Intelligent control system and method of hatching cabinet

Abstract

The invention belongs to the technical field of silkworm hatching cabinet control, and particularly relates to an intelligent control system and method of a hatching cabinet, the temperature and humidity of the hatching cabinet are adjusted and controlled through a single parameter, any two parameters and three parameters to obtain parameter change data, a parameter prediction model predicts the parameter data in the hatching process, and the parameter prediction model predicts the parameter data in the hatching process. The method comprises the following steps: acquiring and adjusting parameter adjustment time nodes in a hatching process, performing feedback model optimization by acquiring silkworm image data and parameter data in real time, adjusting the parameter adjustment time nodes based on an optimized parameter prediction model, and adjusting parameters at the adjusted parameter adjustment time nodes. The technical problems of hysteresis of parameter control of the hatching cabinet and low precision of parameter control in the prior art are effectively solved, and the control precision of the hatching cabinet is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of server heat dissipation, and in particular relates to an intelligent control system and method for a green incubation cabinet. Background Art

[0002] Accelerating silkworm eggs is a key step in silkworm breeding. Temperature, humidity, and light are crucial for the hatching and growth of silkworms. Controlling these levels is crucial, and as the silkworms grow, their requirements change. Therefore, the temperature, humidity, and light levels during this process must be adjusted accordingly.

[0003] A brooding cabinet is a device used to accelerate the hatching of silkworm eggs. It primarily controls temperature, humidity, and light conditions during the silkworm rearing process to ensure successful hatching. These cabinets typically maintain a constant temperature and humidity, and electronically control these conditions to ensure optimal conditions for the development of the eggs.

[0004] The control process of the control system of the existing incubation cabinet is: real-time collection of temperature, humidity and light data during the growth process of silkworms, and then comparing the real-time temperature value, humidity value and light value with the standard temperature value range, humidity value range and light value range respectively, to determine whether the real-time temperature value, humidity value and light value are within the standard temperature value range, humidity value range and light value range; when one or more parameters are not within the standard temperature value range, humidity value range and light value range, the parameters that are not within the standard range are adjusted.

[0005] In fact, there are control problems with the temperature, humidity, and light control systems of existing green incubation cabinets. On the one hand, when the temperature, humidity, or light are not within the corresponding standard range, the temperature or humidity needs to be adjusted by the control system in the green incubation cabinet. However, the temperature or humidity cannot be adjusted in real time, that is, there is a certain lag in the temperature or humidity adjustment, so the temperature and humidity adjustment accuracy is low. On the other hand, the adjustment of temperature, humidity, or light affects each other, and the adjustment of existing green incubation cabinets does not take into account the mutual influence factors in the temperature, humidity, or light adjustment process, which further reduces the control accuracy of the green incubation cabinet.

[0006] Therefore, how to improve the control process of the existing green incubator to reduce the impact of the hysteresis in the temperature and humidity control process of the existing green incubator on the control and the mutual influence between the temperature, humidity or light adjustment, and improve the accuracy of temperature, humidity or light control, is a technical problem that needs to be solved urgently. Summary of the Invention

[0007] The purpose of the present invention is to provide an intelligent control system and method for a green incubation cabinet, so as to improve the control process of the existing green incubation cabinet, so as to reduce the influence of the hysteresis of the temperature and humidity control process of the existing green incubation cabinet on the control process and the mutual influence between the temperature, humidity or light adjustment, and improve the accuracy of temperature, humidity or light control.

[0008] In order to solve the above technical problems, the technical solutions adopted by the present invention are as follows:

[0009] In a first aspect, a method for intelligently controlling a greening cabinet is provided, comprising the following steps:

[0010] S1: separately adjust and control the temperature and humidity of the green incubator, and obtain the parameter change data of each parameter during the single parameter control process of the green incubator;

[0011] S2: Adjust and control any two parameters among temperature, humidity, and light, and obtain the change data of each parameter during the control of any two parameters; simultaneously control the three parameters of temperature, humidity, and light, and obtain the change data of each parameter during the simultaneous control of the three parameters;

[0012] S3: Establish a parameter prediction model and input the parameter change data under various controls into the parameter prediction model to train the model;

[0013] S4: Predicting the parameter data of each parameter in the incubation process using the trained parameter prediction model, obtaining the adjustment time nodes of each parameter in the incubation process, and adjusting each parameter at each parameter adjustment time node;

[0014] S5: Acquire silkworm image data and parameter data in real time, identify the silkworm hatching stage based on the silkworm image data, obtain the standard range of each parameter corresponding to the silkworm hatching stage, obtain the difference between the real-time data of each parameter and the corresponding standard range, and feed it back to the parameter prediction model for model optimization;

[0015] S6: Adjust each parameter adjustment time node based on the optimized parameter prediction model, and adjust each parameter at each adjusted parameter adjustment time node.

