Intelligent monitoring system for artificial feed breeding environment of silkworms
By drawing data change curves in the silkworm breeding environment, identifying sensor locations and using artificial intelligence models to predict risks, the problems of inaccurate monitoring and difficulty in prediction in the existing technology are solved, and efficient monitoring and health protection of the silkworm environment are achieved. .
Patent Information
- Application Number
- CN202411176402.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-26
- Publication Date
- 2024-12-03
AI Technical Summary
In the existing silkworm breeding environment monitoring system, it is difficult to focus on monitoring the location where environmental data changes rapidly, resulting in insufficient monitoring and difficult to predict risk status, resulting in the occurrence of diseases caused by environmental abnormalities in silkworms.
An intelligent monitoring system for artificial feed breeding of silkworms has been designed, including a central processing module, a data acquisition module and a risk warning module. By drawing environmental data change curves, deriving and determining the installation location of the sensor, calculating the environmental risk coefficient and using an artificial intelligence model to perform it. Forecast, determine the risk warning level to adjust the environmental conditions.
It improves the accuracy of monitoring the silkworm feeding environment, estimates environmental risks in advance, reduces the occurrence of diseases caused by environmental abnormalities, and ensures the health of silkworms.
Smart Images

Figure CN119067446A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of silkworm breeding, and in particular is an intelligent monitoring system for silkworm artificial feed breeding environment. Background Art
[0002] Environmental factors play a vital role in the breeding process of silkworms. The suitability of environmental parameters such as temperature, humidity, light intensity, and air circulation directly affects the growth and development of silkworms, their health, and the yield and quality of cocoons. Therefore, how to timely and accurately monitor the breeding environment of silkworms has become a technical problem that needs to be solved urgently.
[0003] The prior art monitors environmental parameters such as temperature, humidity, carbon dioxide concentration and light intensity in the silkworm breeding environment through relevant sensors, and compares the environmental parameters with corresponding preset thresholds to determine whether the silkworm breeding environment is abnormal; however, the relevant sensors in the monitoring system of the prior art are evenly distributed in the silkworm breeding environment, and it is difficult to focus on monitoring the locations where environmental data in the breeding environment changes rapidly, resulting in inaccurate monitoring of the silkworm breeding environment; in addition, the prior art requires that the environmental parameters exceed the corresponding thresholds to determine whether the breeding environment is abnormal, and it is difficult to predict the risk status of the breeding environment, resulting in the monitoring of the silkworm breeding environment not being timely and accurate, which in turn causes the silkworms to become sick due to the abnormal breeding environment, making it difficult to ensure the health of the silkworms.
[0004] The present invention provides an intelligent monitoring system for a silkworm artificial feed breeding environment to solve the above technical problems. Summary of the invention
[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent monitoring system for an artificial silkworm breeding environment, which is used to solve the problem that in the monitoring system of the prior art, relevant sensors are evenly distributed in the silkworm breeding environment, making it difficult to focus on monitoring the locations in the breeding environment where environmental data changes rapidly, resulting in inaccurate monitoring of the silkworm breeding environment; in addition, the prior art requires that environmental parameters exceed corresponding thresholds to determine whether the breeding environment is abnormal, making it difficult to predict the risk status of the breeding environment, resulting in insufficiently timely and accurate monitoring of the silkworm breeding environment, and further resulting in the technical problem that the silkworms become diseased due to abnormal breeding environments and it is difficult to ensure the health of the silkworms.
