Intelligent granary based ventilation and energy consumption linkage adaptive control method

Abstract

The application discloses a kind of ventilation and energy consumption linkage self-adaptive control method based on intelligent granary, it is related to the field of intelligent granary, including: establishing multidimensional data acquisition matrix, real-time collection of temperature and humidity data of different depth of grain pile in flat warehouse, environmental temperature and humidity data outside warehouse and real-time running power data of ventilation equipment in warehouse;The multidimensional data collected are denoised and normalized, a coupling correlation model of temperature and humidity change and ventilation equipment energy consumption is constructed, and the model is based on the model to divide and set energy consumption grade interval;The application combines denoising and normalization processing to reduce data error, accurately establishes the correlation between temperature and humidity change and energy consumption through coupling correlation model, divides the grade interval that meets the upper limit of energy consumption and storage grain safety, and generates ventilation equipment speed and running time strategy based on double-target control threshold automatic matching energy consumption grade.

Description

Technical Field

[0001] This invention relates to the field of intelligent grain storage technology, specifically to an adaptive control method for ventilation and energy consumption linkage based on intelligent grain storage. Background Technology

[0002] Intelligent flat-roofed warehouses are the mainstream facilities for modern grain storage, with a flat-roof structure adapted to the needs of large-scale grain storage. They integrate systems for real-time temperature and humidity monitoring, automatic ventilation and temperature control, and intelligent grain condition early warning, enabling precise regulation of the internal environment and effectively inhibiting grain mold and insect infestation. Compared to traditional grain warehouses, they significantly reduce grain storage losses, decrease labor costs, and ensure grain quality through data-driven management, providing efficient technical support for safe grain storage.

[0003] The invention patent application with application number 202511225563.5 discloses a decision-making method for intelligent grain warehouse ventilation and energy consumption optimization based on reinforcement learning. The method aims to solve the problem that "at present, when dealing with complex grain storage environments, many automated systems in grain warehouses mainly rely on preset fixed threshold logic. Ventilation is only passively triggered when the temperature of a certain local point exceeds the standard. This control method is inherently lagging and passive, unable to predict the trend of grain condition changes, often missing the best intervention opportunity, and ignoring the complex spatial heat conduction relationship between different areas inside the grain pile, making it difficult to achieve global collaborative optimization."

[0004] However, the existing smart grain warehouses lack a close linkage between the energy consumption controls of their ventilation machines, which means that most smart grain warehouses can only have an advantage in either smart ventilation or energy saving.

[0005] To address this, we propose an adaptive control method for ventilation and energy consumption linkage based on intelligent grain storage. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a ventilation and energy consumption linkage adaptive control method based on intelligent grain warehouse, which can effectively solve the problems of the existing technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions;

[0008] This invention discloses a method for adaptive control of ventilation and energy consumption in a smart grain warehouse, comprising:

[0009] A multi-dimensional data acquisition matrix was established to collect real-time temperature and humidity data at different depths of the grain pile inside the flat warehouse, temperature and humidity data of the external environment, and real-time operating power data of the ventilation equipment inside the warehouse. The collected multi-dimensional data underwent noise reduction and normalization processing to construct a coupled correlation model between grain pile temperature and humidity changes and ventilation equipment energy consumption. Based on this model, energy consumption level intervals were defined. According to the safe storage temperature and humidity standards for the grain varieties stored in the flat warehouse and the preset energy consumption control upper limit, dual-objective control thresholds for ventilation effect and energy consumption were set. Based on the coupled correlation model and the dual-objective control thresholds, the energy consumption level interval corresponding to the current data was automatically matched, generating an adaptive control strategy for the operating speed and single-run duration of the ventilation equipment. The ventilation equipment was driven based on the generated adaptive control strategy, while simultaneously monitoring the temperature and humidity change trends of the grain pile and the actual energy consumption value of the equipment in real time. If the data deviated from the dual-objective control thresholds, the control parameters were corrected immediately. Using the theoretical energy consumption and temperature and humidity dual-characteristic curves output by the coupled correlation model, combined with the real-time monitored temperature and humidity change rate and energy consumption fluctuation value, grain storage safety warning levels and energy consumption anomaly warning levels were defined. When the monitored value exceeded the corresponding warning level threshold, graded warning information was automatically pushed.

[0010] Furthermore, the multi-dimensional data acquisition matrix includes:

[0011] Divide the grain pile in the flat warehouse into at least 3 sampling depth layers in the vertical direction. In each depth layer, at least 2 temperature and humidity sampling points are evenly set around the central axis of the grain pile, and the distance between the sampling points and the warehouse wall is not less than one-third of the warehouse radius.

[0012] One ambient temperature and humidity sampling point is set up in the upwind direction, downwind direction and directly above the warehouse in the external environment. The horizontal distance between the upwind and downwind sampling points and the warehouse is 1 to 2 times the height of the warehouse.

[0013] A power acquisition module is connected in series in the main power supply circuit of the ventilation equipment in the warehouse to collect the instantaneous power and cumulative power of the equipment in real time.

[0014] The temperature and humidity data and power data of each collection point are associated with the collection timestamp to construct a multi-dimensional data collection matrix in which the collection time, collection location, data type and data value correspond one-to-one.

[0015] The time interval for collecting timestamps is dynamically adjusted based on the rate of change of temperature and humidity in the grain pile inside the warehouse.

[0016] When the rate of change of temperature and humidity exceeds the preset rate threshold, the sampling time interval is shortened to half of the original interval; when the rate of change of temperature and humidity is lower than the preset rate threshold, the sampling time interval is maintained or the sampling time interval is extended to three-half of the original interval.

[0017] Furthermore, when denoising the collected multidimensional data, the db4 wavelet basis function is selected to perform three-level wavelet decomposition on the data, retaining the low-frequency approximation coefficients, and using hard threshold quantization on the high-frequency detail coefficients. The threshold is 1.5-2 times the standard deviation of the high-frequency coefficients in that level. Then, the data is reconstructed through inverse wavelet transform to remove random noise.

[0018] When normalizing the collected multidimensional data, the denoised data is obtained and normalized using the following formula:

[0019] ;

[0020] In the formula: These are the normalized data values; These are the original data values ​​after noise reduction; The maximum and minimum values ​​of the same data type from the same data collection location within a preset historical period; For collecting depth weighting coefficients;

[0021] in, ∈[0.8,1.2], express The maximum value within the range of values.

