Constant temperature control method and system for plateau greenhouse

Abstract

The invention provides a constant temperature control method and system for a plateau greenhouse, and relates to the technical field of temperature control, and the method comprises the steps: obtaining the temperature information of each monitoring point in the plateau greenhouse; wherein each monitoring point position is provided with at least two temperature sensors; de-noising processing is carried out on the temperature information collected by each monitoring point position to obtain de-noised temperature information; fusing the denoised temperature information of all the monitoring point locations to obtain fused temperature information of all the monitoring point locations; the optimal working temperature is obtained according to the historical data of the plateau greenhouse; constructing a temperature control model by using the optimal working temperature and the fused temperature information; and adjusting the temperature of the plateau greenhouse by using the temperature control model. The temperature can be adjusted in real time, intelligent management is achieved, energy consumption is effectively reduced, and high economic benefits and practical value are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of temperature control, and in particular to a constant temperature control method and system for a plateau greenhouse. Background Art

[0002] Due to its special geographical and climatic conditions, the plateau region faces problems such as drastic temperature changes and large temperature differences between day and night, which brings many challenges to agricultural production, especially animal husbandry. The construction of plateau greenhouses aims to provide yaks with a relatively stable growth environment to improve the yield and quality of yaks. However, traditional temperature control methods often cannot meet the needs of plateau areas. There are mainly the following background technical problems:

[0003] 1. The temperature difference between day and night in the plateau area can reach more than 20°C. The traditional temperature control system is difficult to respond to temperature changes quickly, which affects the growth of yaks.

[0004] 2. Many plateau greenhouses still use manual monitoring and temperature adjustment methods and lack intelligent temperature control systems, resulting in low management efficiency and difficulty in achieving refined management.

[0005] 3. Existing temperature control systems are often designed for plain areas and fail to fully consider the special climatic conditions of plateau areas, resulting in insufficient adaptability of the system in plateau environments.

[0006] Therefore, it is particularly important to develop an efficient, intelligent and automated constant temperature control method and system to meet the special needs of plateau areas. Summary of the invention

[0007] In order to overcome the deficiencies of the prior art, the purpose of the present invention is to provide a constant temperature control method and system for a plateau greenhouse, which can adjust the temperature in real time, realize intelligent management, effectively save energy consumption, and have high economic benefits and practical value.

[0008] To achieve the above object, the present invention provides the following solutions:

[0009] A constant temperature control method for a plateau greenhouse, comprising:

[0010] Acquire temperature information of each monitoring point set inside the plateau greenhouse; wherein each monitoring point has at least two temperature sensors;

[0011] De-noising the temperature information collected at each monitoring point to obtain the de-noised temperature information;

[0012] The denoised temperature information of each monitoring point is fused to obtain the fused temperature information of all detection points;

[0013] Obtaining the optimal working temperature according to the historical data of the plateau greenhouse;

[0014] Building a temperature control model using the optimal operating temperature and the fused temperature information;

[0015] The temperature control model is used to adjust the temperature of the plateau greenhouse.

[0016] Preferably, the temperature information collected at each monitoring point is denoised to obtain denoised temperature information, including:

[0017] Construct a connectivity model based on the spatial distance between each temperature sensor;

[0018] Constructing a temperature denoising function using the connection degree model;

[0019] The temperature denoising function is used to remove abnormal values ​​of the temperature information at each monitoring point to generate denoised temperature information.

[0020] Preferably, the formula of the connection degree model is:

[0021]

[0022] Wherein, σ represents the average value of the temperature difference between the sensor at point m and the sensor at point n in the historical period, dis(m,n) represents the spatial distance between the sensor at point m and the sensor at point n, and R represents the preset parameter.

[0023] Preferably, the use of the connection degree model to construct a temperature denoising function includes:

[0024] The denoising threshold is constructed using the weighted value of the temperature at each monitoring point; wherein the denoising threshold is: Among them, Y i represents the i-th temperature value collected by the temperature sensor, W i Represents Y i The weighted value of

[0025] A temperature denoising function is constructed based on the denoising threshold and the connection degree model; the temperature denoising function is: in, represents the temperature information of the corresponding monitoring point after denoising at time t, x(m,t) represents the measurement value of the sensor at point m at time t, x(n,t) represents the measurement value of the sensor at point n at time t, ρ represents the adjustable threshold, |N m | indicates the number of temperature sensors at the same monitoring point.