[0006] To achieve the above-mentioned object, the first aspect of the present invention provides an intelligent monitoring system for silkworm artificial feed breeding environment, comprising: a central processing module, and a data acquisition module and a risk warning module connected thereto;
[0007] The data acquisition module is used to collect environmental data of several positions in a breeding environment at several consecutive periods and the health status of silkworms at each position; the environmental status and the health status are preprocessed to obtain environmental data and health data; wherein the environmental data include the temperature, humidity, carbon dioxide concentration and light intensity of the breeding environment, and the health data include the food intake and body length of the silkworms;
[0008] The central processing module is used to determine the installation location of relevant sensors based on the environmental data of several consecutive cycles at each location; wherein the relevant sensors include temperature sensors, humidity sensors, carbon dioxide concentration sensors and light intensity sensors; and,
[0009] Calculate the environmental risk coefficient based on environmental data and health data; input the environmental risk coefficient into the environmental risk prediction model to obtain the predicted value of the environmental risk coefficient; wherein the environmental risk prediction model is obtained by training the artificial intelligence model;
[0010] The risk warning module is used to determine the risk warning level based on the predicted value of the environmental risk coefficient; adjust the environmental conditions based on the risk warning level; wherein the risk warning level includes normal, level one warning and level two warning, and the risk severity is level one warning>level two warning>normal.
[0011] Preferably, the preprocessing of the environmental state and the health state respectively comprises:
[0012] A1: Extract the environmental status and health status of each location for several consecutive periods;
[0013] A2: Eliminate abnormal data in environmental conditions and health status;
[0014] A3: Fill in missing data on environmental and health conditions;
[0015] A4: Mark the preprocessed environment state as environment data;
[0016] A5: Mark the preprocessed health status as healthy data.
[0017] Preferably, the step of determining the installation location of the relevant sensors based on the environmental data of several consecutive periods at each location includes:
[0018] B1: Extract environmental data of several consecutive periods at each location;
[0019] B2: Draw the corresponding environmental data change curve based on the environmental data of several consecutive periods at each location;
[0020] B3: Deriving the environmental data change curve to obtain the environmental data derivative function;
[0021] B4: Obtain the derivative value of the environmental data at each location and compare it with the corresponding derivative threshold;
[0022] B5: If the derivative value of the environmental data is greater than the corresponding derivative threshold, the corresponding position is marked as the installation position; otherwise, the corresponding position is marked as the normal position.
[0023] It should be noted that the environmental data change curves drawn include temperature change curves, humidity change curves, carbon dioxide concentration change curves and light intensity change curves, and the environmental data derivative functions include temperature derivative functions, humidity derivative functions, carbon dioxide concentration derivative functions and light intensity derivative functions.
[0024] The present invention draws an environmental data change curve, and derives the environmental data change curve to obtain an environmental data derivative function; when the environmental data derivative value is greater than the corresponding derivative threshold, the corresponding position is marked as the installation position, thereby determining the installation position of the relevant sensor, so that the relevant sensor can focus on monitoring the position where the environmental data changes quickly in the breeding environment, which is beneficial to improving the accuracy of monitoring the breeding environment of silkworms.
[0025] Preferably, the step of drawing a corresponding environmental data change curve based on environmental data of several consecutive periods at each location includes:
[0026] The environmental data of several contact periods are extracted, and the environmental data change curve is established with time as the independent variable and environmental data as the dependent variable.
[0027] Preferably, the calculation of the environmental risk coefficient based on the environmental data and the health data includes:
[0028] C1: Extract environmental data and health data;
[0029] C2: Calculate the environmental score HP based on environmental data;
[0030] C3: Calculate the health score KP based on health data;
[0031] C4: The environmental risk factor HFX is calculated by the formula HFX = (a × HP) / [(b × KP) + c]; where a and b are influencing factors greater than 0; c is a constant to prevent division by zero, and the value range of c is (0,1].
[0032] Preferably, the calculating of the environmental score HP based on the environmental data comprises:
[0033] The temperature WD, humidity SD, carbon dioxide concentration TN and light intensity GZ in the environmental data are extracted, and the environmental score HP is calculated by the formula HP=k1×|WD-WD0|+k2×|SD-SD0|+k3×|TN-TN0|+k4×|GZ-GZ0|; among which WD0 is the temperature reference value, SD0 is the humidity reference value, TN0 is the carbon dioxide concentration reference value, GZ0 is the light intensity reference value, k1, k2, k3, k4 are proportional coefficients greater than 0, and || is the absolute value operator.
[0034] Preferably, the calculating of the health score KP based on the health data comprises:
[0035] The health data of food intake JSL and body length TC are extracted, and the health score KP is calculated by the formula KP=k5×JSL+k6×arctan(TC); wherein k5 and k6 are proportional coefficients greater than 0, and arctan() is the inverse tangent function.