[0022] Furthermore, in the stage of constructing a coupled correlation model between the temperature and humidity changes of the grain pile and the energy consumption of the ventilation equipment, the temperature and humidity changes at each depth layer of the grain pile inside the warehouse after noise reduction and normalization, as well as the temperature and humidity difference between the external environment and the grain pile inside the warehouse, are used as model input variables. The energy consumption per unit time of the ventilation equipment is used as the model output variable. An improved BP neural network is used to construct the coupled correlation model. The improved BP neural network includes one input layer, two hidden layers, and one output layer. The activation function of the hidden layer is... The following loss function is used for parameter optimization during model training:

[0023] ;

[0024] In the formula: The output value of the ELU activation function used for the hidden layer neurons; To improve the weighted input values ​​of neurons in the hidden layer of a BP neural network; The value of the loss function; This represents the number of training samples; Predict the energy consumption per unit time for the i-th sample using the model; This represents the actual energy consumption per unit time for the i-th sample. This is the energy consumption weighting coefficient.

[0025] Furthermore, when defining energy consumption level ranges based on the coupled correlation model, the following applies:

[0026] Based on the correlation between energy consumption per unit time and changes in temperature and humidity of grain piles output by the coupled correlation model, at least 500 sets of valid sample data were extracted, and the extracted samples covered different external environmental conditions and different grain storage stages.

[0027] By analyzing the distribution density of sample data in the characteristic space of energy consumption and temperature and humidity changes, the initial grouping center is determined based on the data density abrupt change points in the characteristic space, and the number of groups is set to 3 to 5.

[0028] Set constraints:

[0029] The maximum energy consumption per unit time within each energy consumption level range shall not exceed the preset energy consumption control limit.

[0030] The temperature and humidity changes of the grain pile corresponding to each energy consumption level range all fall within the allowable fluctuation range of the safe storage temperature and humidity standards for the stored grain varieties.

[0031] During the grouping process, the samples are checked in real time to see if they meet the dual constraints, and only samples that meet the constraints are assigned to the corresponding interval.

[0032] Finally, the initially formed energy consumption level intervals are iteratively optimized: if an interval does not meet any constraint condition, the deviation between the characteristic value of the interval and the constraint condition is calculated first, and the group center position is dynamically corrected according to the deviation. The above grouping steps are then repeated until all intervals meet the dual constraint conditions.

[0033] Furthermore, when setting the dual-objective control thresholds for ventilation efficiency and energy consumption, the following conditions must be met:

[0034] Determine the basic safe temperature and humidity thresholds for grain storage varieties;

[0035] Calculate the external environment correction factor , These are the weighting coefficients for temperature difference and humidity difference, respectively. The values ​​all fall within the range of [0.4, 0.6], and The sum is 1; The ambient temperature outside the warehouse; The average temperature of the grain pile inside the warehouse; Humidity of the external environment; The average humidity of the grain pile inside the warehouse;

[0036] The basic safe temperature and humidity thresholds are corrected to dynamic safe temperature and humidity thresholds, that is, the basic safe temperature and humidity thresholds are respectively compared with... Multiply them to obtain the corrected safe temperature and humidity threshold, which is also known as the dynamic safe temperature and humidity threshold.

[0037] Determine the basic energy consumption control threshold for the ventilation equipment, initially set at 80% of the equipment's rated power, and set a grain storage time correction coefficient. The grain storage time correction coefficient increases linearly with the increase of grain storage time, with a value range of [1.0, 1.5]. In the initial stage of storage, the grain storage time correction coefficient is set to 1. The basic energy consumption control threshold is then corrected to the energy consumption control threshold, which is also the dynamic energy consumption control threshold. This is the product of the basic energy consumption control threshold of the ventilation equipment and the grain storage time correction coefficient.

[0038] The dynamic safety temperature and humidity threshold and the dynamic energy consumption control threshold are denoted as the dual-objective control threshold.

[0039] Furthermore, based on the coupled correlation model and dual-objective control threshold, the operation phase of automatically matching the energy consumption level range corresponding to the current data and generating an adaptive control strategy for the operating speed and single-run duration of the ventilation equipment includes the following steps:

[0040] Step 1: Obtain the temperature and humidity data of each depth layer of the grain pile after noise reduction and normalization at the current moment, and the temperature and humidity data of the external environment. Input them into the coupled correlation model to obtain the predicted energy consumption value per unit time at the current moment.

[0041] Step 2: Match the current predicted energy consumption per unit time with the upper and lower limits of each energy consumption level range to determine the current energy consumption level. For example, the low energy consumption range is 0 to 0.3 times the dynamic energy consumption control threshold (inclusive), the medium energy consumption range is 0.3 to 0.7 times the dynamic energy consumption control threshold (inclusive), and the high energy consumption range is 0.7 to 1 times the dynamic energy consumption control threshold (inclusive).

[0042] Step 3: If the current temperature and humidity of the grain pile exceed the dynamic safe temperature and humidity threshold, then the following applies: When the energy consumption level is low, increase the ventilation equipment speed to the maximum allowable speed corresponding to that level; when the energy consumption level is high, adjust the speed to the minimum allowable speed corresponding to that level; simultaneously calculate the temperature and humidity deviation rate. , This indicates the actual average temperature and humidity of the grain pile inside the warehouse at the current moment. Indicates the dynamic safe temperature and humidity threshold. If the preset deviation threshold is exceeded, the duration of each run will be extended by the ratio of ΔE to the preset deviation threshold; if the current temperature and humidity of the grain pile are within the dynamic safe temperature and humidity threshold, energy consumption will be controlled first.

[0043] Adjust the speed of the ventilation equipment to the optimal speed corresponding to its energy consumption level. The duration of a single run is determined by the temperature and humidity difference between the external environment and the grain pile inside the warehouse. That is, the larger the difference, the shorter the run time, and the smaller the difference, the longer the run time.

[0044] Furthermore, if the data deviates from the dual-target control threshold, the control parameters are corrected immediately as follows:

[0045] The deviations are calculated in real time, including: the absolute value of the difference between the actual grain pile temperature and the dynamic safe temperature threshold, the absolute value of the difference between the actual grain pile humidity and the dynamic safe humidity threshold, and the absolute value of the difference between the actual equipment energy consumption and the dynamic energy consumption control threshold.

[0046] Setting deviation thresholds: Temperature deviation threshold, humidity deviation threshold, and energy consumption deviation threshold are all preset by the user based on grain storage safety requirements and equipment operational stability. When any deviation exceeds the corresponding threshold, parameter correction is initiated.

