An intelligent monitoring and early warning system and method for large-scale farms based on deep learning

Abstract

The application relates to the technical field of early warning data processing, and discloses a large-scale breeding farm intelligent monitoring and early warning system and method based on deep learning, which automatically matches the environment risk level by collecting and analyzing environment parameters in real time, and carries out corresponding disinfection treatment according to the level. Compared with traditional manual monitoring and experience judgment, the intelligent monitoring and treatment mode can more accurately and timely identify and respond to the environment risk of the breeding farm, improves the pertinence and effectiveness of disinfection treatment, and through a dynamic regulation module, the system can dynamically adjust the disinfection treatment scheme according to the evaluation result of the disinfection treatment process parameters, further improves the intelligent level and management efficiency of the system, and provides a comprehensive and intelligent environment monitoring and disinfection management solution for the large-scale breeding farm, and provides strong support for the sustainable development of the breeding industry.

Description

Technical Field

[0001] This invention relates to the field of early warning data processing technology, specifically to a large-scale intelligent monitoring and early warning system and method for livestock farms based on deep learning. Background Technology

[0002] With the rapid development of information technology, all industries are undergoing profound digital transformation. As an important component of the national economy, the livestock industry is also gradually developing towards large-scale, intensive, and intelligent operations. Large-scale farms, through centralized management, standardized production, and optimized resource allocation, have significantly improved production efficiency and economic benefits, becoming the mainstream model in the modern livestock industry.

[0003] For example, the invention patent with publication number CN119599435A discloses a biosafety management system for aquaculture farms, which relates to the field of biological aquaculture and solves the problem of poor management effect in existing biosafety management systems for aquaculture farms. The system includes an environmental data module for acquiring regional division data, temperature monitoring coefficients, and humidity monitoring coefficients for each aquaculture sub-site, thus obtaining aquaculture farm environmental monitoring data; a biological data module for monitoring biological diseases and pests in each aquaculture sub-site and acquiring the corresponding site disease and pest monitoring coefficients, thus obtaining aquaculture farm biological monitoring data; and an aquaculture risk module for analyzing the aquaculture farm environmental monitoring data and aquaculture farm biological monitoring data to obtain the aquaculture risk assessment coefficient for each aquaculture sub-site, and providing aquaculture risk warnings for each aquaculture sub-site based on the aquaculture risk assessment coefficients.

[0004] For example, invention patent CN117787736A discloses a method and system for constructing a healthy aquaculture safety system, belonging to the field of healthy aquaculture technology. The method includes: obtaining the scale and type of the target farm; determining the corresponding aquaculture standards based on the scale and type of the target farm, and obtaining multiple aquaculture risk factors according to the aquaculture standards; constructing an aquaculture risk model, inputting the multiple aquaculture risk factors into the aquaculture risk model, and obtaining the output results; constructing a monitoring system and early warning mechanism based on the output results to monitor the target farm; and constructing a healthy aquaculture safety system based on the monitoring results to achieve health and safety supervision of the target farm.

[0005] However, in the process of implementing the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems: the existing intelligent farm monitoring and early warning system has significant deficiencies in data integration, dynamic mapping, intelligent analysis and full-chain traceability, which makes it difficult to achieve efficient and accurate monitoring and early warning when dealing with complex farming environments and biosecurity risks. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a deep learning-based intelligent monitoring and early warning system and method for large-scale livestock farms, which can effectively solve the problems mentioned in the background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: The first aspect of the present invention provides a deep learning-based intelligent monitoring and early warning system for large-scale livestock farms, comprising: an environmental analysis module, used to mark large-scale livestock farms as target farms, collect and analyze environmental parameters of the target farms to match the environmental risk level of the target farms, and disinfect the target farms based on the environmental risk level; a dynamic control module, used to collect and evaluate the disinfection process parameters of the target farms, determine whether to update the environmental risk level of the target farms, and thus optimize the disinfection process of the target farms; and a safety early warning module, used to monitor and intelligently analyze the result parameters of the target farms through deep learning, and thus determine whether to issue a disinfection early warning for the target farms.

[0008] As a further step, the environmental risk level of the target farm is matched. The specific matching process is as follows:

[0009] Environmental parameters of the target farm are collected and analyzed to obtain its environmental risk index. If the environmental risk index falls within the low-risk range, the farm is marked as low-risk and disinfected using the first disinfection measure. If the index falls within the medium-risk range, the farm is marked as medium-risk and disinfected using the second disinfection measure. If the index falls within the low-risk range, the farm is marked as high-risk and disinfected using the third disinfection measure, while simultaneously issuing an environmental risk warning.

[0010] As a further solution, the disinfection process of the target farm is optimized. The specific optimization process is as follows: obtain the adjustment compliance deviation value of the target farm; if the environmental risk level of the target farm is low risk, update the environmental risk level of the target farm to medium risk, and disinfect the target farm based on the second disinfection measure. At the same time, compare the adjustment compliance deviation value of the target farm with the adjustment compliance deviation threshold. If the adjustment compliance deviation value of the target farm is within the adjustment compliance deviation threshold, the second disinfection measure is not updated. If the adjustment compliance deviation value of the target farm is less than or equal to the adjustment compliance deviation threshold, the second disinfection measure is updated. The specific update process is as follows: match the reduction coefficient from the early warning database based on the adjustment compliance deviation value of the target farm, and then update the second disinfection measure.

[0011] As a further measure, a disinfection warning is issued for the target farm. The specific warning process is as follows: Based on the environmental warning coefficient of the target farm, if the environmental warning coefficient is less than the defined environmental warning coefficient, no disinfection warning is issued for the target farm, and routine disinfection measures are not adjusted. If the environmental warning coefficient is greater than the defined environmental warning coefficient, a disinfection warning is issued for the target farm, and a disinfection warning plan for the target farm is matched from the warning database. At the same time, the reduction rate of the environmental warning coefficient of the target farm in the first cycle is detected and compared with the reduction rate of the defined environmental warning coefficient. If the reduction rate of the environmental warning coefficient of the target farm in the first cycle is greater than the reduction rate of the defined environmental warning coefficient, the disinfection warning for the target farm is lifted. If the reduction rate of the environmental warning coefficient of the target farm in the first cycle is less than or equal to the reduction rate of the defined environmental warning coefficient, the disinfection warning feedback for the target farm continues.