[0036] Preferably, the environmental risk prediction model is obtained by training an artificial intelligence model, including:
[0037] The environmental risk coefficients of several consecutive cycles are extracted and integrated into several groups of training data and test data; the artificial intelligence model is trained using the training data; the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted according to the test results; finally, a compressor failure prediction model is obtained, whose input is the environmental risk coefficients of the most recent several consecutive cycles and whose output is the environmental risk coefficient of the predicted cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.
[0038] The present invention extracts environmental risk coefficients for training artificial intelligence models, obtains an environmental risk prediction model after the training is completed, and obtains the environmental risk coefficients of the prediction period by inputting the environmental risk coefficients of several consecutive periods into the environmental risk prediction model. The risk changes of the silkworm breeding environment can be predicted in advance and corresponding measures can be taken, thereby reducing the situation where silkworms develop diseases due to abnormal breeding environment, which is beneficial to ensuring the health of silkworms.
[0039] Preferably, the step of determining the risk warning level based on the predicted value of the environmental risk coefficient includes:
[0040] T1: Extract the predicted value of environmental risk factor;
[0041] T2: Set the risk range threshold [FY1, FY2]; the specific values of FY1 and FY2 are set according to the experience of the technicians, and FY1 <FY2;
[0042] T3: Determine whether the predicted value of the environmental risk factor is within the risk range threshold [FY1, FY2]; if yes, mark the risk warning level as level 2 warning; if no, jump to T4;
[0043] T4: Determine whether the predicted value of the environmental risk factor is greater than the maximum value FY2 in the risk range threshold; if yes, mark the risk warning level as level one warning; if not, mark the risk warning level as normal.
[0044] Preferably, the adjusting of environmental conditions based on risk warning levels includes:
[0045] Z1: Extract risk warning level;
[0046] Z2: Determine whether the risk warning level is level 1 warning; if yes, adjust the environmental conditions of the breeding environment; if no, jump to Z3;
[0047] Z3: Determine whether the risk warning level is level 2 warning; if yes, increase the frequency of collecting data on the environmental conditions of the breeding environment and the health status of the silkworms; if no, jump to Z1.
[0048] Compared with the prior art, the present invention has the following beneficial effects:
[0049] 1. The present invention draws an environmental data change curve, and derives the environmental data change curve to obtain an environmental data derivative function; if the environmental data derivative value is greater than the corresponding derivative threshold, the corresponding position is marked as the installation position, thereby determining the installation position of the relevant sensor, so that the relevant sensor can focus on monitoring the position where the environmental data changes quickly in the breeding environment, which is beneficial to improving the accuracy of monitoring the breeding environment of silkworms.
[0050] 2. The present invention extracts environmental risk coefficients for training artificial intelligence models, and obtains an environmental risk prediction model after the training is completed. By inputting environmental risk coefficients of several consecutive periods into the environmental risk prediction model, the environmental risk coefficients of the prediction period are obtained. The risk changes in the breeding environment of silkworms can be predicted in advance and corresponding measures can be taken, thereby reducing the situation where silkworms become diseased due to abnormal breeding environment, which is beneficial to ensuring the health of silkworms. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0052] Figure 1It is a schematic diagram of the principle of the intelligent monitoring system for the artificial feed breeding environment of silkworms of the present invention;
[0053] Figure 2 It is an overall flow chart of the intelligent monitoring method for the artificial feed breeding environment of silkworms of the present invention;
[0054] Figure 3 A flow chart for determining the installation position of relevant sensors in the present invention;
[0055] Figure 4 The present invention is a flow chart for determining the risk warning level of the silkworm breeding environment. DETAILED DESCRIPTION
[0056] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0057] See also Figure 1-Figure 4 , the first aspect of the present invention provides an intelligent monitoring system for a silkworm artificial feed breeding environment, comprising: a central processing module, and a data acquisition module and a risk warning module connected thereto;
[0058] Data acquisition module: used to collect environmental data of several positions in several consecutive periods in the breeding environment and the health status of the silkworms at each position; pre-process the environmental status and health status respectively to obtain environmental data and health data; wherein the environmental data includes the temperature, humidity, carbon dioxide concentration and light intensity of the breeding environment, and the health data includes the food intake and body length of the silkworms;
[0059] Central processing module: used to determine the installation location of relevant sensors based on several consecutive cycles of environmental data at each location; wherein the relevant sensors include temperature sensors, humidity sensors, carbon dioxide concentration sensors and light intensity sensors; and,
[0060] Calculate the environmental risk coefficient based on environmental data and health data; input the environmental risk coefficient into the environmental risk prediction model to obtain the predicted value of the environmental risk coefficient; wherein the environmental risk prediction model is obtained by training the artificial intelligence model;
[0061] Risk warning module: used to determine the risk warning level based on the predicted value of the environmental risk factor; adjust the environmental conditions based on the risk warning level; among them, the risk warning level includes normal, level one warning and level two warning, and the risk severity is level one warning>level two warning>normal.