[0047] ;

[0048] In the formula: This is the correction amount for the operating speed of the ventilation equipment; This is the proportionality coefficient; This is the overall deviation. The integral coefficient; These are the differential coefficients;

[0049] in, ;

[0050] For the single-run duration of ventilation equipment, pulse width modulation duty cycle adjustment is used: according to Determine the pulse width modulation duty cycle , Indicates the basic duty cycle. This is the duty cycle adjustment coefficient. ∈[0.1,0.3];

[0051] Finally, based on Adjust the current operating speed according to Adjust the single run duration, correct and continuously monitor the data until all deviations are below the corresponding threshold.

[0052] Furthermore, the system divides the grain storage safety early warning level into two stages: one for grain storage safety and the other for abnormal energy consumption. Based on the theoretical energy consumption and temperature and humidity dual-characteristic curves output by the coupled correlation model, the system calculates the real-time temperature and humidity change rate and the standard deviation of the fluctuation of the actual energy consumption of the equipment within the past preset time interval, and simultaneously calculates the grain storage safety early warning index. ;

[0053] In the formula: This represents the real-time rate of change in temperature and humidity. The actual average temperature and humidity of the grain pile inside the warehouse at the current moment. Dynamic safety temperature and humidity thresholds;

[0054] according to Classifying energy consumption anomaly warning levels: When < At the time of safety level, when ≤ < At the time, it was at the attention level, when ≥ It was at the emergency level, among which To preset the warning index threshold, and < ;

[0055] Further calculate the energy consumption anomaly early warning index , This indicates the actual energy consumption of the equipment. Indicates the dynamic energy consumption control threshold. This represents the standard deviation of the actual energy consumption fluctuation of the equipment within a preset time interval in the past.

[0056] according to Classifying energy consumption anomaly warning levels: When < When it is in the normal level, when J1≤ < When it is an abnormal level, ≥ It was classified as severe, among which To preset the energy consumption warning threshold, and .

[0057] Furthermore, when tiered early warning information is pushed out, it is fed back to the corresponding preset matching receiver according to the warning level.

[0058] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0059] This invention provides a smart grain silo-based adaptive control method for ventilation and energy consumption linkage. During execution, this method can accurately collect real-time data on different depths of the grain pile, external temperature and humidity, and equipment operating power. It can also dynamically adjust the collection interval based on the rate of change in temperature and humidity of the grain pile, and reduce data errors through noise reduction and normalization processing. By using a coupled correlation model, it accurately establishes the relationship between temperature and humidity changes and energy consumption, classifying levels that meet both grain storage safety and energy consumption limits. Based on dual-objective control thresholds, it automatically matches energy consumption levels and generates strategies for ventilation equipment speed and operating time, ensuring both grain storage safety and avoiding energy waste. It monitors data in real-time and corrects control parameters immediately to prevent data deviation from thresholds. Furthermore, it can classify grain storage safety and energy consumption anomaly warning levels and push information to promptly avoid risks, thus improving the overall intelligence level of grain silo ventilation control and effectively reducing energy costs while ensuring grain storage safety. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0061] Figure 1 This is a flowchart illustrating an adaptive control method for ventilation and energy consumption linkage based on intelligent grain storage. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0063] The present invention will be further described below with reference to embodiments.

[0064] Example:

[0065] This embodiment presents an adaptive control method for ventilation and energy consumption linkage based on a smart grain warehouse, such as... Figure 1 As shown, it includes:

[0066] Establish a multi-dimensional data acquisition matrix to collect real-time temperature and humidity data at different depths of grain piles inside the flat warehouse, temperature and humidity data of the external environment, and real-time operating power data of ventilation equipment inside the warehouse.

[0067] The multi-dimensional data collection matrix includes:

[0068] Divide the grain pile in the flat warehouse into at least 3 sampling depth layers in the vertical direction. In each depth layer, at least 2 temperature and humidity sampling points are evenly set around the central axis of the grain pile, and the distance between the sampling points and the warehouse wall is not less than one-third of the warehouse radius.

[0069] One ambient temperature and humidity sampling point is set up in the upwind direction, downwind direction and directly above the warehouse in the external environment. The horizontal distance between the upwind and downwind sampling points and the warehouse is 1 to 2 times the height of the warehouse.

[0070] A power acquisition module is connected in series in the main power supply circuit of the ventilation equipment in the warehouse to collect the instantaneous power and cumulative power of the equipment in real time.

[0071] The temperature and humidity data and power data of each collection point are associated with the collection timestamp to construct a multi-dimensional data collection matrix in which the collection time, collection location, data type and data value correspond one-to-one.

[0072] The time interval for collecting timestamps is dynamically adjusted based on the rate of change of temperature and humidity in the grain pile inside the warehouse.

[0073] When the rate of change of temperature and humidity exceeds the preset rate threshold, the sampling time interval is shortened to half of the original interval; when the rate of change of temperature and humidity is lower than the preset rate threshold, the sampling time interval is maintained or the sampling time interval is extended to three-half of the original interval.

[0074] The collected multi-dimensional data were denoised and normalized to construct a coupled correlation model between the temperature and humidity changes of the grain pile and the energy consumption of the ventilation equipment, and energy consumption level ranges were defined based on the model.

[0075] When denoising the collected multidimensional data, the db4 wavelet basis function is selected to perform three-level wavelet decomposition on the data, retaining the low-frequency approximation coefficients, and using hard threshold quantization for the high-frequency detail coefficients. The threshold is 1.5-2 times the standard deviation of the high-frequency coefficients in that level. Then, the data is reconstructed through inverse wavelet transform to remove random noise.

[0076] When normalizing the collected multidimensional data, the denoised data is obtained and normalized using the following formula:

[0077] ;

[0078] In the formula: These are the normalized data values; These are the original data values ​​after noise reduction; The maximum and minimum values ​​of the same data type from the same data collection location within a preset historical period; For collecting depth weighting coefficients;

[0079] in, ∈[0.8,1.2], The value of increases with increasing grain pile collection depth and decreases with decreasing grain pile collection depth. express The maximum value within the range of values;

[0080] Based on the safe storage temperature and humidity standards for grain varieties in flat warehouses and the preset energy consumption control upper limit, dual target control thresholds for ventilation effect and energy consumption are set.