[0012] The second aspect of this invention provides a deep learning-based intelligent monitoring and early warning method for large-scale livestock farms, comprising: Step 1, marking the large-scale livestock farm as a target farm, collecting and analyzing environmental parameters of the target farm to match its environmental risk level, and disinfecting the target farm based on its environmental risk level; Step 2, collecting and evaluating parameters of the disinfection process of the target farm, determining whether to update the environmental risk level of the target farm, thereby optimizing the disinfection process of the target farm; Step 3, monitoring and intelligently analyzing the result parameters of the target farm using deep learning to determine whether to issue a disinfection early warning for the target farm.

[0013] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0014] (1) This invention provides a deep learning-based intelligent monitoring and early warning system and method for large-scale farms. By collecting and analyzing environmental parameters in real time, it automatically matches the environmental risk level and performs corresponding disinfection treatment according to the level. Compared with traditional manual monitoring and experience judgment, this intelligent monitoring and treatment method can more accurately and timely identify and respond to the environmental risks of farms, and improve the pertinence and effectiveness of disinfection treatment. Through the dynamic control module, the system can dynamically adjust the disinfection treatment plan according to the evaluation results of the disinfection treatment process parameters, further improving the intelligence level and management efficiency of the system. This invention provides a comprehensive and intelligent environmental monitoring and disinfection management solution for large-scale farms, and provides strong support for the sustainable development of the aquaculture industry.

[0015] (2) The present invention is based on a dynamic control module of disinfection efficacy comprehensive evaluation index and adjustment compliance index. By evaluating process parameters such as disinfectant residual concentration and pathogen load difference in real time, and combining the secondary coupling of environmental risk index, the system can adaptively update disinfection measures (such as matching and reducing coefficients to optimize disinfection intensity), realize the whole closed-loop management from "monitoring-evaluation-optimization", and make up for the shortcomings of weak dynamic response capability of traditional systems.

[0016] (3) This invention constructs a dynamic environmental resistance index and combines it with the time-series prediction of the environmental early warning coefficient to realize the graded triggering of disinfection early warning in future cycles (such as the first and second cycles). For example, when the predicted future environmental early warning coefficient exceeds the threshold, the system will start the pre-disinfection early warning in advance and dynamically adjust the disinfection plan, effectively blocking the disease transmission chain and solving the core problem of the passive and lagging traditional early warning mechanism. Attached Figure Description

[0017] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the system module connections of the present invention.

[0019] Figure 2 This is a schematic diagram of the method steps of the present invention.

[0020] Figure 3 This is a first schematic diagram showing the detailed process of the method steps of the present invention.

[0021] Figure 4 This is a second schematic diagram showing the detailed process of the method steps of the present invention.

[0022] Figure 5 The third diagram shows the detailed process flow of the methods and steps in this section. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0024] Reference Figure 1 As shown, the first aspect of the present invention provides a large-scale intelligent monitoring and early warning system for livestock farms based on deep learning, including: an environmental analysis module, a dynamic control module, a safety early warning module, and an early warning database.

[0025] The early warning database is used to store parameters of a large-scale intelligent monitoring and early warning system for livestock farms based on deep learning.

[0026] The environmental analysis module is connected to the dynamic control module, which in turn is connected to the safety early warning module. All three modules—environmental analysis, dynamic control, and safety early warning—are connected to the early warning database.

[0027] The environmental analysis module is used to mark large-scale farms as target farms, collect and analyze the environmental parameters of the target farms, thereby matching the environmental risk level of the target farms, and disinfecting the target farms based on the environmental risk level.

[0028] Specifically, the environmental risk level of the target farm is matched. The matching process is as follows: environmental parameters of the target farm are collected and analyzed to obtain the environmental risk index of the target farm. If the environmental risk index of the target farm belongs to the low-risk index range, the environmental risk level of the target farm is marked as low-risk, and the first disinfection measure is used to disinfect the target farm; if the environmental risk index of the target farm belongs to the medium-risk index range, the environmental risk level of the target farm is marked as medium-risk, and the second disinfection measure is used to disinfect the target farm; if the environmental risk index of the target farm belongs to the high-risk index range, the environmental risk level of the target farm is marked as high-risk, and the third disinfection measure is used to disinfect the target farm, while an environmental risk warning is issued for the target farm.

[0029] The aforementioned environmental risk warning for the target farm specifically refers to the system displaying a visual pop-up window to the administrator, indicating that the target farm's environment poses a high risk.

[0030] It should be explained that different risk levels correspond to different disinfection measures. Specifically, the first disinfection measure is to maintain the routine disinfection plan and ventilation management (spraying disinfectant daily with standard disinfectant according to the established disinfection frequency, and operating the ventilation system at the established power). The second disinfection measure is to switch to a more effective disinfection plan, strengthen ventilation management, and increase the frequency of water source replacement in the target farm (spraying disinfectant daily with broad-spectrum disinfectant according to the established disinfection frequency, increasing the ventilation power in higher-risk areas, and increasing the frequency of water source replacement in the target farm). The third disinfection measure is to implement an emergency disinfection plan, forcibly turn on full-power ventilation in the entire area, replace the water source with fresh water, and turn on the purification device (disinfecting with ozone or high-pressure pulse in the entire closed area, operating all ventilation systems in the target farm at full power, turning on the dynamic fresh water system, and turning on the purification device to purify the air in the target farm).

[0031] The aforementioned higher-risk areas are identified by comparing the pathogen load with the pathogen load threshold defined in the early warning database. If the pathogen load is greater than the defined pathogen load threshold, it is marked as a higher-risk area.

[0032] It should be explained that increasing the frequency of water source changes at the target farm involves querying the corresponding water source change frequency from the early warning database and updating it, thereby increasing the frequency of water source changes at the target farm.

[0033] Environmental parameters of the target farm were collected and analyzed using a multimodal sensor network. The environmental risk index of the target farm was also collected. The specific analysis process included: the pathogen load, the temperature-humidity coupling index, and the rate of environmental risk decay. The pathogen load refers to the distribution density of pathogenic microorganisms in the target farm; the temperature-humidity coupling index is a comprehensive indicator of the coupling between temperature and humidity in the target farm; and the rate of environmental risk decay refers to the rate at which the target farm's ability to inhibit pathogenic microorganisms weakens. The pathogen load and the rate of environmental risk decay can be monitored using pathogen concentration monitoring tools (by detecting the fluorescence signal of pathogenic microorganisms using an online pathogen concentration monitor, and monitoring the pathogen load based on the fluorescence signal).