[0062] In this embodiment, the environmental state and the health state are preprocessed separately, including:
[0063] A1: Extract the environmental status and health status of each location for several consecutive periods;
[0064] A2: Eliminate abnormal data in environmental conditions and health status;
[0065] A3: Fill in missing data on environmental and health conditions;
[0066] A4: Mark the preprocessed environment state as environment data;
[0067] A5: Mark the preprocessed health status as healthy data.
[0068] In this embodiment, the installation location of the relevant sensor is determined based on the environmental data of several consecutive periods at each location, including:
[0069] B1: Extract environmental data of several consecutive periods at each location;
[0070] B2: Draw the corresponding environmental data change curve based on the environmental data of several consecutive periods at each location;
[0071] B3: Deriving the environmental data change curve to obtain the environmental data derivative function;
[0072] B4: Obtain the derivative value of the environmental data at each location and compare it with the corresponding derivative threshold;
[0073] B5: If the derivative value of the environmental data is greater than the corresponding derivative threshold, the corresponding position is marked as the installation position; otherwise, the corresponding position is marked as the normal position.
[0074] It should be noted that the environmental data change curves drawn include temperature change curves, humidity change curves, carbon dioxide concentration change curves and light intensity change curves, and the environmental data derivative functions include temperature derivative functions, humidity derivative functions, carbon dioxide concentration derivative functions and light intensity derivative functions.
[0075] Exemplarily, there are three positions A1, A2 and A3 in the silkworm rearing environment; among them,
[0076] The temperature derivative value corresponding to A1 is 0.25, the humidity derivative value is 0.62, the carbon dioxide concentration derivative value is 0.18, and the light intensity derivative value is 0.3;
[0077] The temperature derivative value corresponding to A2 is 0.15, the humidity derivative value is 0.47, the carbon dioxide concentration derivative value is 0.15, and the light intensity derivative value is 0.1;
[0078] The temperature derivative value corresponding to A3 is 0.18, the humidity derivative value is 0.45, the carbon dioxide concentration derivative value is 0.1, and the light intensity derivative value is 0.2;
[0079] Set the derivative threshold corresponding to temperature to DY1 = 0.2, the derivative threshold corresponding to humidity to DY2 = 0.5, the derivative threshold corresponding to carbon dioxide concentration to DY3 = 0.2, and the derivative threshold corresponding to light intensity to DY4 = 0.4;
[0080] The derivative values of the environmental data corresponding to the three positions A1, A2 and A3 are compared with the corresponding derivative thresholds, respectively, and it is obtained that A1 is the installation position, A2 is the normal position, and A3 is the normal position.
[0081] The present invention draws an environmental data change curve, and derives the environmental data change curve to obtain an environmental data derivative function; when the environmental data derivative value is greater than the corresponding derivative threshold, the corresponding position is marked as the installation position, thereby determining the installation position of the relevant sensor, so that the relevant sensor can focus on monitoring the position where the environmental data changes quickly in the breeding environment, which is beneficial to improving the accuracy of monitoring the breeding environment of silkworms.
