoring point With each hour Create an hourly data record and include air temperature relative humidity Precipitation Sunshine duration Photosynthetically active radiation Soil temperature air pressure and dew point temperature ,in The hourly value is obtained by collecting data from the corresponding sensor at a fixed sampling period within one hour and taking the arithmetic mean. The hourly value is obtained by summing the cumulative rainfall of the tipping bucket rain gauge over the hour. The value is accumulated by the light sensor within an hour, and when used on an hourly scale, its range is limited to 0 to 1, representing the effective sunshine hours within that hour. The PAR sensor samples the photon flux density over an hour, first taking the hourly average and then converting it into an hourly integral to unify the values. dimension;

[0073] Data of the same type from different sources are classified by a unified time identifier. Perform matching and time alignment, with the time alignment rule set to use the hour as the starting boundary and to match elements falling within the specified range. Samples within the same interval are merged into the same hourly record, when the same monitoring point In the same hour When multiple sources of repeated observations occur, single-source values ​​are retained to ensure consistency of single values, following a fixed priority that prioritizes field sensor data over meteorological station data.

[0074] Missing segment repair is performed by branching based on the number of consecutive hours of missing segments, assuming the length of the missing segment is 1. and For the same monitoring point The number of consecutive missing hours of the same environmental element, when When valid observations exist at both ends of the missing segment, linear interpolation is used, and the interpolation model is defined as follows:

[0075] ;

[0076] in Indicates missing hours The hourly value of this element after repair. Indicates the same monitoring point The nearest valid hour before the start of the missing segment The hourly value of this element at that location. Indicates the same monitoring point The nearest valid hour after the end of the missing segment The hourly value of this element at that location. Indicates that it is located at and The missing hour sequence number between them and satisfy This indicates the number of hours in the interpolation span. Indicates the monitoring point number;

[0077] when If the missing segment is located at the sequence boundary, resulting in no valid observations at either end of the missing segment, interpolation is performed using adjacent monitoring points on the same day, with a fixed distance from that monitoring point. The three nearest monitoring points are used as supply points. The hourly values ​​of the supply points are summed using a weighted average of the squared distances to obtain the repaired value, with the weights calculated based on the distance from each supply point to the monitoring point. The inverse square of the geographical distance is normalized within 3 supply points, so that the spatial proximity of the same day and hour is used to repair and avoid introducing system bias across days;

[0078] Amplitude limiting correction is applied to anomalous mutation data, along with both range constraints and hourly rate of change constraints. The range constraints are set as follows: And the unit is And the unit is And the unit is And the unit is And the unit is And the unit is And the unit is And the unit is The hourly rate of change constraint is set as follows:

[0079] , ;

[0080] The units of the rate of change thresholds are consistent with the units of the corresponding elements and are used to suppress sensor jitter and spikes. When the observed value violates the value range constraint or the hourly rate of change constraint, the hourly value is replaced with the limit value that meets the constraint and the effective value of the adjacent previous hour is used as the reference benchmark to perform truncation, so as to ensure the continuity of the sequence and physical rationality.

[0081] Unify all elements to the preset dimensions and standardize units. The unit standardization rule is set to convert air pressure input in Pa to hPa. by Convert to input Convert the sunshine duration from s to h when input, and keep it... for for To form a dimensionlessly consistent hourly sequence in mm;

[0082] After completing the above-mentioned time alignment, missing data repair, abnormal amplitude limit correction, and unit unification, each monitoring point will be... In consecutive hours The 8-element hourly records are output as a standardized environmental sequence in chronological order.

[0083] In this specific embodiment, S2 includes:

[0084] Based on standardized environmental sequences, for each monitoring point With each hour Construct an environmental factor fusion vector and form a multi-environmental factor fusion system, in which For the total number of monitoring points, To determine the continuous hours, first set the air temperature relative humidity Precipitation Sunshine duration Photosynthetically active radiation Soil temperature air pressure and dew point temperature Mapped to values ​​located at Dimensionless environmental factors and Factor values ​​exceeding the effective range for Phytophthora spore germination, spread, or crop susceptibility are truncated. A unified linear saturation mapping model is used for truncation and normalization.