[0034] It should be explained that the rate of environmental risk decay is obtained by real-time monitoring of pathogen concentration and immediate response to the rate of change of pathogen concentration decline. The aforementioned rate of change of pathogen concentration decline is obtained by monitoring the pathogen concentration in real time using an online pathogen concentration monitoring instrument and performing second-order differentiation.

[0035] It should be explained that by introducing influence coefficients to quantify the degree of influence of the target farm's ambient temperature on the temperature-humidity coupling index and the degree of influence of the target farm's ambient humidity on the temperature-humidity coupling index, the various degrees of influence are coupled to analyze the temperature-humidity coupling index. The specific expression is as follows:

[0036]

[0037] In the formula, THI is the temperature-humidity coupling index of the target farm, T is the ambient temperature of the target farm, RH is the ambient humidity of the target farm, ΔT is the preset reference ambient temperature in the early warning database, ΔRH is the preset reference ambient temperature in the early warning database, kx1 is the influence coefficient corresponding to the preset ambient temperature in the early warning database, and kx2 is the influence coefficient corresponding to the preset ambient humidity in the early warning database.

[0038] By introducing influence coefficients, the influence of the pathogen load component, the temperature and humidity coupling index component, and the environmental risk decay rate component of the target farm on the environmental risk index are quantified. These influences are then coupled to analyze the environmental risk index of the target farm. The specific expression is as follows:

[0039] ERI=wi1×FL_C+wi2×FL_THI+wi3×FL_R;

[0040]

[0041]

[0042] In the formula, ERI is the environmental risk index of the target farm, FL_C is the pathogen load component of the target farm, FL_THI is the temperature-humidity coupling index component of the target farm, FL_R is the environmental risk decay rate component of the target farm, C is the pathogen load of the target farm, THI is the temperature-humidity coupling index of the target farm, J_C is the preset defined pathogen load in the early warning database, J_THI is the preset defined temperature-humidity coupling index in the early warning database, R is the environmental risk decay rate of the target farm, J_R is the preset defined environmental risk decay rate in the early warning database, wi1 is the influence coefficient corresponding to the preset pathogen load component in the early warning database, wi2 is the influence coefficient corresponding to the preset temperature-humidity coupling index component in the early warning database, and wi3 is the influence coefficient corresponding to the preset environmental risk decay rate component in the early warning database.

[0043] The environmental risk index of a livestock farm is used to characterize the degree of environmental risk of a target livestock farm.

[0044] The pathogen load component of the aforementioned target farm refers to the ratio of pathogen load to the defined pathogen load (i.e., the ratio); the temperature and humidity coupling index component of the aforementioned target farm refers to the ratio of the temperature and humidity coupling index to the defined temperature and humidity coupling index (i.e., the ratio); the environmental risk decay rate component of the aforementioned target farm refers to the ratio of the environmental risk decay rate to the defined environmental risk decay rate (i.e., the ratio); the defined pathogen load represents the maximum allowable value of pathogen load; the defined temperature and humidity coupling index represents the maximum allowable value of the temperature and humidity coupling index; and the defined environmental risk decay rate represents the minimum allowable value of the environmental risk decay rate.

[0045] The influence coefficients corresponding to the pathogen load component represent the degree of influence of the pathogen load component on the environmental risk index; the influence coefficients corresponding to the temperature and humidity coupling index component represent the degree of influence of the temperature and humidity coupling index component on the environmental risk index; and the influence coefficients corresponding to the environmental risk decay rate component represent the degree of influence of the environmental risk decay rate on the environmental risk index. The early warning database stores the correspondence between the pathogen load component, the temperature and humidity coupling index component, and the environmental risk decay rate component and their corresponding influence coefficients. For example, by inputting the pathogen load component, the temperature and humidity coupling index component, and the environmental risk decay rate component into the early warning database, the database can match the influence coefficients corresponding to the pathogen load component, the temperature and humidity coupling index component, and the environmental risk decay rate component, all of which have values ​​between 0 and 1.

[0046] It should be explained that the rate of environmental risk decay reflects the rate at which the environment of the target farm inhibits pathogenic microorganisms. A high rate of environmental risk decay means that the rate of increase in pathogenic microorganisms is faster, and the pathogen load increases accordingly. That is, the environmental risk decay rate component increases, and the pathogen load component increases. On the other hand, the pathogen load component reflects the proportion of pathogenic microorganism distribution density in the target farm. When this proportion is high, it means that the distribution density of pathogenic microorganisms has increased significantly, and the environmental risk index has increased accordingly. This causes the temperature and humidity coupling index to deviate from the preset temperature and humidity coupling index. That is, the temperature and humidity coupling index component of the target farm increases, which in turn leads to a further increase in the environmental risk index. In summary, the environmental risk decay rate component indirectly affects the environmental risk index by influencing the pathogen load component and the temperature and humidity coupling index component.

[0047] The dynamic control module is used to collect and evaluate the disinfection process parameters of the target farm, determine whether to update the environmental risk level of the target farm, and thus optimize the disinfection process of the target farm.

[0048] In one specific embodiment, the present invention dynamically regulates based on a comprehensive disinfection efficacy evaluation index and a compliance adjustment index. By real-time evaluation of process parameters such as disinfectant residual concentration and pathogen load difference, combined with the secondary coupling of the environmental risk index, the system can adaptively update disinfection measures (such as matching and reducing coefficients to optimize disinfection intensity), realizing a closed-loop management from "monitoring-evaluation-optimization", thus making up for the shortcomings of traditional systems in terms of weak dynamic response capabilities.

[0049] Specifically, the process for determining whether to update the environmental risk level of the target farm is as follows: collect and obtain the disinfection process parameters of the target farm, obtain the regulation compliance index of the target farm, and compare it with the regulation compliance threshold. If the regulation compliance index is greater than or equal to the regulation compliance index threshold, then it is determined that the environmental risk level of the target farm will not be updated.

[0050] If the adjustment compliance index is less than the adjustment compliance index threshold, the environmental risk level of the target farm will be updated, thereby optimizing the disinfection process of the target farm.