[0081] In the stage of constructing a coupled correlation model between temperature and humidity changes in the grain pile and energy consumption of ventilation equipment, the temperature and humidity changes at each depth layer of the grain pile inside the warehouse after noise reduction and normalization, as well as the temperature and humidity difference between the external environment and the grain pile inside the warehouse, are used as model input variables. The energy consumption per unit time of the ventilation equipment is used as the model output variable. An improved BP neural network is used to construct the coupled correlation model. The improved BP neural network includes one input layer (4 neurons, corresponding to 4 input variables), two hidden layers (12 neurons in the first hidden layer and 8 neurons in the second hidden layer), and one output layer (1 neuron, corresponding to Punit). The activation function of the hidden layer is... The following loss function is used for parameter optimization during model training:

[0082] ;

[0083] In the formula: The output value of the ELU activation function used for the hidden layer neurons; To improve the weighted input values ​​of neurons in the hidden layer of a BP neural network; The value of the loss function; This represents the number of training samples; Predict the energy consumption per unit time for the i-th sample using the model; This represents the actual energy consumption per unit time for the i-th sample. Energy consumption weighting coefficient;

[0084] The above formula averages the squared difference between the model's predicted energy consumption per unit time and the actual energy consumption per unit time for each sample by the energy consumption weighting coefficient. This design highlights the importance of energy consumption data in model training through the energy consumption weighting coefficient, which can guide the model to learn more accurately the correlation between changes in grain pile temperature and humidity and the energy consumption of ventilation equipment. At the same time, the use of squared error effectively amplifies the impact of larger error samples on the optimization of model parameters, prompting the model to continuously adjust parameters to reduce prediction errors and improve the accuracy of the coupled correlation model in predicting energy consumption.

[0085] in, The value is set based on the preset energy consumption control upper limit: when the preset energy consumption control upper limit is low, The value ranges from 1.2 to 1.5; when the preset energy consumption control upper limit is high, The value ranges from 0.8 to 1.0;

[0086] Based on the coupled correlation model and dual-objective control threshold, the system automatically matches the energy consumption level range corresponding to the current data and generates an adaptive control strategy for the operating speed and single running time of the ventilation equipment.

[0087] When setting the dual-objective control thresholds for ventilation efficiency and energy consumption, the following conditions must be met:

[0088] Determine the basic safe temperature and humidity thresholds for grain varieties, which are based on the safe storage standards for grain varieties issued by the state or industry.

[0089] Calculate the external environment correction factor , These are the weighting coefficients for temperature difference and humidity difference, respectively. The values ​​all fall within the range of [0.4, 0.6], and The sum is 1; The ambient temperature outside the warehouse; The average temperature of the grain pile inside the warehouse; Humidity of the external environment; The average humidity of the grain pile inside the warehouse;

[0090] The above formula is obtained by multiplying the temperature difference weighting coefficient and the humidity difference weighting coefficient by the proportion of the temperature and humidity difference between the outside and inside grain piles to the temperature and humidity of the grain pile inside the warehouse, and adding 1. The weighting coefficient balances the effects of temperature and humidity differences, avoiding the one-sidedness of judging the safety threshold by a single environmental factor. This allows the calculated correction coefficient to dynamically reflect the changes in grain storage safety requirements under different external environments, providing a scientific basis for correcting the basic safety temperature and humidity threshold to a dynamic safety temperature and humidity threshold, and ensuring that the safety threshold setting is more in line with the actual grain storage environment.

[0091] The basic safe temperature and humidity thresholds are corrected to dynamic safe temperature and humidity thresholds, that is, the basic safe temperature and humidity thresholds are respectively compared with... Multiply them to obtain the corrected safe temperature and humidity threshold, which is also known as the dynamic safe temperature and humidity threshold.

[0092] Determine the basic energy consumption control threshold for the ventilation equipment, initially set at 80% of the equipment's rated power, and set a grain storage time correction coefficient. The grain storage time correction coefficient increases linearly with the increase of grain storage time, with a value range of [1.0, 1.5]. In the initial stage of storage, the grain storage time correction coefficient is set to 1. The basic energy consumption control threshold is then corrected to the energy consumption control threshold, which is also the dynamic energy consumption control threshold. This is the product of the basic energy consumption control threshold of the ventilation equipment and the grain storage time correction coefficient.

[0093] The dynamic safe temperature and humidity threshold and the dynamic energy consumption control threshold are denoted as the dual-objective control threshold.

[0094] The ventilation equipment is driven by the generated adaptive control strategy, while the temperature and humidity of the grain pile and the actual energy consumption of the equipment are monitored in real time. If the data deviates from the dual target control threshold, the control parameters are corrected immediately.

[0095] By combining the theoretical energy consumption and temperature and humidity dual characterization curves output by the coupled correlation model with the real-time monitoring of temperature and humidity change rates and energy consumption fluctuation values, the grain storage safety warning level and the energy consumption abnormality warning level are divided. When the monitored value exceeds the corresponding warning level threshold, the graded warning information is automatically pushed.

[0096] When defining energy consumption level ranges based on a coupled correlation model, the following applies:

[0097] Based on the correlation between energy consumption per unit time and changes in temperature and humidity of grain piles output by the coupled correlation model, at least 500 sets of valid sample data were extracted, and the extracted samples covered different external environmental conditions and different grain storage stages.

[0098] By analyzing the distribution density of sample data in the characteristic space of energy consumption and temperature and humidity changes, the initial grouping center is determined based on the data density abrupt change points in the characteristic space, and the number of groups is set to 3 to 5.

[0099] Set constraints:

[0100] The maximum energy consumption per unit time within each energy consumption level range shall not exceed the preset energy consumption control limit.

[0101] The temperature and humidity changes of the grain pile corresponding to each energy consumption level range all fall within the allowable fluctuation range of the safe storage temperature and humidity standards for the stored grain varieties.

[0102] During the grouping process, the samples are checked in real time to see if they meet the dual constraints, and only samples that meet the constraints are assigned to the corresponding interval.

[0103] Finally, the initially formed energy consumption level intervals are iteratively optimized: if there is an interval that does not meet any constraint condition, first calculate the deviation of the characteristic value of the interval from the constraint condition. The characteristic value includes the maximum energy consumption value and the extreme value of temperature and humidity change in the interval. Then, the group center position is dynamically corrected according to the deviation. The dynamic correction follows the rule that the larger the deviation, the larger the center adjustment range. Then, the above grouping steps are repeated until all intervals meet the dual constraint conditions.

[0104] Based on a coupled correlation model and dual-objective control thresholds, the operation phase of automatically matching the energy consumption level range corresponding to the current data and generating an adaptive control strategy for the operating speed and single-run duration of ventilation equipment includes the following steps:

[0105] Step 1: Obtain the temperature and humidity data of each depth layer of the grain pile after noise reduction and normalization at the current moment, and the temperature and humidity data of the external environment. Input them into the coupled correlation model to obtain the predicted energy consumption value per unit time at the current moment.