[0026] Preferably, the denoised temperature information of each monitoring point is fused to obtain the fused temperature information of all detection points, including:

[0027] The fusion function is constructed using the denoised temperature information of each monitoring point; the formula of the fusion function is: Among them, q ij represents the fusion function, X i represents the temperature information of the ith monitoring point after denoising, X j represents the temperature information of the jth monitoring point after denoising;

[0028] A fusion matrix is ​​constructed according to the fusion function; the formula of the fusion matrix is:

[0029]

[0030] Calculate the eigenvalues ​​corresponding to the fusion matrix;

[0031] Using the formula: Get the fusion weighted coefficient; where L represents the denoised wavelet coefficient, r i represents the i-th eigenvalue of the fusion matrix;

[0032] The fusion weighted coefficient is used to perform weighted averaging on the denoised temperature values ​​of all monitoring points to obtain the fused temperature information of all detection points.

[0033] Preferably, the temperature control model is:

[0034]

[0035] Among them, y(t) represents the temperature information after fusion of all detection points at time t, K p Represents the proportionality coefficient, y i (t) represents the optimal working temperature, T I Indicates the integration time, T D Indicates the derivative time.

[0036] Preferably, obtaining the optimal working temperature according to the historical data of the plateau greenhouse includes:

[0037] Collecting historical data of the plateau greenhouse; the historical data includes: historical temperature data and corresponding yak growth status data; the growth status data includes: weight change data, height change data and quality grade data;

[0038] Construct an initial convolutional neural network;

[0039] Respectively training the initial convolutional neural network according to the historical temperature data, the weight change data, the height change data and the quality grade data to obtain a trained first classifier, a second classifier and a third classifier;

[0040] The trained first classifier, the second classifier and the third classifier are cascaded to obtain a classification network;

[0041] After the classification network, an LSTM neural network and a fully connected layer are sequentially connected to obtain the working temperature prediction model;

[0042] The optimal operating temperature is obtained by using the operating temperature prediction model.

[0043] Preferably, the formula of the loss function of the working temperature prediction model is:

[0044]

[0045] in, is the predicted optimal operating temperature, y i is the actual optimal working temperature, w i is the growth indicator weight, n is the total number of data points, α is the weight hyperparameter of the MSE loss term, and β is the weight hyperparameter of the weighted loss term.

[0046] Preferably, the method for determining the growth indicator weight is:

[0047] Cleaning and feature extracting the growth status data to obtain a feature vector;

[0048] The Pearson correlation coefficient is used to evaluate the correlation between each feature vector and the preset target variable to obtain the weight of each feature vector;

[0049] The weight of each of the characteristic vectors is normalized, and the comprehensive weight of each data point is calculated to obtain the growth indicator weight.

[0050] A constant temperature control system for a plateau greenhouse, comprising:

[0051] The sensor setting module is used to obtain the temperature information of each monitoring point set inside the plateau greenhouse; wherein each monitoring point has at least two temperature sensors;

[0052] A denoising module is used to denoise the temperature information collected at each monitoring point to obtain denoised temperature information;

[0053] A fusion module is used to fuse the denoised temperature information of each monitoring point to obtain the fused temperature information of all detection points;

[0054] A temperature acquisition module, used for acquiring the optimal working temperature according to the historical data of the plateau greenhouse;

[0055] A model building module, used to build a temperature control model using the optimal operating temperature and the fused temperature information;

[0056] The regulating module is used to regulate the temperature of the plateau greenhouse using the temperature control model.

[0057] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0058] The present invention provides a constant temperature control method and system for a plateau greenhouse, the method comprising: obtaining temperature information of each monitoring point set inside the plateau greenhouse; wherein each monitoring point has at least two temperature sensors; performing denoising on the temperature information collected by each monitoring point to obtain denoised temperature information; fusing the denoised temperature information of each monitoring point to obtain fused temperature information of all detection points; obtaining the optimal working temperature according to the historical data of the plateau greenhouse; constructing a temperature control model using the optimal working temperature and the fused temperature information; and adjusting the temperature of the plateau greenhouse using the temperature control model. The present invention can adjust the temperature in real time, realize intelligent management, effectively save energy consumption, and has high economic benefits and practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] 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 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 labor.