[0051] It should be explained that adjusting the compliance threshold refers to adjusting the compliance index threshold, which is the minimum value of the compliance index that is preset in the early warning database.

[0052] Specifically, the compliance index for the target farm is assessed through the following process: Disinfection process parameters for the target farm include the disinfection success rate, the residual disinfectant concentration, and the difference in pathogen load. The comprehensive evaluation index of disinfection efficiency is obtained by multiplying and coupling the disinfection success rate, residual disinfectant concentration, and pathogen load difference with their corresponding influence coefficients. The disinfection success rate refers to the pathogen inactivation rate per unit time at the target farm. The residual disinfectant concentration refers to the concentration of disinfectant remaining in the environment after disinfection. The pathogen load difference refers to the difference in pathogen distribution density before and after disinfection. The disinfection success rate and residual disinfectant concentration can be monitored using disinfection-related monitoring tools (the disinfection success rate is obtained by monitoring the reduction in adenosine triphosphate (ATP) content using a pathogenic microorganism ATP monitoring tool; the disinfection success rate is directly proportional to the reduction in ATP content; the residual disinfectant concentration is obtained through real-time monitoring using a disinfectant concentration monitoring tool).

[0053] The aforementioned disinfection success rate is used to measure the ability of disinfection measures to inhibit and eliminate pathogens in a target farm within a specific time period. It represents the percentage reduction in the number of pathogens per unit time (e.g., per hour, per day), and is an important parameter for evaluating disinfection effectiveness, directly reflecting the efficiency and effectiveness of disinfection measures in reducing or eliminating pathogenic microorganisms.

[0054] The above-mentioned pathogen load difference is obtained by subtracting the pathogen load after disinfection from the pathogen load before disinfection in the target farm.

[0055] It should be explained that the comprehensive evaluation index of disinfection efficiency for the target farm is as follows:

[0056]

[0057] In the formula, DEI is the comprehensive evaluation index of disinfection efficiency of the target farm, G is the disinfection success rate of the target farm, RC is the disinfectant residue concentration of the target farm, PC is the pathogen load difference of the target farm, J_G is the preset definition of disinfection success rate in the early warning database, J_RC is the preset definition of disinfectant residue concentration in the early warning database, J_PC is the preset definition of pathogen load difference in the early warning database, gf1 is the influence coefficient corresponding to the preset disinfection success rate in the early warning database, gf2 is the influence coefficient corresponding to the preset disinfectant residue concentration in the early warning database, and gf3 is the influence coefficient corresponding to the preset pathogen load difference in the early warning database.

[0058] The disinfection process parameters of the target farm were collected and evaluated. The specific evaluation process included the disinfection success rate, the residual concentration of disinfectant, and the pathogen load deviation value of the target farm. The collected data were preprocessed using an improved LSTM-Attention mechanism.

[0059] The above-described definition of disinfection success rate represents the minimum allowable value for disinfection success rate; the above-described definition of disinfectant residue concentration represents the maximum allowable value for disinfectant residue concentration; the above-described definition of pathogen load difference represents the minimum allowable value for pathogen load difference; the above-described influence coefficient corresponding to the disinfection success rate represents the degree of influence of the disinfection success rate on the comprehensive evaluation index of disinfection efficacy; the above-described influence coefficient corresponding to the disinfectant residue concentration represents the degree of influence of the disinfectant residue concentration on the comprehensive evaluation index of disinfection efficacy; the above-described influence coefficient corresponding to the pathogen load difference represents the degree of influence of the pathogen load difference on the comprehensive evaluation index of disinfection efficacy.

[0060] By introducing influence coefficients, the impact of the target farm's successful disinfection rate on the comprehensive disinfection efficacy evaluation index, the impact of the target farm's disinfectant residue concentration on the comprehensive disinfection efficacy evaluation index, and the impact of the target farm's environmental risk decay rate component on the comprehensive disinfection efficacy evaluation index are quantified. These influence factors are then coupled to assess the comprehensive disinfection efficacy evaluation index of the target farm. Simultaneously, the environmental risk index of the target farm is introduced for secondary coupling to derive the target farm's regulatory compliance index, specifically expressed as follows:

[0061]

[0062] In the formula, ACI is the regulation compliance index of the target farm, ERI is the environmental risk index of the target farm, DEI is the comprehensive evaluation index of disinfection efficiency of the target farm, J_ERI is the pre-defined environmental risk index in the early warning database, J_DEI is the pre-defined comprehensive evaluation index of disinfection efficiency in the early warning database, cd1 is the influence coefficient corresponding to the pre-defined environmental risk index in the early warning database, and cd2 is the influence coefficient corresponding to the pre-defined comprehensive evaluation index of disinfection efficiency in the early warning database.

[0063] It should be explained that the comprehensive evaluation index of disinfection efficiency of the target farm is used to characterize the overall effect of disinfection measures.

[0064] It should be explained that the regulation compliance index of the target farm is used to characterize the degree of compliance of the regulation operation, thereby determining whether to update the environmental risk level of the target farm.

[0065] The impact coefficients corresponding to the aforementioned environmental risk indices indicate the degree of influence of the environmental risk indices on the regulatory compliance index; the impact coefficients corresponding to the aforementioned comprehensive disinfection efficacy assessment index indicate the degree of influence of the comprehensive disinfection efficacy assessment index on the regulatory compliance index.

[0066] It's important to explain that a higher compliance index means that current disinfection measures are more effective at reducing the environmental risk level of the target farm, thus more effectively regulating it. However, as environmental risk increases, the compliance index may face challenges. This is because rising environmental risk and pathogen load can cause the environmental risk index to gradually approach or exceed the high-risk level range. Simultaneously, the pathogen load difference also increases, indirectly affecting the compliance index. The disinfection efficacy assessment index measures the effectiveness of disinfection measures against pathogens. When the disinfection success rate increases (i.e., the pathogen inactivation rate at the target farm increases), the disinfection effect is enhanced, the disinfection efficacy assessment index increases, and the compliance index rises. Conversely, when the pathogen load difference decreases (i.e., the pathogen inactivation level at the target farm decreases), current disinfection measures are less effective at regulating the target farm, the disinfection efficacy assessment index decreases, and the compliance index declines. Disinfection measures for the target farm are dynamically adjusted based on the compliance index.