[0085] ;

[0086] in Representation elements At the monitoring point with hours Dimensionless environmental factors at the location Indicates the standardized hourly value participating in the mapping and , Representation elements The lower bound of the effective range, Representation elements The upper limit of the effective range, Represents the amplitude limiting function and when When taking 0, Take 1 for the first case, and 1 for the rest. And fix the effective range parameters of each element as follows: And the unit is And the unit is And the unit is mm. And the unit is And the unit is And the unit is , And the unit is hPa. And the unit is Therefore, when the hourly value of any element is below its lower bound... or above its upper limit At that time, the dimensionless environmental factor is truncated to the boundary value through clip to achieve truncation suppression of out-of-range inputs;

[0087] Subsequently, the changing trends over adjacent hours were calculated to characterize the dynamic evolution of the environmental state, and temperature changes were fixed. Defined as and Divide the difference by 8 and then limit the amplitude to Changes in humidity Defined as and Divide the difference by 30 and then limit the amplitude to Changes in air pressure Defined as and Divide the difference by 4 and then limit the amplitude to Changes in light Defined as and Divide the difference by 8 and then limit the amplitude to and in When and Set to 0 to eliminate uncertainty caused by the lack of preceding data in the first hour;

[0088] Finally, according to the same hour The time marker combines the dimensionless environmental factors and the trend components in a fixed order to form an environmental factor fusion vector. The fixed order is as follows:

[0089] ;

[0090] and all exist to This monitoring point is located within the area. A multi-environmental factor fusion system.

[0091] In this specific embodiment, S3 includes:

[0092] Based on environmental factor fusion vector For each monitoring point With each hour Perform multi-factor coupled wetting state determination and output hourly wetting candidate values. ,in For the total number of monitoring points, For consecutive hours, The range of values ​​is Furthermore, the larger the value, the stronger the evidence that the leaf surface was moist during that hour;

[0093] First, based on a standardized environmental sequence, three types of wetting evidence and one type of low-light calibration condition are generated hourly. Direct wetting evidence of precipitation is determined by a precipitation threshold, and when… Established at that time, among which For monitoring points In hours Rainfall and the unit is And the unit is mm;

[0094] Condensation-type wetting evidence is determined by coupling condensation conditions and when and Established at that time, among which This represents the difference between air temperature and dew point temperature, and the unit is _____. Indicates air temperature and the unit is Represents dew point temperature and the unit is . And the unit is , Represents soil temperature and the unit is And the unit is ;

[0095] Evidence of persistent humidity is determined by the combined presence of "changes in air pressure, insufficient sunshine duration, and high relative humidity," which serves as the criteria for determining persistence. and and And the above three conditions are within hours. When the same condition is met, the hour is determined. The evidence for persistent wetting is established, among which Relative humidity and the unit is And the unit is Sunshine duration in units of And the unit is The component representing the trend of air pressure change over adjacent hours and its value range is [value range missing]. Furthermore, it is dimensionless and corresponds to a change in air pressure of at least 1 hPa per hour;

[0096] Low light calibration conditions are determined based on the photosynthetically active radiation threshold and when Established at that time, among which Photosynthetically active radiation and its unit is And the unit is ;

[0097] Subsequently, an evidence scoring mechanism was used to integrate various types of wet evidence into a wet evidence score. The strength of the evidence was calibrated under low-light conditions, and the wetness evidence score was calculated according to the following model:

[0098] ;

[0099] in Indicates monitoring point In hours The wet evidence score and the range of values ​​are This indicates that the upper limit of the result within the parentheses is truncated to 1 to avoid score overflow caused by the accumulation of multiple pieces of evidence. Indicates the low-light calibration coefficient and when Time to take And when Time to take This indicates the base score for evidence of direct wetting by precipitation and takes... This represents the base score for evidence of condensation-type wetting and takes... Indicates the baseline score for evidence of persistent wetting and takes Indicates the amount of direct wetting evidence from precipitation and when Take 1 if it is true, otherwise take 1. Indicates the amount of evidence for condensation-type wetting and when and Take 1 if it is true, otherwise take 1. Indicates the amount of evidence for persistent wetting and when and and And hours The value is 1 if all three conditions are met; otherwise, the value is 0.