[0106] Step 2: Match the current predicted energy consumption per unit time with the upper and lower limits of each energy consumption level range to determine the current energy consumption level. For example, the low energy consumption range is 0 to 0.3 times the dynamic energy consumption control threshold (inclusive), the medium energy consumption range is 0.3 to 0.7 times the dynamic energy consumption control threshold (inclusive), and the high energy consumption range is 0.7 to 1 times the dynamic energy consumption control threshold (inclusive).

[0107] Step 3: If the current temperature and humidity of the grain pile exceed the dynamic safe temperature and humidity threshold, then the following applies: When the energy consumption level is low, increase the ventilation equipment speed to the maximum allowable speed corresponding to that level; when the energy consumption level is high, adjust the speed to the minimum allowable speed corresponding to that level; simultaneously calculate the temperature and humidity deviation rate. , This indicates the actual average temperature and humidity of the grain pile inside the warehouse at the current moment. Indicates the dynamic safe temperature and humidity threshold. If the preset deviation threshold is exceeded, the duration of each run will be extended by the ratio of ΔE to the preset deviation threshold; if the current temperature and humidity of the grain pile are within the dynamic safe temperature and humidity threshold, energy consumption will be controlled first.

[0108] The above formula amplifies the impact of large deviations through squaring operations. At the same time, it uses the square root form to convert the deviation rate into a single quantitative indicator, which can intuitively and accurately reflect the degree to which the temperature and humidity of the grain pile deviate from the safety threshold. This provides a clear basis for judging whether it is necessary to extend the single running time of the ventilation equipment, and avoids the risk to grain storage safety or unnecessary increase in energy consumption caused by inaccurate judgment of temperature and humidity deviation.

[0109] The speed of the ventilation equipment is adjusted to the optimal speed corresponding to the energy consumption level. This speed is calculated by a coupled correlation model. In order to minimize energy consumption per unit time and maintain stable temperature and humidity, the running time of a single operation is determined based on the temperature and humidity difference between the external environment and the grain pile inside the warehouse. That is, the larger the difference, the shorter the running time, and the smaller the difference, the longer the running time.

[0110] If the data deviates from the dual-target control threshold, the control parameters should be corrected immediately as follows:

[0111] The deviations are calculated in real time, including: the absolute value of the difference between the actual grain pile temperature and the dynamic safe temperature threshold, the absolute value of the difference between the actual grain pile humidity and the dynamic safe humidity threshold, and the absolute value of the difference between the actual equipment energy consumption and the dynamic energy consumption control threshold.

[0112] Setting deviation thresholds: Temperature deviation threshold, humidity deviation threshold, and energy consumption deviation threshold are all preset by the user based on grain storage safety requirements and equipment operational stability. When any deviation exceeds the corresponding threshold, parameter correction is initiated.

[0113] ;

[0114] In the formula: This is the correction amount for the operating speed of the ventilation equipment; This is the proportionality coefficient; This is the overall deviation. The integral coefficient; These are the differential coefficients;

[0115] The above formula responds quickly to the current comprehensive deviation through the proportional term and adjusts the speed in time to deal with the deviation; it also eliminates the static error of the system through the integral term, ensuring that the deviation continues to decrease during long-term operation; and it predicts the changing trend of the comprehensive deviation through the differential term, adjusting the speed in advance to suppress the expansion of the deviation. The three work together to achieve accurate and rapid correction of the speed of the ventilation equipment, effectively solving the problem of speed adjustment when the data deviates from the dual target control threshold, and ensuring that the ventilation effect and energy consumption are controlled within a reasonable range.

[0116] in, ∈[0.2,0.5], when When the deviation is large and the ventilation equipment parameters need to be adjusted quickly to reduce the deviation, the value should be larger. When the parameter is relatively small, and it is necessary to avoid overshooting of parameter correction that could lead to system instability, the smaller the value should be. ∈[0.01,0.05], the larger the value is when there is a continuous deviation in the temperature and humidity of the grain pile or the energy consumption of the equipment, and it is necessary to eliminate the static error; the smaller the value is when the deviation fluctuates frequently and it is necessary to avoid the integral saturation leading to excessive parameter correction. ∈[0.05,0.1], when the comprehensive deviation The value should be larger when the rate of change is rapid and the trend of deviation expansion needs to be suppressed in advance; the value should be smaller when the deviation changes slowly and excessive intervention to avoid system oscillation needs to be avoided. ;

[0117] For the single-run duration of ventilation equipment, pulse width modulation duty cycle adjustment is used: according to Determine the pulse width modulation duty cycle , Indicates the basic duty cycle. This is the duty cycle adjustment coefficient. ∈[0.1,0.3], when The more severe the deviation of the grain pile temperature, humidity and energy consumption from the dual-target control threshold, the larger the value; conversely, the smaller the value.

[0118] Finally, based on Adjust the current operating speed according to Adjust the single run duration, and continue to monitor the data after correction until all deviations are below the corresponding thresholds.

[0119] The system divides the grain storage safety early warning level into two stages: one for grain storage safety and the other for abnormal energy consumption. Based on the theoretical energy consumption and temperature / humidity dual characterization curves output by the coupled correlation model, it calculates the real-time temperature and humidity change rate and the standard deviation of the fluctuation of the actual energy consumption of the equipment within a preset time interval, and simultaneously calculates the grain storage safety early warning index. ;

[0120] In the formula: This represents the real-time rate of change in temperature and humidity. The actual average temperature and humidity of the grain pile inside the warehouse at the current moment. Dynamic safety temperature and humidity thresholds;

[0121] The above formula takes into account the dynamic factor of the rate of change of temperature and humidity during operation, which can capture the safety risks brought about by rapid changes in temperature and humidity in a timely manner. It also incorporates the static deviation between the current temperature and humidity and the safety threshold, comprehensively reflecting the grain storage safety status. At the same time, the influence of dynamic changes and static deviations is balanced by the weighting coefficient, so that the early warning index can scientifically quantify the grain storage safety level, providing an accurate indicator for classifying the grain storage safety early warning level, which facilitates the timely detection and handling of potential grain storage safety hazards.