[0067] The disinfection process of the target farm is optimized as follows: The adjustment compliance deviation value of the target farm is obtained. If the environmental risk level of the target farm is low, it is updated to medium risk. Disinfection is then carried out on the target farm based on the second disinfection measure. Simultaneously, the adjustment compliance deviation value of the target farm is compared with the adjustment compliance deviation threshold. If the adjustment compliance deviation value is greater than the threshold, the second disinfection measure is not updated. If the adjustment compliance deviation value is less than or equal to the threshold, the second disinfection measure is updated. Specifically, the update process involves matching the reduction coefficient corresponding to the adjustment compliance deviation value from the early warning database based on the target farm's adjustment compliance deviation value, thereby updating the second disinfection measure.

[0068] The aforementioned reduction coefficient is used to multiply the disinfection frequency and ventilation power of the second disinfection measure, and the result is marked as the updated disinfection measure. After the second disinfection measure is implemented, if the deviation value of the adjustment compliance index of the target farm is less than or equal to the adjustment compliance deviation threshold, it means that the intensity of the current disinfection measure is too high. In this case, the disinfection frequency and ventilation power are adjusted by reducing the coefficient.

[0069] It should be explained that the adjustment compliance deviation threshold refers to the adjustment compliance index deviation value threshold, which is the maximum value of the adjustment compliance deviation value preset in the early warning database.

[0070] If the environmental risk level of the target farm is medium risk, the environmental risk level of the target farm will be updated to high risk, and disinfection treatment will be carried out on the target farm based on the third disinfection measure. At the same time, the adjustment compliance deviation value of the target farm will be compared with the adjustment compliance deviation threshold. If the adjustment compliance deviation value of the target farm is greater than the adjustment compliance deviation threshold, the third disinfection measure will not be updated, and an environmental risk warning will be issued to the target farm. If the adjustment compliance deviation value of the target farm is less than or equal to the adjustment compliance deviation threshold, the third disinfection measure will be updated. The specific update process is as follows: a reduction coefficient will be matched from the warning database based on the adjustment compliance deviation value of the target farm, thereby updating the third disinfection measure.

[0071] If the environmental risk level of the target farm is high, the target farm will be disinfected based on the third disinfection measure, and an environmental risk warning will be issued for the target farm at the same time.

[0072] It should be explained that the target farm's environmental risk level is medium risk, which means that the target farm is currently facing significant environmental risks. When making adjustments, the environmental risk level of the target farm will first be updated to high risk, and the corresponding third disinfection measures will be implemented. The adjustment compliance deviation value of the target farm will be obtained and compared with the threshold. If it is greater than the threshold, it means that the current disinfection measures of the target farm are reasonable and a high-risk environmental risk warning is required. If it is less than or equal to the threshold, it means that the current disinfection intensity of the target farm's disinfection measures is too high and the disinfection intensity needs to be reduced.

[0073] It should be explained that the adjustment compliance deviation value is obtained by subtracting the adjustment compliance index from the adjustment compliance index threshold defined in the early warning database, specifically: E_ACI=|ACI-J_ACI|;

[0074] In the formula, E_ACI is the deviation value of the target farm's regulation compliance index, ACI is the target farm's regulation compliance index, and J_ACI is the pre-defined regulation compliance index in the early warning database.

[0075] It should be explained that the adjustment compliance deviation threshold is the maximum value of the adjustment compliance deviation preset in the early warning database.

[0076] The safety early warning module is used to monitor and analyze the result parameters of the target farm through deep learning intelligence, thereby determining whether to issue a disinfection warning for the target farm.

[0077] In one specific embodiment, the present invention constructs a dynamic environmental resistance index and combines it with the time-series prediction of the environmental early warning coefficient to achieve graded triggering of disinfection early warning in future cycles (such as the first and second cycles). For example, when the predicted future environmental early warning coefficient exceeds the threshold, the system initiates pre-disinfection early warning in advance and dynamically adjusts the disinfection plan, effectively blocking the disease transmission chain and solving the core problem of the passive and lagging nature of the traditional early warning mechanism.

[0078] Specifically, the process for determining whether to issue a disinfection alert for the target farm is as follows:

[0079] By using deep learning to intelligently analyze the result parameters of the target farm, an environmental warning coefficient for the target farm is obtained and compared with an environmental warning threshold. If the environmental warning coefficient of the target farm is less than or equal to the environmental warning threshold, it is determined that no disinfection warning will be issued for the target farm; if the environmental warning coefficient of the target farm is greater than the environmental warning threshold, it is determined that a disinfection warning will be issued for the target farm.

[0080] The aforementioned environmental warning thresholds are the maximum values ​​allowed by the preset environmental warning coefficients in the warning database.

[0081] It needs to be explained that the disinfection early warning system generates decision suggestions based on the early warning results and displays them through the application service layer. If it is determined that the target farm needs a disinfection early warning, the disinfection measures for the next two cycles are adjusted according to the environmental early warning coefficient. The disinfection early warning is used to predict the environmental risk level for the next two cycles based on the predicted environmental early warning coefficient and issue an early warning. The disinfection early warning is implemented by modifying the preset disinfection measures (modifying the established disinfection frequency). After modifying and executing the preset disinfection measures, it is determined again according to the environmental early warning coefficient whether a disinfection early warning is needed. If so, the direction of the disinfection measures is changed (the type of disinfectant is modified). If it is determined that the target farm does not need a disinfection early warning, the current disinfection measures are maintained.

[0082] The aforementioned two future cycles refer to the cycles in which the environmental early warning coefficient can be efficiently predicted and adjusted. A predetermined period of time constitutes one cycle, and the two future cycles are predetermined cycles in which disinfection early warning can be carried out through the environmental early warning coefficient.

[0083] Furthermore, when the environmental warning coefficient is less than or equal to the environmental warning threshold, it indicates that the current environmental risk is within a controllable range, and conventional disinfection measures can maintain the environmental stability of the target farm, so no environmental warning is issued. When the environmental warning coefficient is greater than the environmental warning threshold, it means that the current environmental risk is high, and conventional disinfection measures cannot maintain the environmental stability of the target farm. In this case, an environmental warning is issued, and stronger disinfection measures need to be replaced. In summary, the disinfection warning for the target farm is dynamically adjusted by comparing the environmental warning coefficient with the environmental warning threshold.