[0100] Finally, the moisture evidence will be scored. Convert to hourly humidity candidate value , among which when Fixed output The candidate value for wet is wet, when Fixed output The wet candidate value indicates non-wetting, when Fixed output The intermediate evidence values ​​are retained and then passed to the hysteresis state machine of S4 for stabilization determination, whereby... This indicates a wettability confirmation threshold and is dimensionless. This indicates a dimensionless threshold for excluding wetting, resulting in a sequence of hourly wettability candidate values ​​for subsequent statistics on effective leaf wettability duration.

[0101] In this specific embodiment, S4 includes:

[0102] hourly humidity candidate values Input a hysteresis state machine to determine the hourly humidity status and calculate the effective leaf surface humidity duration for each day. Indicates the monitoring point number and The total number of monitoring points. Indicates consecutive hour numbers and For consecutive hours, Furthermore, a higher value indicates stronger evidence of moisture, and the hysteresis state machine includes dry states. Moisture confirmation status Keep it moist Different thresholds are set for entering and exiting the wet state to suppress hourly jitter, with the threshold for entering the wet state set to... Furthermore, it is dimensionless, and the threshold for exiting the wet state is set to... And dimensionless and satisfying The number of consecutive hours required to enter confirmation is set to... And the unit is h, the required buffer time for exit is set to The unit is h, and the state machine initialization rule is set for each monitoring point. exist Always keep the state in a dry state. And will enter the counter With exit counter Set to 0, where This represents the number of hours in which the entry threshold is met consecutively, and the unit is h. This represents the number of hours in which the exit threshold has been met consecutively, and the unit is h.

[0103] When the state is dry At that time, if Then switch the status to wet confirmation status. juxtaposition And this hour will be recorded as the "confirmation cache hour" to await subsequent confirmation results. Then keep it dry and determine that hour as non-humid;

[0104] When the status is confirmed as moist At that time, if Then let ,when Switch the state to moisturizing state. and will include the most recent hour Each confirmed cache hour is uniformly determined as "wet" to ensure a stable start time for wetting and to prevent omissions due to single-hour fluctuations. If the confirmation fails, the status will be switched back to dry state. And clear the confirmation cache and set it. ,like Then maintain the moist state and keep it moist. The record remains unchanged, and this hour continues to be entered into the confirmation cache to await further strengthening or weakening of evidence in subsequent hours.

[0105] When the state is moistened At that time, if Then it is determined that the hour is wet and placed ,like The hour will still be considered wet and... ,when The status will be switched to dry mode starting from the next hour. juxtaposition This achieves a hysteresis mechanism that "exits only when the threshold is below the exit threshold and the buffer duration is reached." Furthermore, S3 improves the wetting evidence score and outputs a higher score under low-light conditions through a low-light calibration coefficient. Therefore, when weak light calibration makes Not less than Or exit the counter When reset, the moisture remains in the state. All hours that are extended and continue to be included in the effective leaf surface wetness duration;

[0106] After obtaining the hourly determination results, the hourly humidity state is defined. And if the hour is determined to be wet, take 1; otherwise, take 0, and adjust according to the calendar day. The effective daily leaf surface moisture duration is obtained by accumulating these amounts. ,in Indicates the calendar day index and starts with the hour. The date to which it belongs is used as the date to which that hour belongs, and a mapping function is used. Indicates hour sequence number The corresponding calendar day index, the statistical model for the effective leaf surface wetness duration per day is defined as follows: ;

[0107] in The unit is This represents the summation over the hours that satisfy the conditions. This indicates all dates assigned to a calendar day. Hour number Perform a traversal. Indicates monitoring point In hours The hourly humidity level, due to each The cumulative humidity corresponding to a full hour is summed to directly obtain the effective daily leaf surface humidity duration. Furthermore, when a continuous humidity process spans two dates, the duration is calculated using... The attribution rules will include the wet hours across days into their respective dates to ensure that the cross-day statistics are not distorted.