[0016] Preferably, the specific process in step S1 is as follows:

[0017] S11: Regulating and controlling the temperature of the incubation cabinet, including presetting several target temperatures, and performing temperature regulation at multiple starting temperatures, and obtaining real-time temperature change data and time data for each starting temperature reaching each target temperature;

[0018] S12: Adjust and control the humidity of the green incubator, including presetting several target humidity levels, adjusting the humidity at multiple starting humidity levels, and obtaining real-time humidity change data and time data from each starting humidity level to each target humidity level.

[0019] Preferably, the order of executing step S11 and step S12 can be exchanged.

[0020] Preferably, the specific process in step S2 is as follows:

[0021] S21: Setting several target temperatures, performing temperature adjustment at multiple starting temperatures, and obtaining temperature change data, humidity change data, and time data for each starting temperature reaching each target temperature in real time;

[0022] S22: Setting a number of target humidities, performing humidity adjustment at a number of starting humidities, and obtaining in real time humidity change data, temperature change data, and time data from each starting humidity to each target humidity stage;

[0023] S23: Setting a number of target lightings, adjusting the lighting in the green incubator to each target lighting, and collecting temperature and humidity change data of the green incubator over time under each lighting.

[0024] Preferably, the parameter prediction model in step S3 is a long-short time neural network model, comprising an input gate, a forget gate, an update gate and an output gate, wherein the input gate is connected to the forget gate and the update gate respectively, and the output gate is connected to the forget gate and the update gate;

[0025] The forget gate is used to output a value from 0 to 1 to represent the discarded proportion of the input data at the current time node through the input data at the current time node and the hidden state at the previous time node, thereby determining the discarded data in the data input by the input gate;

[0026] The update gate is used to multiply the output value of the forget gate with the output value of the previous time node, and add the input gate value and the current time step to obtain the unupdated value multiplied as a component of the input of the next time node.

[0027] Preferably, in step S3, the specific process of inputting the parameter change data under various controls into the parameter prediction model to train the model is as follows:

[0028] S31: converting the parameter change data into a data sequence, inputting the data sequence through the input gate, and passing through a forget gate and an update gate to obtain an output value, and outputting the output value through the output gate;

[0029] S32: Compare the output value with the actual label value and perform calculation based on a preset loss function;

[0030] S33: Derivative the parameters of the parameter prediction model according to the calculation result of the loss function to obtain the influence degree of the specified parameters of the model on the loss function;

[0031] S34: Based on the degree of influence of each parameter on the loss function, backpropagation is performed, and the gradient is returned to each gated unit. The specified parameters of the model are adjusted according to the specified step size through the gradient descent method to minimize the loss function.

[0032] Preferably, the specific process of predicting each parameter data, obtaining each parameter adjustment time node and adjusting it in step S4 is as follows:

[0033] S41: The parameter prediction model predicts the parameter data during the incubation process, obtains the time period corresponding to each hatching stage of the normal development of the silkworm, and obtains the parameter prediction data corresponding to the time period of each hatching stage;

[0034] S42: comparing the predicted data of each parameter with the standard range of each parameter in each incubation stage, and obtaining a time node at which the predicted parameter data is not within the corresponding standard range of each parameter;

[0035] S43: reversely inferring the adjustment time points of each parameter in the greening process based on the time points at which the predicted parameter data is not within the corresponding parameter standard ranges;

[0036] S44: regulating and controlling the parameters of the green incubation cabinet at the parameter adjustment time nodes.