[0084] Specifically, the environmental early warning coefficient of the target farm is analyzed through the following process: deep learning intelligent analysis of the target farm's parameters, including the residual pathogen load and the change in disinfection frequency. The residual pathogen load refers to the distribution density of pathogens after disinfection. The change in disinfection frequency refers to the change in disinfection frequency before and after the change in disinfection measures. The residual pathogen load can be obtained through pathogen concentration monitoring tools.

[0085] By introducing influence coefficients, the impact of the remaining pathogen load in the target farm on the dynamic environmental resistance index and the impact of changes in the disinfection frequency in the target farm on the dynamic environmental resistance index are quantified respectively. These influence levels are then coupled to analyze the dynamic environmental resistance index of the target farm. It should be noted that the dynamic environmental resistance index is specifically defined as follows:

[0086]

[0087] In the formula, DERI is the dynamic environmental resistance index of the target farm, RPL is the residual pathogen load of the target farm, DFC is the change in disinfection frequency of the target farm, J_RPL is the preset residual pathogen load in the early warning database, J_DFC is the preset change in disinfection frequency in the early warning database, kz1 is the influence coefficient corresponding to the preset residual pathogen load in the early warning database, and kz2 is the influence coefficient corresponding to the preset change in disinfection frequency in the early warning database.

[0088] It should be explained that the change in disinfection frequency is obtained by subtracting the disinfection frequency before and after disinfection of the target farm, specifically:

[0089] DFC = |DC - ΔDC|;

[0090] In the formula, DFC represents the change in disinfection frequency of the target farm, DC represents the disinfection frequency of the target farm before the disinfection measures were changed, and ΔDC represents the disinfection frequency of the target farm after the disinfection measures were changed.

[0091] The disinfection frequency before the change of disinfection measures refers to the disinfection frequency of the target farm before the adjustment of disinfection measures, and the disinfection frequency after the change of disinfection measures refers to the disinfection frequency of the target farm after the adjustment of disinfection measures.

[0092] Simultaneously, the environmental risk index and the regulation compliance index of the target farm are introduced and coupled twice to analyze the environmental early warning coefficient of the target farm. The specific expression is as follows:

[0093]

[0094] In the formula, EAC is the environmental early warning coefficient of the target farm, ERI is the environmental risk index of the target farm, ACI is the regulation compliance index of the target farm, DERI is the dynamic environmental resistance index of the target farm, J_ERI is the pre-defined environmental risk index in the early warning database, J_ACI is the pre-defined regulation compliance index in the early warning database, J_DERI is the pre-defined dynamic environmental resistance index in the early warning database, kh1 is the influence coefficient corresponding to the pre-defined environmental risk index in the early warning database, kh2 is the influence coefficient corresponding to the pre-defined regulation compliance index in the early warning database, and kh3 is the influence coefficient corresponding to the pre-defined dynamic environmental resistance index in the early warning database.

[0095] The regulation compliance attenuation term for the target farm is used to characterize the inhibitory effect of the regulation compliance index on environmental early warning.

[0096] The aforementioned definition of the environmental risk index represents the maximum permissible value of the environmental risk index; the aforementioned definition of the adjustment compliance index represents the minimum permissible value of the adjustment compliance index; the aforementioned definition of the dynamic environmental resistance index represents the minimum permissible value of the environmental risk index; the aforementioned impact coefficients corresponding to the environmental risk indices represent the degree of influence of the environmental risk indices on the environmental early warning coefficients; the aforementioned impact coefficients corresponding to the adjustment compliance indices represent the degree of influence of the adjustment compliance indices on the environmental early warning coefficients; and the aforementioned impact coefficients corresponding to the dynamic environmental resistance indexes represent the degree of influence of the dynamic environmental resistance indexes on the environmental early warning coefficients.

[0097] It should be explained that the environmental risk index reflects the degree of environmental risk of the target farm. A high environmental risk index means a high-risk environment, and the environmental warning coefficient increases accordingly. The regulation compliance index reflects the compliance of the disinfection measures of the target farm. The higher the regulation compliance index, the more suitable the current disinfection measures are for the current target farm, and the environmental warning coefficient decreases accordingly. The dynamic environmental resistance index reflects the ability of the target farm to maintain environmental homeostasis. An increase in the dynamic environmental resistance index means that the current intensity of suppressing the risk of the target farm is stronger, and the environmental warning coefficient decreases accordingly.

[0098] Disinfection alerts are issued for target farms. The specific alert process is as follows: Based on the environmental alert coefficient of the target farm, if the environmental alert coefficient is less than the defined environmental alert coefficient, no disinfection alert is issued for the target farm, and routine disinfection measures are not adjusted. If the environmental alert coefficient is greater than the defined environmental alert coefficient, a disinfection alert is issued for the target farm, and a disinfection alert plan for the target farm is matched from the alert database. At the same time, the reduction rate of the environmental alert coefficient of the target farm in the first cycle is detected and compared with the reduction rate of the defined environmental alert coefficient. If the reduction rate of the environmental alert coefficient of the target farm in the first cycle is greater than the reduction rate of the defined environmental alert coefficient, the disinfection alert for the target farm is lifted. If the reduction rate of the environmental alert coefficient of the target farm in the first cycle is less than or equal to the reduction rate of the defined environmental alert coefficient, disinfection alert feedback is continuously issued to the target farm (based on the disinfection alert feedback, the disinfection alert plan for the target farm is updated, and the reduction rate of the environmental alert coefficient of the target farm in the first cycle is detected again and compared with the reduction rate of the defined environmental alert coefficient, until the reduction rate of the environmental alert coefficient of the target farm in the first cycle is greater than the reduction rate of the defined environmental alert coefficient).

[0099] The aforementioned reduction rate of the environmental warning coefficient was obtained by processing the difference between the environmental warning coefficient before and after the implementation of the disinfection warning plan, and then comparing it with the environmental warning coefficient after the implementation of the disinfection warning plan.

[0100] It should be explained that the reduction rate of the environmental early warning coefficient is defined as the minimum allowable value preset in the early warning database.