[0108] In this specific embodiment, S5 includes:

[0109] By calendar day To calculate particle size, the effective leaf surface wetting time per day is used. The standardized environmental sequence of S1 is summarized daily and input into the nonlinear disease quantification model, which outputs the daily disease unit value. ,in Indicates the monitoring point number and The total number of monitoring points. Indicates the calendar day index and To analyze the number of days, the environmental factors summarized on a daily basis include the daily average air temperature. Average daily soil temperature Daily average relative humidity Daily cumulative precipitation Sunshine duration Daily average photosynthetically active radiation Daily average air pressure and daily average dew point temperature ,in The air temperature over the course of 24 hours. Arithmetic mean and unit is The soil temperature over the 24 hours of that day. Arithmetic mean and unit is , The relative humidity over a 24-hour period of that day. Arithmetic mean and unit is The rainfall in 24 hours that day Summation in mm The duration of sunshine in 24 hours on that day Summation and unit is The photosynthetically active radiation of that day over 24 hours Arithmetic mean and unit is The air pressure over the 24 hours of that day Arithmetic mean and unit is The dew point temperature over the 24 hours of that day. Arithmetic mean and unit is ;

[0110] The nonlinear Phytophthora quantification model constructs temperature and humidity interaction terms on a daily scale. Light suppression item Suitable soil temperature Barometric Propagation With time accumulation And the range of values ​​for each item is set to Among them, the temperature and humidity interaction term The coupling effect of "high humidity + suitable temperature + continuous humidity" required for spore germination and infection is determined and used by temperature suitability, humidity suitability, dew point proximity, and humidification duration. Temperature suitability is determined by... exist The text describes a linear ascent and descent within a given area. or Set to zero and in Take 1 for the time, and adjust the humidity according to the appropriate level. exist Inner linear increase and in Set to zero and in Take 1 for time, and the closeness of the dew point is based on the dew point difference. exist Internal linear decrease and in When taking 1 and in Set the time to zero, and adjust the humidity duration factor according to... The ratio limit to 12 hours Thus in Take 1 at the time;

[0111] Light suppression term This study characterizes the promoting effect of low light intensity on reduced leaf evaporation and prolonged humidity, and is determined jointly by solar radiation and photosynthetically active radiation. The solar radiation factor is calculated as follows: The ratio limit to 8 hours And using "1 minus this limit value" as the contribution to low solar radiation, the radiation factor is calculated as follows: The ratio to 6 is limited to Using "1 minus the limit value" as the low radiation contribution and the light suppression term Take the arithmetic mean of the low solar radiation contribution and the low solar radiation contribution;

[0112] Suitable soil temperature for growth Used to characterize the supporting role of the soil environment in maintaining near-surface moisture and pathogen reproduction, and according to exist The linear rise and fall within the range and in or Set to zero and in Take 1 at the time;

[0113] Pressure propagation term This is used to characterize the promoting effects of low-pressure processes and pressure drops on pathogen transmission and water vapor maintenance, and is determined by the intraday pressure drop and the daily average pressure. The intraday pressure drop is denoted as... The unit is hPa, and it is calculated by accumulating all positive decreases hourly for that day, i.e., for each hour. And only include the difference in the summation if the difference is positive, then... The ratio to 6 is limited to As a contribution to the intensity of decline and The contribution of the decrease in intensity is multiplied by 0.7 to reflect the adverse effect of the high-pressure background on propagation;

[0114] Time accumulation item The cumulative disease unit value over 7 days is used to characterize recent risk and is derived from the cumulative disease unit value of the preceding days. for The calculations already made Summation and in Sum the available days, then... The ratio to 300 is limited to As ;

[0115] Dynamic weights are used to reflect the relative importance of different environmental contributions under different humidity backgrounds and are recalculated daily. A basic weight for the temperature-humidity interaction term is set. Basic weight of the light suppression term Soil temperature suitability as a basic factor Basic weight of air pressure propagation term The basic weight of the time accumulation item and set the total base weight Then divide each basic weight by Obtain the corresponding dynamic weights , and And the sum of all dynamic weights is 1;

[0116] Persistence Index As an adjustment term for the time accumulation period, and used to enhance or reduce the daily epidemic unit value, the persistence index... Based on continuous wet hours Number of consecutive effective leaf surface wetness periods between adjacent dates Number of consecutive increases in daily disease unit value and low light calibration duration hours The calculation yielded, where Defined as the hourly humidity level within that day. The maximum length of the continuous sequence of 1s, in units of Defined as follows: if the first hour of the day is wet and the preceding hour is also wet, take value 1; otherwise, take value 0. It is dimensionless. Defined as the number of consecutive days of growth up to the previous day and when At that time Add 1 to the base and truncate the upper limit to 3; otherwise, set it to 0 and dimensionless. Defined as simultaneously satisfying the following conditions within that day and The sum of hours, in hours, and the normalized quantity is calculated accordingly. , Then calculate and limit it to get ,in And the unit is ;