[0037] In a second aspect, an intelligent control system for a green incubation cabinet is provided, which is used to implement the intelligent control method of the green incubation cabinet, including a data acquisition module, an adjustment control module, a parameter prediction model, a feedback module and a parameter optimization module, wherein the data acquisition module is connected to the adjustment control module and the parameter prediction model, and the parameter prediction model is connected to the feedback module and the parameter optimization module;

[0038] The regulating and controlling module is used to independently regulate and control the temperature, humidity and light of the green incubation cabinet;

[0039] The data acquisition module is used to collect parameter data of the green incubator during the control process;

[0040] The parameter prediction model is used to predict the parameter data of the green tea incubation process after training the model by inputting the parameter change data under various controls, and obtain the adjustment time nodes of each parameter in the green tea incubation process;

[0041] The feedback module is used to obtain silkworm image data and parameter data in real time, identify the silkworm hatching stage based on the silkworm image data, obtain the standard range of each parameter corresponding to the silkworm hatching stage, obtain the difference between the real-time data of each parameter and the corresponding standard range, and feed it back to the parameter prediction model;

[0042] The parameter optimization module is used to optimize the parameters of the model based on the data fed back by the model.

[0043] The beneficial effects of the present invention include:

[0044] The intelligent control system and method of the green incubation cabinet provided by the present invention adjust and control the temperature and humidity of the green incubation cabinet through a single parameter, any two parameters and three parameters to obtain the change data of each parameter. The parameter prediction model predicts the data of each parameter in the green incubation process, obtains the adjustment time nodes of each parameter in the green incubation process and adjusts them. The feedback model is optimized by obtaining mulberry silkworm image data and each parameter data in real time. The adjustment time nodes of each parameter are adjusted based on the optimized parameter prediction model. The parameters are adjusted at the adjusted parameter adjustment time nodes, which effectively solves the technical problems of the hysteresis of the parameter control of the green incubation cabinet and the low precision of the control of each parameter in the prior art, and improves the control precision of the green incubation cabinet. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of the intelligent control method of the green incubation cabinet of the present invention.

[0046] Figure 2 Schematic diagram of the training process of the parameter prediction model of the present invention.

[0047] Figure 3 Schematic diagram of the architecture of the intelligent control system of the greening cabinet of the present invention. DETAILED DESCRIPTION

[0048] The following is combined with Figures 1 to 3 The present invention is described in further detail:

[0049] Example 1

[0050] See attached Figure 1 As shown, an intelligent control method for a greening cabinet includes the following steps:

[0051] S1: separately adjust and control the temperature and humidity of the green incubator, and obtain the parameter change data of each parameter during the single parameter control process of the green incubator;

[0052] S2: Adjust and control any two parameters among temperature, humidity, and light, and obtain the change data of each parameter during the control of any two parameters; simultaneously control the three parameters of temperature, humidity, and light, and obtain the change data of each parameter during the simultaneous control of the three parameters;

[0053] S3: Establish a parameter prediction model and input the parameter change data under various controls into the parameter prediction model to train the model;

[0054] S4: Predicting the parameter data of each parameter in the incubation process using the trained parameter prediction model, obtaining the adjustment time nodes of each parameter in the incubation process, and adjusting each parameter at each parameter adjustment time node;

[0055] S5: Acquire silkworm image data and parameter data in real time, identify the silkworm hatching stage based on the silkworm image data, obtain the standard range of each parameter corresponding to the silkworm hatching stage, obtain the difference between the real-time data of each parameter and the corresponding standard range, and feed it back to the parameter prediction model for model optimization;

[0056] S6: Adjust each parameter adjustment time node based on the optimized parameter prediction model, and adjust each parameter at each adjusted parameter adjustment time node.

[0057] Because the control system of the existing green tea incubator is in the process of controlling the temperature, humidity and light, when the temperature, humidity or light is not within the corresponding standard range, the temperature or humidity needs to be adjusted by the control system in the green tea incubator. However, the adjustment of temperature or humidity cannot be realized in real time, that is, there is a certain lag in the adjustment of temperature or humidity. Therefore, when the temperature and humidity of the green tea incubator are not within the corresponding standard range at a certain time point, it cannot be immediately adjusted to be within the standard range. It can only be within the standard range after a certain period of adjustment. In addition, the adjustment of temperature, humidity or light affects each other. For example, when the temperature in the green tea incubator is adjusted, its humidity will change dynamically with the change of temperature, that is, when the temperature changes, the humidity will also change accordingly. For another example, when adjusting the light in the green tea incubator, different light intensities will also cause the temperature in the green tea incubator to change. Therefore, when adjusting the various parameters in the green tea incubator, the various parameters affect each other.