[0060] Figure 1 A flow chart of a method provided by an embodiment of the present invention;

[0061] Figure 2 A schematic diagram of the system structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not 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.

[0063] The purpose of the present invention is to provide a constant temperature control method and system for a plateau greenhouse, which can adjust the temperature in real time, realize intelligent management, effectively save energy consumption, and have high economic benefits and practical value.

[0064] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0065] Figure 1 A flow chart of a method provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a constant temperature control method for a plateau greenhouse, comprising:

[0066] Step 100: Acquire temperature information of each monitoring point set inside the plateau greenhouse; wherein each monitoring point has at least two temperature sensors;

[0067] Step 200: performing denoising processing on the temperature information collected at each monitoring point to obtain denoised temperature information;

[0068] Step 300: Fusing the denoised temperature information of each monitoring point to obtain the fused temperature information of all detection points;

[0069] Step 400: Obtaining the optimal working temperature according to the historical data of the plateau greenhouse;

[0070] Step 500: constructing a temperature control model using the optimal operating temperature and the fused temperature information;

[0071] Step 600: Utilize the temperature control model to adjust the temperature of the plateau greenhouse.

[0072] Specifically, when setting up monitoring points inside the plateau greenhouse, multiple factors should be considered to ensure the accuracy and representativeness of temperature information. First, the monitoring points should be evenly distributed in various areas of the greenhouse to avoid temperature measurement deviations caused by local environmental differences. It is recommended to set up at least three monitoring points at different heights and different locations in the greenhouse (such as close to yaks, central areas, and edge areas) to capture vertical and horizontal changes in temperature. In addition, each monitoring point should be equipped with at least two temperature sensors to achieve redundant measurement and data comparison to ensure that reliable temperature information can still be obtained when a sensor fails. Sensors should be selected with fast response speed and high precision, and should be calibrated regularly to ensure data accuracy. Finally, the setting of monitoring points should take into account the convenience of subsequent data collection and maintenance to ensure that data reading and equipment maintenance can be conveniently performed.

[0073] Specifically, the sensor of this embodiment uses a DS18B20 digital temperature sensor, which has high accuracy (±0.5°C), can operate in the range of -55°C to +125°C, supports multi-point connection, and is suitable for use at multiple monitoring points.

[0074] Furthermore, when raising yaks in a greenhouse, the sensor setting implementation process is as follows:

[0075] 1. Conduct environmental assessment in the plateau greenhouse to determine the yak's activity area, rest area and drinking water area. According to the habits of yaks, select appropriate monitoring points to ensure that the areas where yaks often move are covered.

[0076] 2. Choose a suitable temperature sensor, such as DHT22 or BME280. These sensors can not only measure temperature, but also monitor humidity, which is suitable for the needs of plateau environment. Make sure the sensor has waterproof and dustproof functions to adapt to the environment in the greenhouse.

[0077] 3. Set up multiple monitoring points in the greenhouse. It is recommended to place sensors at the following locations:

[0078] Rest area: At least two sensors are set up in the resting area of ​​yaks, placed at different heights (such as the ground and 1 meter high) to monitor temperature changes in the area.

[0079] Activity area: Set up sensors in areas where yaks often move around to ensure that temperature changes when yaks are active can be captured.

[0080] Drinking water area: Set up sensors near the drinking water area to monitor the temperature in the area to ensure that the water source does not freeze due to low temperature.

[0081] 4. Install the sensor in an appropriate location, making sure it is away from direct heat sources (such as heating equipment) and cold sources (such as vents) to avoid local temperature interference. The sensor should be fixed on a stable bracket to avoid damage caused by the activities of yaks.

[0082] 5. Connect all sensors to the central data acquisition system and transmit data wirelessly or wiredly. Ensure that the system can receive and store temperature and humidity data of each monitoring point in real time.

[0083] 6. After installation, perform a preliminary test of the system to ensure that all sensors are working properly and can accurately collect temperature data. Regularly calibrate the sensors to ensure long-term data accuracy.

[0084] 7. Monitor temperature changes in real time through the data acquisition system, and adjust the temperature control strategy of the greenhouse based on the data analysis results to ensure that yaks grow and live within the optimal temperature range.