[0117] The daily disease unit value was finally calculated using the following nonlinear synthesis model. :

[0118]

[0119] ;

[0120] in Indicates monitoring point On calendar days The daily disease incidence rate is a dimensionless risk intensity measure. This represents a unit scaling factor and is used to map the aggregate contribution to a numerical range that facilitates classification. It represents a persistence index and is used to characterize the cumulative effect of continuous suitable environments on spore germination and dispersal, as well as crop susceptibility. , and These represent the dynamic weights of the temperature-humidity interaction term, the light inhibition term, the soil-temperature adaptability term, the air pressure propagation term, and the time accumulation term, respectively. , and Each represents an environmental contribution item corresponding to one of the dynamic weights, and each takes a value of . and the calculated Used for subsequent risk level determination and early warning generation.

[0121] In this specific embodiment, S6 includes:

[0122] Daily epidemic unit value The system uses the cumulative time results to classify risks, identify periods of rising risk, and output early warning results. Indicates the monitoring point number and The total number of monitoring points. Indicates the calendar day index and To analyze the number of days, Indicates monitoring point On calendar days The daily epidemic unit value is a dimensionless risk intensity quantity;

[0123] First, calculate the risk level for each day. It is further categorized into three levels: low risk, medium risk, and high risk. The set of values ​​is These correspond to low risk, medium risk, and high risk, respectively. The risk classification is determined by comparing fixed thresholds and following rules:

[0124] ;

[0125] in Indicates the risk level code. This represents the daily disease unit value. This represents the lower limit threshold for medium risk and is a dimensionless numerical threshold. This represents a high-risk lower limit threshold and is a dimensionless numerical threshold.

[0126] Subsequently based on consecutive dates The cumulative time-based results identify periods of increased risk, using the 7-day cumulative disease unit value already obtained in S5. As a discriminant and Indicates monitoring point On calendar days The cumulative daily case counts for the most recent 7 days were used to characterize the short-term persistent risk background. The identification of periods of rising risk was determined by simultaneously meeting either of two conditions: the first condition being a risk level jump and... The calendar day Marked as a day of increased risk, the second type of condition is a continuous increase in value and when and The calendar day Marked as a day of increased risk, where the differential threshold of 15 is a dimensionless numerical threshold used to suppress false alarms caused by small fluctuations, and when The risk level for that day was... Upgrade to level one on the basis and The process should be maintained without further improvement to reflect the cumulative effect of a continuously suitable environment;

[0127] When generating early warning data, the dominant environmental factors are simultaneously provided to support prevention and control decisions. The determination of the dominant environmental factors uses the environmental contribution items calculated internally by S5 and their dynamic weights, and is based on the data on the current day. The component that contributes the most is taken as the dominant term. The dominant term is uniquely determined among the temperature and humidity interaction term, light inhibition term, soil temperature suitability term, air pressure propagation term and time accumulation term, and its corresponding term name is output, thereby pointing the warning cause to the specific environmental driver.

[0128] Finally, the early warning data table is output, and the monitoring points are used for analysis. With calendar days The data is written row by row using the primary key, and the early warning data table fields are fixed to include the date. Monitoring points Effective daily leaf surface moisture duration The unit is h, the name of the dominant environmental factor, and the daily disease unit value. Risk level The risk level is marked as high, and the warning data table is saved in comma-separated text format and synchronously output to the warning interface for decision-making on the prevention and control of Phytophthora blight.