[0058] Therefore, the present invention adjusts and controls the temperature and humidity of the green incubation cabinet by using a single parameter, any two parameters, and three parameters to obtain the change data of each parameter, trains the parameter prediction model based on the obtained change data of each parameter, and captures the change of parameters over time during the parameter adjustment process, as well as the impact of one parameter change on another parameter during the parameter adjustment process through the parameter prediction model.

[0059] Then, the parameter prediction model is used to predict the parameter data of each parameter in the greening process. Through the prediction results of each parameter data, the time nodes when each parameter is not within the corresponding range are obtained. The temperature and humidity of the greening cabinet are adjusted and controlled by combining single parameters, any two parameters, and three parameters to obtain the parameter change data, and the actual time nodes of temperature or humidity adjustment are reversed, thereby realizing early adjustment of temperature or humidity, so that the temperature or humidity can continue to be within the standard parameter range.

[0060] By adjusting the time nodes of various parameters in the incubation process, real-time acquisition of silkworm image data and various parameter data is performed to optimize the feedback model, and the adjustment time nodes of various parameters are adjusted based on the optimized parameter prediction model. At the adjusted parameter adjustment time nodes, various parameters are adjusted, which effectively solves the technical problems of the lag of parameter control of the incubation cabinet and the low precision of the control of various parameters in the existing technology, and improves the control precision of the incubation cabinet.

[0061] Example 2

[0062] Based on Example 1, the specific process in step S1 is as follows:

[0063] S11: Regulating and controlling the temperature of the incubation cabinet, including presetting several target temperatures, and performing temperature regulation at multiple starting temperatures, and obtaining real-time temperature change data and time data for each starting temperature reaching each target temperature;

[0064] S12: Adjust and control the humidity of the green incubator, including presetting several target humidity levels, adjusting the humidity at multiple starting humidity levels, and obtaining real-time humidity change data and time data from each starting humidity level to each target humidity level.

[0065] The order of executing step S11 and step S12 can be exchanged.

[0066] In this embodiment, the specific process in step S2 is as follows:

[0067] S21: Setting several target temperatures, performing temperature adjustment at multiple starting temperatures, and obtaining temperature change data, humidity change data, and time data for each starting temperature reaching each target temperature in real time;

[0068] S22: Setting a number of target humidities, performing humidity adjustment at a number of starting humidities, and obtaining in real time humidity change data, temperature change data, and time data from each starting humidity to each target humidity stage;

[0069] S23: Setting a number of target lightings, adjusting the lighting in the green incubator to each target lighting, and collecting temperature and humidity change data of the green incubator over time under each lighting.

[0070] Example 3

[0071] Based on Example 1 or Example 2, the parameter prediction model in step S3 is a long-short time neural network model, including an input gate, a forget gate, an update gate, and an output gate, wherein the input gate is connected to the forget gate and the update gate, respectively, and the output gate is connected to the forget gate and the update gate. The forget gate is used to output a value from 0 to 1 representing the proportion of the input data at the current time node that is discarded based on the input data at the current time node and the hidden state of the previous time node, thereby determining the discarded data in the data input by the input gate. The update gate is used to multiply the output value of the forget gate with the output value of the previous time node, and add the input gate value and the current time step to obtain the unupdated value multiplied as a component of the input of the next time node.

[0072] See also Figure 2 In step S3, the specific process of inputting the parameter change data under various controls into the parameter prediction model to train the model is as follows:

[0073] S31: converting the parameter change data into a data sequence, inputting the data sequence through the input gate, and passing through a forget gate and an update gate to obtain an output value, and outputting the output value through the output gate;

[0074] S32: Compare the output value with the actual label value and perform calculation based on a preset loss function;

[0075] S33: Derivative the parameters of the parameter prediction model according to the calculation result of the loss function to obtain the influence degree of the specified parameters of the model on the loss function;

[0076] S34: Based on the degree of influence of each parameter on the loss function, backpropagation is performed, and the gradient is returned to each gated unit. The specified parameters of the model are adjusted according to the specified step size through the gradient descent method to minimize the loss function.