[0122] according to Classifying energy consumption anomaly warning levels: When < At the safe level (no intervention required), when ≤ < At the time, it was at the level of concern (monitoring needs to be strengthened). ≥ The situation is classified as emergency (requiring immediate intervention), among which... To preset the warning index threshold, and < ;

[0123] Further calculate the energy consumption anomaly early warning index , This indicates the actual energy consumption of the equipment. Indicates the dynamic energy consumption control threshold. This represents the standard deviation of the actual energy consumption fluctuation of the equipment within a preset time interval in the past.

[0124] The above-mentioned early warning index formula design not only focuses on the situation where the current actual energy consumption exceeds the control threshold, but also highlights the impact of excessive energy consumption by using a large weighting coefficient. It also considers the stability of energy consumption fluctuations, avoids misjudgment caused by short-term energy consumption fluctuations, and can comprehensively and accurately assess the degree of energy consumption anomaly. It provides a reliable basis for classifying energy consumption anomaly early warning levels, helps to promptly detect and solve the problem of abnormal energy consumption of ventilation equipment, and ensures that energy consumption is controlled within the preset range.

[0125] according to Classifying energy consumption anomaly warning levels: When < When it is in the normal level, when J1≤ < When it is an abnormal level, ≥ It was classified as severe, among which To preset the energy consumption warning threshold, and ;

[0126] When tiered early warning information is pushed out, it is fed back to the corresponding preset matching receiver according to the early warning level.

[0127] The method described in the above embodiments, in the scenario of storing grain in a flat warehouse, can monitor the different depths of the grain pile, the temperature and humidity outside the warehouse, and the energy consumption of the equipment in real time. After data processing, the energy consumption range is accurately matched, and the operating status of the equipment is adaptively adjusted to ensure that the temperature and humidity of the grain pile meet the safety standards. When the data deviates, the parameters can be corrected in time, and the warning can be pushed out in stages. This not only effectively protects the safety of grain storage and avoids grain storage risks and abnormal energy consumption, but also reduces energy waste, thereby improving the efficiency and economy of grain warehouse management.

[0128] Referring to the method in the above embodiments, the following is an application example of this method:

[0129] A grain depot uses a 15m x 20m flat warehouse to store wheat (safe storage temperature ≤25℃, humidity ≤13%). The specific application of this method is as follows:

[0130] A multi-dimensional data acquisition matrix was established: Three depth layers (0.5m, 1.5m, and 2.5m) were set up along the vertical direction of the grain pile inside the warehouse. Two temperature and humidity acquisition points were set up around the central axis of the grain pile in each layer (warehouse radius 3m, acquisition point distance 1.2m from warehouse wall, meeting the requirement of ≥1 / 3 warehouse radius). One environmental temperature and humidity acquisition point was set up outside the warehouse at the upwind and downwind directions (6m from warehouse, warehouse height 5m, meeting the requirement of 1-2 times warehouse height) and directly above the warehouse. A power acquisition module was connected in series with the main power supply circuit of the ventilation equipment (rated power 10kW) inside the warehouse to collect instantaneous and cumulative power in real time. The initial data collection timestamp interval was 10 minutes, shortened to 5 minutes when the rate of change of temperature and humidity in the grain pile exceeded a preset threshold, and extended to 15 minutes when it was below the threshold.

[0131] Data denoising and normalization: The collected data was decomposed into three layers using the db4 wavelet basis function. The high-frequency detail coefficients were quantized with a hard threshold of 0.8 (1.8 times the standard deviation of the high-frequency coefficients in that layer) and then denoised using inverse wavelet transform. Taking the temperature at a depth of 0.5m as an example, the original value after denoising was 25℃. The maximum temperature at this location during the historical period was 30℃ and the minimum temperature was 15℃. The depth weighting coefficient was 1.0 (ωmax=1.2), and the normalized result was 0.67.

[0132] A coupled correlation model was constructed and energy consumption intervals were defined: The temperature and humidity changes at each depth layer and the temperature and humidity difference between inside and outside the storage area were used as inputs, and the energy consumption per unit time of the equipment was used as the output. An improved BP neural network (1 input layer, 2 hidden layers, 1 output layer, with the hidden layer's ELU activation function α=0.1) was used to construct the coupled correlation model. 600 valid samples were extracted, and three energy consumption intervals were defined (the dynamic energy consumption control threshold for the initial stage of grain storage was 8kW): low energy consumption 0-2.4kW (0-0.3 times the dynamic threshold), medium energy consumption 2.4-5.6kW (0.3-0.7 times the dynamic threshold), and high energy consumption 5.6-8kW (0.7-1 times the dynamic threshold). The temperature and humidity fluctuations in each interval met the safety requirements for wheat.

[0133] The dual-target control thresholds are set as follows: basic safe temperature and humidity for wheat are 25℃ and 13%; outside temperature is 28℃ and humidity is 15%; average inside temperature is 24℃ and humidity is 12%. Taking α=0.5 and β=0.5 (summing to 1), the environmental correction coefficient is calculated to be 1.05, and the dynamic safe temperature and humidity are 26.25℃ and 13.65%. The basic energy consumption threshold for equipment is 8kW (80% of rated power), the initial grain storage correction coefficient is 1.0, and the dynamic energy consumption threshold is 8kW.

[0134] Generate an adaptive control strategy: Obtain the current temperature and humidity of the grain pile at various depths after noise reduction and normalization (23-25℃), and the outside temperature and humidity (28℃ / 15%). Input these into the model to obtain a predicted energy consumption value of 2.2kW per unit time, matching the low energy consumption range. Since the current temperature and humidity of the grain pile (24℃, 12%) are within the dynamic safety threshold, energy consumption is prioritized, and the equipment speed is adjusted to the optimal speed of 1200r / min in the low energy consumption range. Due to the large temperature and humidity difference between inside and outside the warehouse, the single operation time is set to 30 minutes.

[0135] Real-time control parameter correction: During operation, the actual average temperature of the grain pile was monitored to be 27℃ (0.75℃ above the dynamic threshold), humidity was 14% (0.35% above the threshold), and the actual energy consumption of the equipment was 2.8kW (0.4kW upper limit of the ultra-low energy consumption range). The calculated comprehensive deviation was 0.45. The speed correction was calculated to be 150r / min using the PID algorithm, and the adjusted speed was 1350r / min. The pulse width modulation duty cycle was calculated to be 0.65, and the single run time was extended to 40 minutes, continuously monitored until the deviation was below the threshold.