[0129] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0130] This invention standardizes and fuses hourly multidimensional environmental data to form an environmental factor fusion vector that simultaneously characterizes the current environmental state and its changing trends. Based on this vector, a hierarchical coupling judgment framework is employed to determine leaf surface moisture hourly and statistically analyze it daily, considering mechanisms such as direct precipitation wetting, condensation-type wetting due to temperature and dew point difference combined with soil temperature, and continuous wetting due to the interaction of air pressure changes, insufficient sunshine, and high humidity. This yields the effective daily leaf surface moisture duration. Subsequently, this moisture duration, along with other environmental factors such as temperature, humidity, precipitation, sunshine, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature, is input into a nonlinear blight quantification model. The model outputs disease units and performs risk classification and early warning, thereby covering more key environmental influencing factors in the assessment dimensions, improving the accuracy of start-end identification and cross-day continuous processes in moisture statistics, and achieving automated batch processing of hourly multidimensional data, reducing manual statistical errors and improving the timeliness and reliability of early warnings.

[0131] To address the problems in existing technologies, such as the susceptibility of humidity determination to single threshold fluctuations, boundary jitter caused by hourly data noise, and insufficient characterization of the cumulative effects of continuous suitable environments, this invention makes targeted improvements to the algorithm structure: First, it introduces an evidence scoring mechanism to weightedly fuse multiple types of evidence corresponding to different humidity formation mechanisms, and calibrates the evidence strength in conjunction with low light conditions, transforming humidity determination from a single-condition trigger to a consensus confirmation of multiple evidences, thus improving adaptability to complex weather processes; Second, it introduces a hysteretic state machine with an entry threshold higher than the exit threshold and a buffer period, transforming candidate humidity evidence into a stable humidity state sequence, suppressing frequent switching caused by hourly fluctuations, thereby improving the statistical accuracy of effective leaf humidity duration; Third, it introduces a persistence index to adjust the time accumulation term, characterizing the cumulative effect of continuous humidity and continuous risk growth, making the output of disease units more consistent with the continuous development of Phytophthora blight, further improving the accuracy of risk assessment and early warning results.

Claims

1. A quantitative model for fungal diseases integrating multiple environmental factors and a method for disease analysis, characterized in that, include: S1. Obtain multidimensional environmental data of the area to be analyzed on a continuous hourly scale, including air temperature, relative humidity, precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure and dew point temperature, perform time alignment, missing data repair and unit unification to obtain a standardized environmental sequence. S2. Construct a multi-environmental factor fusion system based on standardized environmental sequences to obtain the environmental factor fusion vector corresponding to each hour; S3. Determine the hourly humidity candidate value by performing multi-factor coupled humidity state determination based on the environmental factor fusion vector; S4. Determine the hourly humidity status based on the hourly humidity candidate value, and obtain the effective leaf surface humidity duration on a daily basis. S5. Input the effective leaf surface wet duration and corresponding environmental factors into the nonlinear Phytophthora quantification model to calculate the daily disease unit value; S6. Generate disease risk analysis results and early warning results based on the daily disease unit value and its cumulative over time.

2. The method for quantitative analysis of fungi and diseases that integrates multiple environmental factors according to claim 1, characterized in that, S1 includes: Collect or read environmental data from one or more monitoring points within the area to be analyzed on an hourly basis; Data from different sources are matched according to a unified time identifier to form hourly data records; For short-term missing data, interpolation is performed using adjacent hours or adjacent monitoring points on the same day; amplitude-limiting correction is applied to abnormal and sudden data. By unifying precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature into preset dimensions, a standardized environmental sequence is obtained.

3. The method for quantitative analysis of fungi and diseases that integrates multiple environmental factors according to claim 1, characterized in that, S2 include: Air temperature, relative humidity, precipitation, sunshine duration, photosynthetically active radiation, soil temperature, air pressure, and dew point temperature are respectively mapped to dimensionless environmental factors; The dimensionless environmental factors are combined according to the same hour time identifier to form an environmental factor fusion vector; Factor values ​​that exceed the effective range for Phytophthora spore germination, spread, or crop susceptibility are truncated or weighted down. The system calculates the temperature, humidity, air pressure, and light intensity changes in adjacent hours to form a multi-environmental factor fusion system that includes the current environmental state and its changing trends.