[0077] In this embodiment, the specific process of predicting each parameter data, obtaining each parameter adjustment time node and adjusting it in step S4 is as follows:

[0078] S41: The parameter prediction model predicts the parameter data during the incubation process, obtains the time period corresponding to each hatching stage of the normal development of the silkworm, and obtains the parameter prediction data corresponding to the time period of each hatching stage;

[0079] S42: comparing the predicted data of each parameter with the standard range of each parameter in each incubation stage, and obtaining a time node at which the predicted parameter data is not within the corresponding standard range of each parameter;

[0080] S43: reversely inferring the adjustment time points of each parameter in the greening process based on the time points at which the predicted parameter data is not within the corresponding parameter standard ranges;

[0081] S44: regulating and controlling the parameters of the green incubation cabinet at the parameter adjustment time nodes.

[0082] An intelligent control system for a green tea cabinet, used to implement the intelligent control method of a green tea cabinet, see Figure 3 , including a data acquisition module, an adjustment and control module, a parameter prediction model, a feedback module and a parameter optimization module. The data acquisition module is connected to the adjustment and control module and the parameter prediction model, and the parameter prediction model is connected to the feedback module and the parameter optimization module. The adjustment and control module is used to independently adjust and control the temperature, humidity and light of the green incubator. The data acquisition module is used to collect parameter data of the green incubator during the control process. The parameter prediction model is used to train the model by inputting parameter change data under various controls into the parameter prediction model, and then predict the parameter data of each parameter in the green incubation process, and obtain the adjustment time node of each parameter in the green incubation process. The feedback module is used to obtain mulberry silkworm image data and each parameter data in real time, identify the mulberry silkworm hatching stage based on the mulberry silkworm image data, obtain the standard range of each parameter corresponding to the mulberry silkworm hatching stage, obtain the difference between the real-time data of each parameter and the corresponding standard range, and feed it back to the parameter prediction model. The parameter optimization module is used to optimize the parameters of the model based on the data fed back by the feedback model.

[0083] In summary, the intelligent control system and method of the green incubation cabinet provided by the present invention adjusts and controls the temperature and humidity of the green incubation cabinet through a single parameter, any two parameters, and three parameters to obtain the change data of each parameter. The parameter prediction model predicts the data of each parameter in the green incubation process, obtains the adjustment time nodes of each parameter in the green incubation process and adjusts them. The feedback model is optimized by obtaining the mulberry silkworm image data and each parameter data in real time. The adjustment time nodes of each parameter are adjusted based on the optimized parameter prediction model. The parameters are adjusted at the adjusted parameter adjustment time nodes, which effectively solves the technical problems of the hysteresis of the parameter control of the green incubation cabinet and the low accuracy of the control of each parameter in the prior art, and improves the control accuracy of the green incubation cabinet.

Claims

1. An intelligent control method for a greening cabinet, characterized in that: The following steps are involved: S1: separately adjust and control the temperature and humidity of the green incubator, and obtain the parameter change data of each parameter during the single parameter control process of the green incubator; S2: Adjust and control any two parameters among temperature, humidity, and light, and obtain the change data of each parameter during the control of any two parameters; simultaneously control the three parameters of temperature, humidity, and light, and obtain the change data of each parameter during the simultaneous control of the three parameters; S3: Establish a parameter prediction model and input the parameter change data under various controls into the parameter prediction model to train the model; S4: Predicting the parameter data of each parameter in the incubation process using the trained parameter prediction model, obtaining the adjustment time nodes of each parameter in the incubation process, and adjusting each parameter at each parameter adjustment time node; S5: Acquire silkworm image data and parameter data in real time, identify the silkworm hatching stage based on the silkworm image data, obtain the standard range of each parameter corresponding to the silkworm hatching stage, obtain the difference between the real-time data of each parameter and the corresponding standard range, and feed it back to the parameter prediction model for model optimization; S6: Adjust each parameter adjustment time node based on the optimized parameter prediction model, and adjust each parameter at each adjusted parameter adjustment time node.

2. The intelligent control method of a green incubation cabinet according to claim 1, characterized in that: The specific process in step S1 is as follows: S11: Regulating and controlling the temperature of the incubation cabinet, including presetting several target temperatures, and performing temperature regulation at multiple starting temperatures, and obtaining real-time temperature change data and time data for each starting temperature reaching each target temperature; S12: Adjust and control the humidity of the green incubator, including presetting several target humidity levels, adjusting the humidity at multiple starting humidity levels, and obtaining real-time humidity change data and time data from each starting humidity level to each target humidity level.