[0082] In this embodiment, the corresponding environmental data change curve is drawn based on the environmental data of several consecutive periods at each location, including:
[0083] The environmental data of several contact periods are extracted, and the environmental data change curve is established with time as the independent variable and environmental data as the dependent variable.
[0084] In this embodiment, the environmental risk coefficient is calculated based on the environmental data and the health data, including:
[0085] C1: Extract environmental data and health data;
[0086] C2: Calculate the environmental score HP based on environmental data;
[0087] C3: Calculate the health score KP based on health data;
[0088] C4: The environmental risk factor HFX is calculated by the formula HFX = (a × HP) / [(b × KP) + c]; where a and b are influencing factors greater than 0; c is a constant to prevent division by zero, and the value range of c is (0,1].
[0089] Exemplarily, a=0.1, b=1.5, c=0.01 are set; the environmental score HP=18, the health score KP=1.087; and the environmental risk coefficient HFX≈1.097 is calculated by the formula.
[0090] In this embodiment, the environmental score HP is calculated based on the environmental data, including:
[0091] The temperature WD, humidity SD, carbon dioxide concentration TN and light intensity GZ in the environmental data are extracted, and the environmental score HP is calculated by the formula HP=k1×|WD-WD0|+k2×|SD-SD0|+k3×|TN-TN0|+k4×|GZ-GZ0|; among which WD0 is the temperature reference value, SD0 is the humidity reference value, TN0 is the carbon dioxide concentration reference value, GZ0 is the light intensity reference value, k1, k2, k3, k4 are proportional coefficients greater than 0, and || is the absolute value operator.
[0092] Exemplarily, set temperature WD=24°C, humidity SD=73%, carbon dioxide concentration TN=0.3%, and light intensity GZ=320lx; temperature reference value WD0=26°C, humidity reference value SD0=75%, carbon dioxide concentration reference value TN0=0.2%, and light intensity reference value GZ0=400lx; k1=2, k2=150, k3=3000, k4=0.1; and calculate the environmental score HP=18 through the formula.
[0093] In this embodiment, the health score KP is calculated based on the health data, including:
[0094] The health data of food intake JSL and body length TC are extracted, and the health score KP is calculated by the formula KP=k5×JSL+k6×arctan(TC); wherein k5 and k6 are proportional coefficients greater than 0, and arctan() is the inverse tangent function.
[0095] Exemplarily, the food intake JSL=4 pieces, the body length TC=5 cm, the proportional coefficients k5=0.1, k6=0.5 are set; the health score KP≈1.087 is calculated by the formula.
[0096] It should be noted that the food intake of silkworms refers to the number of mulberry leaves consumed by each silkworm per day.
[0097] In this embodiment, the environmental risk prediction model is obtained through artificial intelligence model training, including:
[0098] The environmental risk coefficients of several consecutive cycles are extracted and integrated into several groups of original data, 80% of the original data are used as training data and 20% as test data; the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted according to the test results; finally, a compressor fault prediction model is obtained, whose input is the environmental risk coefficients of the most recent several consecutive cycles and whose output is the environmental risk coefficient of the predicted cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.
[0099] The present invention extracts environmental risk coefficients for training artificial intelligence models, obtains an environmental risk prediction model after the training is completed, and obtains the environmental risk coefficients of the prediction period by inputting the environmental risk coefficients of several consecutive periods into the environmental risk prediction model. The risk changes of the silkworm breeding environment can be predicted in advance and corresponding measures can be taken, thereby reducing the situation where silkworms develop diseases due to abnormal breeding environment, which is beneficial to ensuring the health of silkworms.
[0100] In this embodiment, the risk warning level is determined based on the predicted value of the environmental risk factor, including:
[0101] T1: Extract the predicted value of environmental risk factor;
[0102] T2: Set the risk range threshold [FY1, FY2]; the specific values of FY1 and FY2 are set according to the experience of the technicians, and FY1 <FY2;
[0103] T3: Determine whether the predicted value of the environmental risk factor is within the risk range threshold [FY1, FY2]; if yes, mark the risk warning level as level 2 warning; if no, jump to T4;
[0104] T4: Determine whether the predicted value of the environmental risk factor is greater than the maximum value FY2 in the risk range threshold; if yes, mark the risk warning level as level one warning; if not, mark the risk warning level as normal.