[0085] Through the above steps, the present embodiment can effectively set sensors to ensure accurate and reliable temperature monitoring of yaks in plateau greenhouses, thereby providing a good growth environment for them.

[0086] Preferably, the temperature information collected at each monitoring point is denoised to obtain denoised temperature information, including:

[0087] Construct a connectivity model based on the spatial distance between each temperature sensor;

[0088] Constructing a temperature denoising function using the connection degree model;

[0089] The temperature denoising function is used to remove abnormal values ​​of the temperature information at each monitoring point to generate denoised temperature information.

[0090] Preferably, the formula of the connection degree model is:

[0091]

[0092] Wherein, σ represents the average value of the temperature difference between the sensor at point m and the sensor at point n in the historical period, dis(m,n) represents the spatial distance between the sensor at point m and the sensor at point n, and R represents the preset parameter.

[0093] Specifically, in the present invention, the sensors at point m and point n are neighboring sensors. The closer the positions of the two sensors are, the greater the correlation between the two sensors. Therefore, the present invention measures the degree of connection between the two sensors based on the spatial distance between them. In practical applications, since the sensor may be affected by its own parameters or environmental factors, the measured value collected by the sensor at a certain moment may deviate greatly from the actual value. Therefore, the present invention introduces R to correct the problem of deviation in sensor correlation caused by the influence of the sensor's own parameters or environmental factors. In general, the value of R is 1.

[0094] Preferably, the use of the connection degree model to construct a temperature denoising function includes:

[0095] The denoising threshold is constructed using the weighted value of the temperature at each monitoring point; wherein the denoising threshold is: Among them, Y i represents the i-th temperature value collected by the temperature sensor, W i Represents Y i The weighted value of

[0096] A temperature denoising function is constructed based on the denoising threshold and the connection degree model; the temperature denoising function is: in, represents the temperature information of the corresponding monitoring point after denoising at time t, x(m,t) represents the measurement value of the sensor at point m at time t, x(n,t) represents the measurement value of the sensor at point n at time t, ρ represents the adjustable threshold, |N m | indicates the number of temperature sensors at the same monitoring point.

[0097] Specifically, the present invention uses a temperature denoising function to fuse the measurement values ​​of sensors at the same monitoring point, thereby avoiding a situation where a large deviation occurs between the measurement value collected by the sensor at a certain moment and the actual value, and making the measurement value of the sensor more realistic.

[0098] Preferably, the denoised temperature information of each monitoring point is fused to obtain the fused temperature information of all detection points, including:

[0099] The fusion function is constructed using the denoised temperature information of each monitoring point; the formula of the fusion function is: Among them, q ij represents the fusion function, X i represents the temperature information of the ith monitoring point after denoising, X j represents the temperature information of the jth monitoring point after denoising;

[0100] A fusion matrix is ​​constructed according to the fusion function; the formula of the fusion matrix is:

[0101]

[0102] Calculate the eigenvalues ​​corresponding to the fusion matrix;

[0103] Using the formula: Get the fusion weighted coefficient; where L represents the denoised wavelet coefficient, r i represents the i-th eigenvalue of the fusion matrix;

[0104] The fusion weighted coefficient is used to perform weighted averaging on the denoised temperature values ​​of all monitoring points to obtain the fused temperature information of all detection points.

[0105] The present invention is based on a fusion matrix and dynamically allocates a corresponding fusion weighting coefficient L to each sensor according to the temperature information measured by each sensor in real time, so as to obtain the final temperature information after fusion of all detection points, which can be closer to the actual temperature value of the entire high-temperature heating greenhouse.

[0106] Preferably, the temperature control model is:

[0107]

[0108] Among them, y(t) represents the temperature information after fusion of all detection points at time t, K p Represents the proportionality coefficient, y i (t) represents the optimal working temperature, T I Indicates the integration time, T D Indicates the derivative time.