[0101] It needs to be explained that updating the disinfection warning plan for the target farm involves obtaining the number of times the disinfection warning plan has been updated based on the feedback (the preset maximum number of updates is three; if, after three updates, the reduction rate of the environmental warning coefficient within the first cycle of the target farm is still not greater than the defined reduction rate, an emergency alarm will be immediately triggered, audible and visual alarms will be activated, and the last disinfection warning plan will continue to be executed). The disinfection warning plan level is then matched, and the disinfection warning plan before the first update is marked as the baseline disinfection warning plan. Simultaneously, the level of the disinfection warning plan is marked according to the number of updates (when the number of updates is one, the disinfection warning plan is a Level 1 disinfection warning plan, and so on). Specifically, the baseline disinfection warning plan is indicated by a system pop-up visual window showing the administrator that the target farm needs to adjust the preset disinfection plan and adjust the disinfection frequency of the standard disinfectant according to the regular disinfection plan; the Level 1 disinfection warning plan is indicated by a system pop-up visual window... A pop-up window will display to the administrator that the target farm needs to adjust the preset disinfection plan and adjust the regular disinfection plan according to the baseline disinfection plan, while maintaining regular ventilation management (adjusting the disinfection frequency of the standard disinfectant); the second-level disinfection warning plan will display a visual pop-up window to the administrator that the target farm needs to adjust the preset disinfection plan, and the administrator will replace it with a strong disinfection plan according to the first-level disinfection warning plan, while strengthening ventilation management and increasing the frequency of water source replacement (replacing with a strong disinfectant, increasing the power of the ventilation system, and increasing the frequency of water source replacement for the target farm); the third-level disinfection warning plan will display a visual pop-up window to the administrator that the target farm needs to adjust the preset disinfection plan, and the administrator will replace it with an emergency disinfection plan according to the second-level disinfection warning plan, while implementing emergency ventilation, replacing the live water source, and implementing light disinfection measures (fully enclosing the area, activating ozone high-pressure pulse disinfection, running all ventilation systems of the target farm at full power, activating the dynamic live water system, and activating ultraviolet disinfection throughout the area).

[0102] It needs to be explained that matching a disinfection early warning scheme involves matching the environmental early warning coefficient of the target farm with the result parameters of the target farm corresponding to the environmental early warning coefficient interval of each target farm in the early warning database. This means that the environmental early warning coefficient of the target farm corresponds to the result parameters of the target farm, and a disinfection early warning scheme is matched based on the target farm's environmental early warning coefficient interval to which the environmental early warning coefficient of the target farm belongs.

[0103] In one specific embodiment, the present invention provides a deep learning-based intelligent monitoring and early warning system and method for large-scale livestock farms. By collecting and analyzing environmental parameters in real time, the system automatically matches environmental risk levels and performs corresponding disinfection treatments based on the levels. Compared with traditional manual monitoring and experience-based judgment, this intelligent monitoring and treatment method can more accurately and timely identify and respond to environmental risks in livestock farms, improving the targeting and effectiveness of disinfection treatments. Through a dynamic control module, the system can dynamically adjust the disinfection treatment plan based on the evaluation results of disinfection process parameters, further improving the system's intelligence level and management efficiency. The present invention provides a comprehensive and intelligent environmental monitoring and disinfection management solution for large-scale livestock farms, providing strong support for the sustainable development of the livestock industry.

[0104] Reference Figure 2 As shown, the second aspect of the present invention provides a method for intelligent monitoring and early warning of large-scale farms based on deep learning, comprising: Step 1, marking large-scale farms as target farms to match the environmental risk level of the target farms, and disinfecting the target farms based on the environmental risk level of the target farms; Step 2, collecting and evaluating the disinfection process parameters of the target farms, determining whether to update the environmental risk level of the target farms, thereby optimizing the disinfection process of the target farms; Step 3, monitoring and intelligently analyzing the result parameters of the target farms through deep learning to determine whether to issue a disinfection early warning for the target farms.

[0105] Detailed process as follows Figure 3 As shown: First, environmental parameters of the target farm are collected and analyzed to calculate the environmental risk index. This index is then compared with the environmental risk index range to match the environmental risk level. If the environmental risk index of the target farm is in the low-risk range, the standard disinfection plan is maintained; if the environmental risk index is in the medium-risk range, an enhanced disinfection plan is implemented; if the environmental risk index is in the high-risk range, an emergency disinfection plan is implemented; then... Figure 4 As shown: Calculate the regulation compliance index of the target farm, perform regulation compliance verification. If it is qualified, maintain the current disinfection plan; if it is not qualified, calculate the regulation compliance deviation value, and determine the standard based on the deviation value threshold. If the deviation value does not exceed the standard, maintain the current plan; if it exceeds the standard, trigger disinfection feedback; then proceed as follows... Figure 5 As shown: Obtain the environmental early warning coefficient of the target farm, compare it with the threshold, if it is less than or equal to the threshold, no early warning will be issued and routine disinfection measures will be maintained, if it is greater than the threshold, a disinfection early warning will be triggered.