[0105] For example, the environmental risk factor prediction value is set to HPX 预 =1.2, risk range threshold [FY1, FY2] = [0.9, 1.3]; since the environmental risk factor prediction value is HPX 预 It is within the risk range threshold [FY1, FY2], so the risk warning level is marked as level 2 warning.
[0106] In this embodiment, the environmental conditions are adjusted based on the risk warning level, including:
[0107] Z1: Extract risk warning level;
[0108] Z2: Determine whether the risk warning level is level 1 warning; if yes, adjust the environmental conditions of the breeding environment; if no, jump to Z3;
[0109] Z3: Determine whether the risk warning level is level 2 warning; if yes, increase the frequency of collecting data on the environmental conditions of the breeding environment and the health status of the silkworms; if no, jump to Z1.
[0110] Exemplarily, the risk warning level is set to level 2 warning, and the frequency of collecting environmental conditions through relevant sensors in the breeding environment is increased to once every 30 minutes.
[0111] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.
[0112] Working principle of the present invention:
[0113] The present invention collects environmental data of several positions in several continuous periods in a breeding environment and the health status of silkworms corresponding to each position; pre-processes the environmental status and the health status respectively to obtain environmental data and health data; determines the installation position of relevant sensors based on the environmental data of several continuous periods at each position; calculates the environmental risk coefficient based on the environmental data and the health data; inputs the environmental risk coefficient into an environmental risk prediction model to obtain an environmental risk coefficient prediction value; determines a risk warning level based on the environmental risk coefficient prediction value; and adjusts the environmental conditions based on the risk warning level.
[0114] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. An intelligent monitoring system for silkworm artificial feed breeding environment, comprising: The central processing module, and the data acquisition module and risk warning module connected thereto are characterized in that: The data acquisition module is used to collect environmental data of several positions in a breeding environment at several consecutive periods and the health status of silkworms at each position; the environmental status and the health status are preprocessed to obtain environmental data and health data; wherein the environmental data include the temperature, humidity, carbon dioxide concentration and light intensity of the breeding environment, and the health data include the food intake and body length of the silkworms; The central processing module is used to determine the installation location of relevant sensors based on the environmental data of several consecutive cycles at each location; wherein the relevant sensors include temperature sensors, humidity sensors, carbon dioxide concentration sensors and light intensity sensors; and, Calculate the environmental risk coefficient based on environmental data and health data; input the environmental risk coefficient into the environmental risk prediction model to obtain the predicted value of the environmental risk coefficient; wherein the environmental risk prediction model is obtained by training the artificial intelligence model; The risk warning module is used to determine the risk warning level based on the predicted value of the environmental risk coefficient; adjust the environmental conditions based on the risk warning level; wherein the risk warning level includes normal, level one warning and level two warning, and the risk severity is level one warning>level two warning>normal.
2. The intelligent monitoring system for artificial feeding environment of silkworms according to claim 1 is characterized in that: The preprocessing of the environmental state and the health state respectively comprises: A1: Extract the environmental status and health status of each location for several consecutive periods; A2: Eliminate abnormal data in environmental conditions and health status; A3: Fill in missing data on environmental and health conditions; A4: Mark the preprocessed environment state as environment data; A5: Mark the preprocessed health status as healthy data.
3. The intelligent monitoring system for silkworm artificial feed breeding environment according to claim 1, characterized in that: The step of determining the installation location of the relevant sensors based on the environmental data of a plurality of consecutive periods at each location includes: B1: Extract environmental data of several consecutive periods at each location; B2: Draw the corresponding environmental data change curve based on the environmental data of several consecutive periods at each location; B3: Deriving the environmental data change curve to obtain the environmental data derivative function; B4: Obtain the derivative value of the environmental data at each location and compare it with the corresponding derivative threshold; B5: If the derivative value of the environmental data is greater than the corresponding derivative threshold, the corresponding position is marked as the installation position; otherwise, the corresponding position is marked as the normal position.