[0109] Preferably, obtaining the optimal working temperature according to the historical data of the plateau greenhouse includes:

[0110] Collecting historical data of the plateau greenhouse; the historical data includes: historical temperature data and corresponding yak growth status data; the growth status data includes: weight change data, height change data and quality grade data;

[0111] Construct an initial convolutional neural network;

[0112] Respectively training the initial convolutional neural network according to the historical temperature data, the weight change data, the height change data and the quality grade data to obtain a trained first classifier, a second classifier and a third classifier;

[0113] The trained first classifier, the second classifier and the third classifier are cascaded to obtain a classification network;

[0114] After the classification network, an LSTM neural network and a fully connected layer are sequentially connected to obtain the working temperature prediction model;

[0115] The optimal operating temperature is obtained by using the operating temperature prediction model.

[0116] Optionally, the data input of this embodiment includes: historical temperature data and growth status data. Historical temperature data is suitable for CNN processing, especially performing well in time series. Growth status data includes weight change, height change and quality grade data, which can also be captured by CNN. Potential patterns and features.

[0117] In this embodiment, cascading multiple classifiers can integrate information of different features, but it may lead to information redundancy, and attention should be paid to the complexity and performance of the model. And LSTM is very effective in modeling long and short-term memory of time series data, which is suitable for analyzing the relationship between temperature data changes and growth conditions. The overall architecture of this embodiment, from CNN to LSTM, plus the fully connected layer, is a complex and powerful model combination, which can theoretically better capture the relationship between temperature and growth conditions.

[0118] Furthermore, in addition to LSTM, this embodiment may also consider variants such as BiLSTM, GRU, and more advanced structures such as transformer to improve the ability to extract time series features.

[0119] Through the above measures, this embodiment can improve the accuracy and robustness of the model, better predict the optimal operating temperature, and adapt to the unique climate and environmental requirements of plateau greenhouses.

[0120] Preferably, the formula of the loss function of the working temperature prediction model is:

[0121]

[0122] in, is the predicted optimal operating temperature, y i is the actual optimal working temperature, w i is the growth indicator weight, n is the total number of data points, α is the weight hyperparameter of the MSE loss term, and β is the weight hyperparameter of the weighted loss term.

[0123] Specifically, Defined as the predicted optimal operating temperature, this is the predicted value of the model output, representing the optimal temperature that the model believes under the given conditions. i Defined as the actual optimum operating temperature, this is the true target value that is compared with the predicted value to calculate the error. i Defined as the growth indicator weight, this is a weight factor that reflects the importance of each data point. It is usually calculated from growth indicators (such as growth rate, yield, quality, etc.) and ranges from 0 to 1. A higher weight means that the data point performs more significantly on the growth indicator. n is defined as the total number of data points, which is used to calculate the average value of the loss function. α is defined as the weight hyperparameter of the MSE loss term, which can control the influence of the mean square error (MSE) in the total loss. By adjusting α, the impact of MSE on model training can be increased or decreased. β is defined as the weight hyperparameter of the weighted loss term, which can control the influence of the growth indicator weighted loss term in the total loss. By adjusting β, the influence of the growth indicator on model training can be increased or decreased.

[0124] Preferably, the method for determining the growth indicator weight is:

[0125] Cleaning and feature extracting the growth status data to obtain a feature vector;

[0126] The Pearson correlation coefficient is used to evaluate the correlation between each feature vector and the preset target variable to obtain the weight of each feature vector;

[0127] The weight of each of the characteristic vectors is normalized, and the comprehensive weight of each data point is calculated to obtain the growth indicator weight.

[0128] Furthermore, the method for determining the weight of the growth index of the present embodiment is carried out by the following steps: First, collect a variety of index data related to yak growth, and ensure the integrity and accuracy of the data. Then, use statistical analysis methods (such as correlation analysis) to evaluate the relationship between each index and the target variable (such as final yield or quality), and identify the indicators that have a significant impact on yak growth. Then, use feature selection algorithms (such as random forests, LASSO regression, etc.) to score the importance of these indicators and quantify the contribution of each indicator to the target variable. Next, the weights of each indicator are normalized so that their sum is 1, so as to facilitate subsequent weighted calculations. Finally, combined with expert opinions or domain knowledge, the weights are appropriately adjusted to ensure their rationality and practicality. Through the above steps, the present embodiment can scientifically determine the weights of growth indicators and provide a basis for subsequent temperature control models and decisions.