4. The method for quantitative analysis of fungi and diseases that integrates multiple environmental factors according to claim 1, characterized in that, S3 include: When the precipitation reaches the preset precipitation threshold, the corresponding hour will be marked as evidence of direct wetting by precipitation. When the difference between air temperature and dew point temperature is less than the condensation threshold and the soil temperature meets the conditions for promoting condensation, the corresponding hour is marked as evidence of condensation-type wetting. When changes in air pressure, insufficient sunshine duration, and relative humidity exceeding the humidity threshold together meet the conditions for continuous humidity, the corresponding hour is marked as evidence of continuous humidity. When photosynthetically active radiation is below the weak light threshold, the intensity of wet evidence for the same hour is calibrated for weak light. Hourly humidity candidate values ​​are generated based on the combined results of various humidity evidence.

5. The method for quantitative analysis of fungi and diseases by integrating multiple environmental factors according to claim 1, characterized in that, S4 include: The hourly wetness candidate values ​​are input into the hysteresis state machine, and the hourly wetness state is determined based on the threshold for entering the wetness state and the threshold for exiting the wetness state. The hours identified as being in a moist state are accumulated according to the calendar day to obtain the effective leaf surface moist duration for a single day; When a continuous wet period spans two dates, the wet hours spanning those dates will be counted into the corresponding dates according to their actual time. When low light calibration extends the wet state, the extended hours are included in the effective leaf wet duration for the corresponding date.

6. The method for quantitative analysis of fungi and diseases by integrating multiple environmental factors according to claim 1, characterized in that, S5 include: The effective leaf surface wet duration, average daily air temperature, average daily soil temperature, average daily relative humidity, daily cumulative precipitation, sunshine duration, average daily photosynthetically active radiation, average daily air pressure, and average daily dew point temperature are input into the nonlinear Phytophthora quantification model. The environmental contribution was calculated using the temperature and humidity interaction term, the light inhibition term, the soil temperature adaptation term, the air pressure propagation term, and the time accumulation term, respectively. The daily disease unit value is synthesized based on the dynamic weights of each environmental contribution.

7. The method for quantitative analysis of fungi and diseases by integrating multiple environmental factors according to claim 1, characterized in that, S6 include: The daily disease unit value is compared with the preset risk level threshold to generate low-risk, medium-risk and high-risk analysis results; Identify periods of increased risk based on daily disease unit values ​​and their cumulative changes over consecutive dates; The output includes an early warning data table containing date, monitoring point, effective leaf surface wet duration per day, dominant environmental factors, daily disease unit value, and risk level, which is used for Phytophthora blight control decision-making.

8. The method for quantitative analysis of fungi and diseases that integrates multiple environmental factors according to claim 4, characterized in that, The multi-factor coupled wetting state determination adopts an evidence scoring mechanism, including: Base scores were set for evidence of direct precipitation wetting, evidence of condensation wetting, evidence of persistent wetting, and evidence of low light calibration. The wet evidence score is obtained by weighting and summing the combinations of various wet evidence within the same hour. When the wet evidence score reaches the wet confirmation threshold, a wet candidate value is output. When the wet evidence score is lower than the wet exclusion threshold, a non-wet candidate value is output. When the wet evidence score is between the wet confirmation threshold and the wet exclusion threshold, an intermediate evidence value is retained.

9. The method for quantitative analysis of fungi incorporating multiple environmental factors according to claim 5, characterized in that, The hysteresis state machine includes a dry state, a wetness confirmation state, and a wetness maintenance state; When the wet candidate value reaches the threshold for entering the wet state for one or more consecutive hours, the state switches from dry state to wet confirmation state. When the wet candidate value continues to meet the maintenance condition, the state switches from wet confirmation state to wet maintenance state. When the wetness candidate value is lower than the exit wetness state threshold and continues to reach the preset buffer time, the state is switched from wetness-maintaining state to dry state. The threshold for entering the wetness state is higher than the threshold for exiting the wetness state.

10. The method for quantitative analysis of fungi and diseases by integrating multiple environmental factors according to claim 6, characterized in that, The nonlinear Phytophthora quantification model also includes a persistence index calculation process, which is calculated based on the number of consecutive wet hours, the number of times the effective leaf surface wetness continues on adjacent dates, the number of times the daily disease unit value increases continuously, and the number of hours of weak light calibration. The persistence index is used as an adjustment for the time-cumulative term to enhance or reduce the daily disease unit value, thereby characterizing the cumulative effect of a sustained suitable environment on Phytophthora spore germination, spread, and crop susceptibility.

Images