4. The intelligent monitoring system for silkworm artificial feed rearing environment according to claim 3 is characterized in that: The step of drawing a corresponding environmental data change curve based on environmental data of a plurality of consecutive periods at each location includes: The environmental data of several contact periods are extracted, and the environmental data change curve is established with time as the independent variable and environmental data as the dependent variable.
5. The intelligent monitoring system for silkworm artificial feed rearing environment according to claim 1, characterized in that: The calculation of the environmental risk coefficient based on the environmental data and the health data includes: C1: Extract environmental data and health data; C2: Calculate the environmental score HP based on environmental data; C3: Calculate the health score KP based on health data; C4: The environmental risk factor HFX is calculated by the formula HFX = (a × HP) / [(b × KP) + c]; where a and b are influencing factors greater than 0; c is a constant to prevent division by zero, and the value range of c is (0,1].
6. The intelligent monitoring system for silkworm artificial feed rearing environment according to claim 5, characterized in that: The step of calculating the environmental score HP based on the environmental data includes: The temperature WD, humidity SD, carbon dioxide concentration TN and light intensity GZ in the environmental data are extracted, and the environmental score HP is calculated by the formula HP=k1×|WD-WD0|+k2×|SD-SD0|+k3×|TN-TN0|+k4×|GZ-GZ0|; among which WD0 is the temperature reference value, SD0 is the humidity reference value, TN0 is the carbon dioxide concentration reference value, GZ0 is the light intensity reference value, k1, k2, k3, k4 are proportional coefficients greater than 0, and || is the absolute value operator.
7. The intelligent monitoring system for silkworm artificial feed rearing environment according to claim 5, characterized in that: The step of calculating the health score KP based on the health data includes: The health data of food intake JSL and body length TC are extracted, and the health score KP is calculated by the formula KP=k5×JSL+k6×arctan(TC); wherein k5 and k6 are proportional coefficients greater than 0, and arctan() is the inverse tangent function.
8. The intelligent monitoring system for silkworm artificial feed rearing environment according to claim 1, characterized in that: The environmental risk prediction model is obtained through artificial intelligence model training, including: The environmental risk coefficients of several consecutive cycles are extracted and integrated into several groups of training data and test data; the artificial intelligence model is trained using the training data; the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted according to the test results; finally, a compressor failure prediction model is obtained, whose input is the environmental risk coefficients of the most recent several consecutive cycles and whose output is the environmental risk coefficient of the predicted cycle; wherein the artificial intelligence model includes a BP neural network model or an RBF neural network model.
9. The intelligent monitoring system for silkworm artificial feed rearing environment according to claim 1, characterized in that: The step of determining the risk warning level based on the predicted value of the environmental risk coefficient includes: T1: Extract the predicted value of environmental risk factor; T2: Set the risk range threshold [FY1, FY2]; the specific values of FY1 and FY2 are set according to the experience of the technicians, and FY1 <FY2; T3: Determine whether the predicted value of the environmental risk factor is within the risk range threshold [FY1, FY2]; if yes, mark the risk warning level as level 2 warning; if no, jump to T4; T4: Determine whether the predicted value of the environmental risk factor is greater than the maximum value FY2 in the risk range threshold; if yes, mark the risk warning level as level one warning; if not, mark the risk warning level as normal.
10. The intelligent monitoring system for silkworm artificial feed rearing environment according to claim 1, characterized in that: The adjustment of environmental conditions based on the risk warning level includes: Z1: Extract risk warning level; Z2: Determine whether the risk warning level is level 1 warning; if yes, adjust the environmental conditions of the breeding environment; if no, jump to Z3; Z3: Determine whether the risk warning level is level 2 warning; if yes, increase the frequency of collecting data on the environmental conditions of the breeding environment and the health status of the silkworms; if no, jump to Z1.
Citation Information
Cited By
Disease early warning method and system for silkworm breeding data
CN119851924A
Artificial feed silkworm breeding environment intelligent regulation and control method and system
CN121153660A