[0136] Tiered Early Warning: The current grain storage safety early warning index is 0.3 (preset I1=0.4, I2=0.8), which is at the safe level; the energy consumption anomaly early warning index is 0.045 (actual energy consumption 2.8kW, dynamic threshold 8kW, fluctuation standard deviation 0.3kW), which is at the normal level, and no early warning is pushed. After 3 months of grain storage (dynamic energy consumption threshold rises to 9.6kW, correction coefficient 1.2), if the actual energy consumption of the equipment is monitored at 11kW and the fluctuation standard deviation is 0.8kW, the calculated energy consumption early warning index is 0.127 (preset J1=0.1, J2=0.3), which is at the abnormal level, and an early warning is pushed to the equipment management group; if the actual temperature of the grain pile is 28℃ and the humidity is 14.2%, the calculated safety early warning index is 0.55, which is at the attention level, and an early warning is pushed to the grain depot operation and maintenance group.

[0137] In summary, the methods described in the above embodiments can accurately collect data on different depths of the grain pile, external temperature and humidity, and equipment operating power in real time during execution. They can also dynamically adjust the collection interval based on the rate of change in temperature and humidity of the grain pile, reduce data errors through noise reduction and normalization, accurately establish the correlation between temperature and humidity changes and energy consumption through a coupled correlation model, divide the storage safety and energy consumption upper limit into grade ranges, automatically match energy consumption levels based on dual-objective control thresholds, and generate ventilation equipment speed and operating time strategies. This ensures both grain storage safety and avoids energy waste, monitors data in real time and corrects control parameters immediately to prevent data deviation from thresholds, and can also classify grain storage safety and energy consumption anomaly warning levels and push information to avoid risks in a timely manner. Overall, this improves the intelligence level of grain warehouse ventilation control, effectively reducing energy costs while ensuring grain storage safety.

[0138] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for adaptive control of ventilation and energy consumption linked together in a smart grain warehouse, characterized in that, include: Establish a multi-dimensional data acquisition matrix to collect real-time temperature and humidity data of grain piles at different depths inside the flat warehouse, temperature and humidity data of the external environment, and real-time operating power data of ventilation equipment inside the warehouse. The collected multi-dimensional data were denoised and normalized to construct a coupled correlation model between the temperature and humidity changes of the grain pile and the energy consumption of the ventilation equipment, and energy consumption level ranges were defined based on the model. Based on the safe storage temperature and humidity standards for grain varieties in flat warehouses and the preset energy consumption control upper limit, dual target control thresholds for ventilation effect and energy consumption are set. Based on the coupled correlation model and dual-objective control threshold, the system automatically matches the energy consumption level range corresponding to the current data and generates an adaptive control strategy for the operating speed and single running time of the ventilation equipment. The operational phase of automatically matching the energy consumption level range corresponding to the current data based on the coupled correlation model and dual-objective control threshold to generate the adaptive control strategy for the operating speed and single-run duration of the ventilation equipment includes the following steps: Step 1: Obtain the temperature and humidity data of each depth layer of the grain pile after noise reduction and normalization at the current moment, and the temperature and humidity data of the external environment. Input them into the coupled correlation model to obtain the predicted energy consumption value per unit time at the current moment. Step 2: Match the current energy consumption prediction value per unit time with the upper and lower limits of each energy consumption level range to determine the current energy consumption level; Step 3: If the current temperature and humidity of the grain pile exceed the dynamic safe temperature and humidity threshold, then the following applies: When the energy consumption level is low, increase the ventilation equipment speed to the maximum allowable speed corresponding to that level; when the energy consumption level is high, adjust the speed to the minimum allowable speed corresponding to that level; simultaneously calculate the temperature and humidity deviation rate. , This indicates the actual average temperature and humidity of the grain pile inside the warehouse at the current moment. Indicates the dynamic safe temperature and humidity threshold. If the preset deviation threshold is exceeded, the duration of each run will be extended by the ratio of ΔE to the preset deviation threshold; if the current temperature and humidity of the grain pile are within the dynamic safe temperature and humidity threshold, energy consumption will be controlled first. Adjust the speed of the ventilation equipment to the optimal speed corresponding to its energy consumption level. The duration of a single run is determined based on the temperature and humidity difference between the external environment and the grain pile inside the warehouse. That is, the larger the difference, the shorter the run time, and the smaller the difference, the longer the run time. The ventilation equipment is driven by the generated adaptive control strategy, while the temperature and humidity of the grain pile and the actual energy consumption of the equipment are monitored in real time. If the data deviates from the dual-target control threshold, the control parameters are corrected immediately. By combining the theoretical energy consumption and temperature and humidity dual characterization curves output by the coupled correlation model with the real-time monitoring of temperature and humidity change rates and energy consumption fluctuation values, the grain storage safety warning level and energy consumption abnormality warning level are divided. When the monitored value exceeds the corresponding warning level threshold, the graded warning information is automatically pushed.

2. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, The multi-dimensional data acquisition matrix includes: Divide the grain pile in the flat warehouse into at least 3 sampling depth layers in the vertical direction. In each depth layer, at least 2 temperature and humidity sampling points are evenly set around the central axis of the grain pile, and the distance between the sampling points and the warehouse wall is not less than one-third of the warehouse radius. One ambient temperature and humidity sampling point is set up in the upwind direction, downwind direction and directly above the warehouse in the external environment. The horizontal distance between the upwind and downwind sampling points and the warehouse is 1 to 2 times the height of the warehouse. A power acquisition module is connected in series in the main power supply circuit of the ventilation equipment in the warehouse to collect the instantaneous power and cumulative power of the equipment in real time. The temperature and humidity data and power data of each collection point are associated with the collection timestamp to construct a multi-dimensional data collection matrix in which the collection time, collection location, data type and data value correspond one-to-one. The time interval for collecting the timestamps is dynamically adjusted based on the rate of change of temperature and humidity in the grain pile inside the warehouse. When the rate of change of temperature and humidity exceeds the preset rate threshold, the sampling time interval is shortened to half of the original interval; when the rate of change of temperature and humidity is lower than the preset rate threshold, the sampling time interval is maintained or the sampling time interval is extended to three-half of the original interval.

3. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, When denoising the collected multidimensional data, the db4 wavelet basis function is selected to perform three-level wavelet decomposition on the data, retaining the low-frequency approximation coefficients, and using hard threshold quantization for the high-frequency detail coefficients. The threshold is 1.5-2 times the standard deviation of the high-frequency coefficients in that level. Then, the data is reconstructed through inverse wavelet transform to remove random noise. When normalizing the collected multidimensional data, the denoised data is obtained and normalized using the following formula: ； In the formula: These are the normalized data values; These are the original data values ​​after noise reduction; The maximum and minimum values ​​of the same data type from the same data collection location within a preset historical period; For collecting depth weighting coefficients; in, ∈[0.8,1.2], express The maximum value within the range of values.

4. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, In the stage of constructing a coupled correlation model between temperature and humidity changes in the grain pile and energy consumption of ventilation equipment, the temperature and humidity changes at each depth layer of the grain pile inside the warehouse after noise reduction and normalization, as well as the temperature and humidity difference between the external environment and the grain pile inside the warehouse, are used as model input variables. The energy consumption per unit time of the ventilation equipment is used as the model output variable. An improved BP neural network is used to construct the coupled correlation model. The improved BP neural network includes one input layer, two hidden layers, and one output layer. The activation function of the hidden layer is... The following loss function is used for parameter optimization during model training: ； In the formula: The output value of the ELU activation function used for the hidden layer neurons; To improve the weighted input values ​​of neurons in the hidden layer of a BP neural network; The value of the loss function; This represents the number of training samples; Predict the energy consumption per unit time for the i-th sample using the model; This represents the actual energy consumption per unit time for the i-th sample. This is the energy consumption weighting coefficient.

5. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, When defining energy consumption level ranges based on a coupled correlation model, the following applies: Based on the correlation between energy consumption per unit time and changes in temperature and humidity of grain piles output by the coupled correlation model, at least 500 sets of valid sample data were extracted, and the extracted samples covered different external environmental conditions and different grain storage stages. By analyzing the distribution density of sample data in the characteristic space of energy consumption and temperature and humidity changes, the initial grouping center is determined based on the data density abrupt change points in the characteristic space, and the number of groups is set to 3 to 5. Set constraints: The maximum energy consumption per unit time within each energy consumption level range shall not exceed the preset energy consumption control limit. The temperature and humidity changes of the grain pile corresponding to each energy consumption level range all fall within the allowable fluctuation range of the safe storage temperature and humidity standards for the stored grain varieties. During the grouping process, the samples are checked in real time to see if they meet the dual constraints, and only samples that meet the constraints are assigned to the corresponding interval. Finally, the initially formed energy consumption level intervals are iteratively optimized: if an interval does not meet any constraint condition, the deviation between the characteristic value of the interval and the constraint condition is calculated first, and the group center position is dynamically corrected according to the deviation. The above grouping steps are then repeated until all intervals meet the dual constraint conditions.

6. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, When setting the dual-objective control threshold for ventilation effect and energy consumption, the following applies: Determine the basic safe temperature and humidity thresholds for grain storage varieties; Calculate the external environment correction factor , These are the weighting coefficients for temperature difference and humidity difference, respectively. The values ​​all fall within the range of [0.4, 0.6], and The sum is 1; The ambient temperature outside the warehouse; The average temperature of the grain pile inside the warehouse; Humidity of the external environment; The average humidity of the grain pile inside the warehouse; The basic safe temperature and humidity thresholds are corrected to dynamic safe temperature and humidity thresholds, that is, the basic safe temperature and humidity thresholds are respectively compared with... Multiply them to obtain the corrected safe temperature and humidity threshold, which is also known as the dynamic safe temperature and humidity threshold. Determine the basic energy consumption control threshold for the ventilation equipment, initially set at 80% of the equipment's rated power, and set a grain storage time correction coefficient. The grain storage time correction coefficient increases linearly with the increase of grain storage time, with a value range of [1.0, 1.5]. In the initial stage of storage, the grain storage time correction coefficient is set to 1. The basic energy consumption control threshold is then corrected to the energy consumption control threshold, which is also the dynamic energy consumption control threshold. This is the product of the basic energy consumption control threshold of the ventilation equipment and the grain storage time correction coefficient. The dynamic safety temperature and humidity threshold and the dynamic energy consumption control threshold are denoted as the dual-objective control threshold.

7. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, If the data deviates from the dual-target control threshold, the control parameters should be corrected immediately as follows: The deviations are calculated in real time, including: the absolute value of the difference between the actual grain pile temperature and the dynamic safe temperature threshold, the absolute value of the difference between the actual grain pile humidity and the dynamic safe humidity threshold, and the absolute value of the difference between the actual equipment energy consumption and the dynamic energy consumption control threshold. Setting deviation thresholds: Temperature deviation threshold, humidity deviation threshold, and energy consumption deviation threshold are all preset by the user based on grain storage safety requirements and equipment operational stability. When any deviation exceeds the corresponding threshold, parameter correction is initiated. ； In the formula: This is the correction amount for the operating speed of the ventilation equipment; This is the proportionality coefficient; This is the overall deviation. The integral coefficient; These are the differential coefficients; in, ; For the single-run duration of ventilation equipment, pulse width modulation duty cycle adjustment is used: according to Determine the pulse width modulation duty cycle , Indicates the basic duty cycle. This is the duty cycle adjustment coefficient. ∈[0.1,0.3]; Finally, based on Adjust the current operating speed according to Adjust the duration of each run, and continue monitoring the data after correction until all deviations are below the corresponding thresholds.

8. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, The process of dividing the grain storage safety early warning level into two stages, based on the theoretical energy consumption and temperature and humidity dual characterization curves output by the coupled correlation model, calculates the real-time temperature and humidity change rate and the standard deviation of the fluctuation of the actual energy consumption of the equipment within the past preset time interval, and simultaneously calculates the grain storage safety early warning index. ; In the formula: This represents the real-time rate of change in temperature and humidity. The actual average temperature and humidity of the grain pile inside the warehouse at the current moment. Dynamic safety temperature and humidity thresholds; according to Classifying energy consumption anomaly warning levels: When < At the time of safety level, when ≤ < At the time, it was at the attention level, when ≥ It was at the emergency level, among which To preset the warning index threshold, and < ; Further calculate the energy consumption anomaly early warning index , This indicates the actual energy consumption of the equipment. This indicates the dynamic energy consumption control threshold. This represents the standard deviation of the actual energy consumption fluctuation of the equipment within a preset time interval in the past. according to Classifying energy consumption anomaly warning levels: When < When it is in the normal level, when J1≤ < When it is an abnormal level, ≥ It was classified as severe, among which To preset the energy consumption warning threshold, and .

9. The adaptive control method for ventilation and energy consumption linkage based on intelligent grain silos according to claim 1, characterized in that, When the tiered early warning information is pushed out, it is fed back to the corresponding preset matching receiver according to the early warning level.

Images