[0129] Furthermore, to adjust the temperature of the plateau greenhouse using the temperature control model, it is first necessary to combine the previously collected denoised temperature information of each monitoring point with the optimal working temperature in the historical data to build an accurate control model. The model calculates the deviation between the current temperature and the optimal working temperature in real time by analyzing the information fed back by each temperature sensor. Using adaptive control algorithms, such as PID (proportional-integral-differential) control or fuzzy control, the system can dynamically adjust the operating state of the heating or cooling equipment inside the greenhouse according to the deviation value. For example, when the current temperature is lower than the optimal working temperature, the system will automatically adjust the power of the heating device to increase the temperature; conversely, when the temperature is too high, the ventilation volume will be increased or the cooling equipment will be started. At the same time, the temperature control model can also integrate data of other environmental factors such as humidity and light to optimize growth conditions and ensure that yaks live in an ideal climate environment. This embodiment not only achieves precise temperature control, but also improves the overall energy efficiency of the plateau greenhouse and the stability of yak growth, thereby improving economic benefits.

[0130] Corresponding to the above method, such as Figure 2 As shown, this embodiment provides a constant temperature control system for a plateau greenhouse, including:

[0131] The sensor setting module is used to obtain the temperature information of each monitoring point set inside the plateau greenhouse; wherein each monitoring point has at least two temperature sensors;

[0132] A denoising module is used to denoise the temperature information collected at each monitoring point to obtain denoised temperature information;

[0133] A fusion module is used to fuse the denoised temperature information of each monitoring point to obtain the fused temperature information of all detection points;

[0134] A temperature acquisition module, used for acquiring the optimal working temperature according to the historical data of the plateau greenhouse;

[0135] A model building module, used to build a temperature control model using the optimal operating temperature and the fused temperature information;

[0136] The regulating module is used to regulate the temperature of the plateau greenhouse using the temperature control model.

[0137] The beneficial effects of the present invention are as follows:

[0138] (1) The present invention can effectively eliminate external interference and sensor errors by setting at least two temperature sensors at each monitoring point and performing denoising on the collected temperature information, thereby improving the accuracy and reliability of temperature data.

[0139] (2) The present invention fuses the denoised temperature information of the monitoring points to comprehensively reflect the temperature distribution inside the plateau greenhouse, ensuring that representative temperature data can be obtained in the entire environment and avoiding decision-making errors caused by local temperature anomalies.

[0140] (3) The present invention combines historical data of plateau greenhouses to obtain the optimal working temperature, making the temperature control model more scientific and reasonable, thereby being able to provide a suitable growth environment under different climatic conditions and improving the efficiency and stability of yak growth.

[0141] (4) The present invention utilizes the constructed temperature control model to monitor and adjust the temperature of the greenhouse in real time, ensuring that the temperature is always maintained within the optimal range, avoiding the negative impact of extreme weather on yaks, and improving the stability and safety of production.

[0142] (5) The present invention can effectively reduce energy consumption, optimize resource utilization, reduce operating costs, and improve overall economic benefits through intelligent temperature control.

[0143] (6) The present invention provides a constant and suitable temperature environment for raising yaks, which helps to improve the health and growth performance of yaks and improve the breeding efficiency.

[0144] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0145] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A constant temperature control method for a plateau greenhouse, characterized in that: include: Acquire temperature information of each monitoring point set inside the plateau greenhouse; wherein each monitoring point has at least two temperature sensors; De-noising the temperature information collected at each monitoring point to obtain the de-noised temperature information; The denoised temperature information of each monitoring point is fused to obtain the fused temperature information of all detection points; Obtaining the optimal working temperature according to the historical data of the plateau greenhouse; Building a temperature control model using the optimal operating temperature and the fused temperature information; The temperature control model is used to adjust the temperature of the plateau greenhouse.

2. The constant temperature control method for a plateau greenhouse according to claim 1, characterized in that: The temperature information collected at each monitoring point is denoised to obtain denoised temperature information, including: Construct a connectivity model based on the spatial distance between each temperature sensor; Constructing a temperature denoising function using the connection degree model; The temperature denoising function is used to remove abnormal values ​​of the temperature information at each monitoring point to generate denoised temperature information.

3. The constant temperature control method for a plateau greenhouse according to claim 2, characterized in that: The formula of the connection degree model is: Wherein, σ represents the average value of the temperature difference between the sensor at point m and the sensor at point n in the historical period, dis(m,n) represents the spatial distance between the sensor at point m and the sensor at point n, and R represents the preset parameter.