3. The intelligent control method of a green incubation cabinet according to claim 2, characterized in that: The order of executing step S11 and step S12 can be exchanged.

4. The intelligent control method of a green incubation cabinet according to claim 1, characterized in that: The specific process in step S2 is as follows: S21: Setting several target temperatures, performing temperature adjustment at multiple starting temperatures, and obtaining temperature change data, humidity change data, and time data for each starting temperature reaching each target temperature in real time; S22: Setting a number of target humidities, performing humidity adjustment at a number of starting humidities, and obtaining in real time humidity change data, temperature change data, and time data from each starting humidity to each target humidity stage; S23: Setting a number of target lightings, adjusting the lighting in the green incubator to each target lighting, and collecting temperature and humidity change data of the green incubator over time under each lighting.

5. The intelligent control method of a green incubation cabinet according to claim 1, characterized in that: The parameter prediction model in step S3 is a long-short time neural network model, including an input gate, a forget gate, an update gate and an output gate, wherein the input gate is connected to the forget gate and the update gate respectively, and the output gate is connected to the forget gate and the update gate; The forget gate is used to output a value from 0 to 1 to represent the discarded proportion of the input data at the current time node through the input data at the current time node and the hidden state at the previous time node, thereby determining the discarded data in the data input by the input gate; The update gate is used to multiply the output value of the forget gate with the output value of the previous time node, and add the input gate value and the current time step to obtain the unupdated value multiplied as a component of the input of the next time node.

6. The intelligent control method of a green incubation cabinet according to claim 5, characterized in that: The specific process of inputting the parameter change data under various controls into the parameter prediction model to train the model in step S3 is as follows: S31: converting the parameter change data into a data sequence, inputting the data sequence through the input gate, and passing through a forget gate and an update gate to obtain an output value, and outputting the output value through the output gate; S32: Compare the output value with the actual label value and perform calculation based on a preset loss function; S33: Derivative the parameters of the parameter prediction model according to the calculation result of the loss function to obtain the influence degree of the specified parameters of the model on the loss function; S34: Based on the degree of influence of each parameter on the loss function, backpropagation is performed, and the gradient is returned to each gated unit. The specified parameters of the model are adjusted according to the specified step size through the gradient descent method to minimize the loss function.

7. The intelligent control method of a green incubation cabinet according to claim 1, characterized in that: The specific process of predicting each parameter data, obtaining each parameter adjustment time node and adjusting it in step S4 is as follows: S41: The parameter prediction model predicts the parameter data during the incubation process, obtains the time period corresponding to each hatching stage of the normal development of the silkworm, and obtains the parameter prediction data corresponding to the time period of each hatching stage; S42: comparing the predicted data of each parameter with the standard range of each parameter in each incubation stage, and obtaining a time node at which the predicted parameter data is not within the corresponding standard range of each parameter; S43: reversely inferring the adjustment time points of each parameter in the greening process based on the time points at which the predicted parameter data is not within the corresponding parameter standard ranges; S44: regulating and controlling the parameters of the green incubation cabinet at the parameter adjustment time nodes.

8. An intelligent control system for a green tea incubator, used to implement the intelligent control method for a green tea incubator according to any one of claims 1 to 7, characterized in that: It includes a data acquisition module, an adjustment and control module, a parameter prediction model, a feedback module and a parameter optimization module, wherein the data acquisition module is connected to the adjustment and control module and the parameter prediction model, and the parameter prediction model is connected to the feedback module and the parameter optimization module; The regulating and controlling module is used to independently regulate and control the temperature, humidity and light of the green incubation cabinet; The data acquisition module is used to collect parameter data of the green incubator during the control process; The parameter prediction model is used to predict the parameter data of the green tea incubation process after training the model by inputting the parameter change data under various controls, and obtain the adjustment time nodes of each parameter in the green tea incubation process; The feedback module is used to obtain silkworm image data and parameter data in real time, identify the silkworm hatching stage based on the silkworm image data, obtain the standard range of each parameter corresponding to the silkworm hatching stage, obtain the difference between the real-time data of each parameter and the corresponding standard range, and feed it back to the parameter prediction model; The parameter optimization module is used to optimize the parameters of the model based on the data fed back by the model.

Images