[0106] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A deep learning-based intelligent monitoring and early warning system for large-scale livestock farms, characterized in that, include: The environmental analysis module is used to mark large-scale farms as target farms, collect and analyze the environmental parameters of the target farms, thereby matching the environmental risk level of the target farms, and disinfecting the target farms based on the environmental risk level of the target farms; The specific analysis process for collecting and analyzing environmental parameters from the target farm is as follows: The environmental parameters of the target farm include the pathogen load of the target farm, the temperature-humidity coupling index of the target farm, and the rate of environmental risk decay of the target farm. The temperature and humidity coupling index of the target farm is obtained by introducing an influence coefficient to quantify the degree of influence of the ambient temperature and humidity of the target farm on the temperature and humidity coupling index, and then coupling the degree of influence. By introducing influence coefficients, the influence of pathogen load component of the target farm on the environmental risk index, the influence of temperature and humidity coupling index component of the target farm on the environmental risk index, and the influence of environmental risk decay rate component of the target farm on the environmental risk index are quantified respectively. The influence of each degree is coupled to analyze the environmental risk index of the target farm. The environmental risk index of the farm is used to characterize the degree of environmental risk of the target farm; The dynamic control module is used to collect and evaluate the disinfection process parameters of the target farm, determine whether to update the environmental risk level of the target farm, and further determine whether the measures need to be updated, thereby optimizing the disinfection process of the target farm. The collection and evaluation of disinfection process parameters at the target farm are specifically evaluated as follows: The disinfection process parameters of the target farm include the disinfection success rate of the target farm, the residual concentration of disinfectant in the target farm, and the pathogen load deviation value of the target farm. By introducing influence coefficients, the degree of influence of the target farm's successful disinfection growth rate on the comprehensive evaluation index of disinfection efficiency, the degree of influence of the target farm's disinfectant residue concentration on the comprehensive evaluation index of disinfection efficiency, and the degree of influence of the target farm's pathogen load deviation value on the comprehensive evaluation index of disinfection efficiency are quantified respectively. The degree of influence is coupled to evaluate the comprehensive evaluation index of disinfection efficiency of the target farm. At the same time, the environmental risk index of the target farm is introduced and coupled a second time to obtain the regulation compliance index of the target farm. The comprehensive evaluation index of disinfection efficiency of the target farm is used to characterize the overall effect of disinfection measures; The regulation compliance index of the target farm is used to characterize the degree of compliance of the regulation operation, thereby determining whether to update the environmental risk level of the target farm. The safety early warning module is used to monitor and analyze the result parameters of the target farm through deep learning intelligent analysis. The specific early warning process for monitoring and analyzing the result parameters of the target farm through deep learning intelligent analysis is as follows: The result parameters of the target farm include the remaining pathogen load of the target farm and the change in the disinfection frequency of the target farm. By introducing influence coefficients, the impact of the remaining pathogen load in the target farm on the dynamic environmental resistance index and the impact of the change in the disinfection frequency on the dynamic environmental resistance index are quantified respectively. The influence levels are coupled to analyze the dynamic environmental resistance index of the target farm. At the same time, the environmental risk index and the regulation compliance index of the target farm are introduced and coupled again to analyze the environmental warning coefficient of the target farm. Then, it is determined whether to issue a disinfection warning for the target farm. In addition, the reduction rate of the environmental warning coefficient of the target farm in the first cycle is detected and compared with the defined reduction rate of the environmental warning coefficient to determine whether to provide feedback.

2. The intelligent monitoring and early warning system for large-scale livestock farms based on deep learning according to claim 1, characterized in that: The environmental risk level of the target farm is matched, and the specific matching process is as follows: Collect and analyze environmental parameters of the target farm to obtain the environmental risk index of the target farm. If the environmental risk index of the target farm is in the low risk index range, mark the environmental risk level of the target farm as low risk level and use the first disinfection measure to disinfect the target farm. If the environmental risk index of the target farm is in the medium risk range, the environmental risk level of the target farm will be marked as medium risk level, and the second disinfection measure will be used to disinfect the target farm. If the environmental risk index of the target farm is in the high-risk range, the environmental risk level of the target farm will be marked as high-risk, and the third disinfection measure will be used to disinfect the target farm. At the same time, an environmental risk warning will be issued for the target farm.

3. The intelligent monitoring and early warning system for large-scale livestock farms based on deep learning according to claim 1, characterized in that: The specific process for determining whether to update the environmental risk level of the target farm is as follows: Collect and obtain the disinfection process parameters of the target farm, obtain the regulation compliance index of the target farm, and compare it with the regulation compliance threshold. If the regulation compliance index is greater than or equal to the regulation compliance index threshold, it is determined that the environmental risk level of the target farm will not be updated. If the adjustment compliance index is less than the adjustment compliance index threshold, the environmental risk level of the target farm will be updated, thereby optimizing the disinfection process of the target farm.

4. The intelligent monitoring and early warning system for large-scale livestock farms based on deep learning according to claim 3, characterized in that: The optimization process for the disinfection of the target farm is as follows: The adjustment compliance deviation value of the target farm is obtained. If the environmental risk level of the target farm is low, it is updated to medium risk. Disinfection is then performed on the target farm based on the second disinfection measure. Simultaneously, the adjustment compliance deviation value of the target farm is compared with the adjustment compliance deviation threshold. If the adjustment compliance deviation value is within the threshold, the second disinfection measure is not updated. If the adjustment compliance deviation value is less than or equal to the threshold, the second disinfection measure is updated. Specifically, the update process involves matching a reduction coefficient from the early warning database based on the adjustment compliance deviation value of the target farm, thereby updating the second disinfection measure.

5. The intelligent monitoring and early warning system for large-scale livestock farms based on deep learning according to claim 1, characterized in that: The specific determination process for whether to issue a disinfection alert for the target farm is as follows: By using deep learning to intelligently analyze the result parameters of the target farm, the environmental warning coefficient of the target farm is obtained and compared with the environmental warning threshold. If the environmental warning coefficient of the target farm is less than or equal to the environmental warning threshold, it is determined that no disinfection warning will be issued for the target farm. If the environmental warning coefficient of the target farm is greater than the environmental warning threshold, it is determined that the target farm will be subject to disinfection warning.

6. The intelligent monitoring and early warning system for large-scale livestock farms based on deep learning according to claim 5, characterized in that: The disinfection warning process for the target farm is as follows: if the environmental warning coefficient of the target farm is less than the defined environmental warning coefficient, then no disinfection warning will be issued for the target farm, and routine disinfection measures will not be adjusted. If the environmental warning coefficient is greater than the defined environmental warning coefficient, a disinfection warning is issued for the target farm, and a disinfection warning plan for the target farm is matched from the warning database. At the same time, the reduction rate of the environmental warning coefficient of the target farm in the first cycle is detected and compared with the reduction rate of the defined environmental warning coefficient. If the reduction rate of the environmental warning coefficient of the target farm in the first cycle is greater than the reduction rate of the defined environmental warning coefficient, the disinfection warning for the target farm is lifted. If the reduction rate of the environmental warning coefficient of the target farm in the first cycle is less than or equal to the reduction rate of the defined environmental warning coefficient, the disinfection warning feedback for the target farm continues.

7. A method for applying the deep learning-based intelligent monitoring and early warning system for large-scale livestock farms as described in any one of claims 1-6, characterized in that: include: Step 1: Used to mark large-scale farms as target farms, thereby matching the environmental risk level of the target farms, and disinfecting the target farms based on the environmental risk level of the target farms; Step 2: Collect and evaluate the disinfection process parameters of the target farm, collect and analyze the environmental parameters of the target farm, determine whether to update the environmental risk level of the target farm, and thus optimize the disinfection process of the target farm. Step 3: The results parameters of the target farm are monitored and analyzed using deep learning to determine whether to issue a disinfection warning for the target farm.

Images