4. The constant temperature control method for a plateau greenhouse according to claim 3, characterized in that: The method of constructing a temperature denoising function using the connection degree model includes: The denoising threshold is constructed using the weighted value of the temperature at each monitoring point; wherein the denoising threshold is: Among them, Y i represents the i-th temperature value collected by the temperature sensor, W i Represents Y i The weighted value of A temperature denoising function is constructed based on the denoising threshold and the connection degree model; the temperature denoising function is: in, represents the temperature information of the corresponding monitoring point after denoising at time t, x(m,t) represents the measurement value of the sensor at point m at time t, x(n,t) represents the measurement value of the sensor at point n at time t, ρ represents the adjustable threshold, |N m | indicates the number of temperature sensors at the same monitoring point.

5. The constant temperature control method for a plateau greenhouse according to claim 4, characterized in that: The denoised temperature information of each monitoring point is fused to obtain the fused temperature information of all detection points, including: The fusion function is constructed using the denoised temperature information of each monitoring point; the formula of the fusion function is: Among them, q ij represents the fusion function, X i represents the temperature information of the ith monitoring point after denoising, X j represents the temperature information of the jth monitoring point after denoising; A fusion matrix is ​​constructed according to the fusion function; the formula of the fusion matrix is: Calculate the eigenvalues ​​corresponding to the fusion matrix; Using the formula: Get the fusion weighted coefficient; where L represents the denoised wavelet coefficient, r i represents the i-th eigenvalue of the fusion matrix; The fusion weighted coefficient is used to perform weighted averaging on the denoised temperature values ​​of all monitoring points to obtain the fused temperature information of all detection points.

6. The constant temperature control method for a plateau greenhouse according to claim 5, characterized in that: The temperature control model is: Among them, y(t) represents the temperature information after fusion of all detection points at time t, K p Represents the proportionality coefficient, y i (t) represents the optimal working temperature, T I Indicates the integration time, T D Indicates the derivative time.

7. The constant temperature control method for a plateau greenhouse according to claim 1, characterized in that: The optimal working temperature is obtained based on the historical data of the plateau greenhouse, including: Collecting historical data of the plateau greenhouse; the historical data includes: historical temperature data and corresponding yak growth status data; the growth status data includes: weight change data, height change data and quality grade data; Construct an initial convolutional neural network; Respectively training the initial convolutional neural network according to the historical temperature data, the weight change data, the height change data and the quality grade data to obtain a trained first classifier, a second classifier and a third classifier; The trained first classifier, the second classifier and the third classifier are cascaded to obtain a classification network; After the classification network, an LSTM neural network and a fully connected layer are sequentially connected to obtain the working temperature prediction model; The optimal operating temperature is obtained by using the operating temperature prediction model.

8. The constant temperature control method for a plateau greenhouse according to claim 7, characterized in that: The formula of the loss function of the working temperature prediction model is: in, is the predicted optimal operating temperature, y i is the actual optimal working temperature, w i is the growth indicator weight, n is the total number of data points, α is the weight hyperparameter of the MSE loss term, and β is the weight hyperparameter of the weighted loss term.

9. The constant temperature control method for a plateau greenhouse according to claim 8, characterized in that: The method for determining the growth indicator weight is: Cleaning and feature extracting the growth status data to obtain a feature vector; The Pearson correlation coefficient is used to evaluate the correlation between each feature vector and the preset target variable to obtain the weight of each feature vector; The weight of each of the characteristic vectors is normalized, and the comprehensive weight of each data point is calculated to obtain the growth indicator weight.

10. A constant temperature control system for a plateau greenhouse, characterized in that: include: The sensor setting module is used to obtain the temperature information of each monitoring point set inside the plateau greenhouse; wherein each monitoring point has at least two temperature sensors; A denoising module is used to denoise the temperature information collected at each monitoring point to obtain denoised temperature information; A fusion module is used to fuse the denoised temperature information of each monitoring point to obtain the fused temperature information of all detection points; A temperature acquisition module, used for acquiring the optimal working temperature according to the historical data of the plateau greenhouse; A model building module, used to build a temperature control model using the optimal operating temperature and the fused temperature information; The regulating module is used to regulate the temperature of the plateau greenhouse using the temperature control model.

Images