Grain bin sidewall temperature control system and method based on big data analysis

Abstract

The present application relates to the technical field of grain storage, in particular to a grain storehouse side wall temperature control system and method based on big data analysis, in the present application, the dynamic correlation modeling module is used to integrate computing power, heat and environmental parameters in real time, to construct a multi-dimensional data tensor with space-time alignment, and to generate a dynamic weight parameter set containing environmental interference factors, real-time correlation matrix and historical correlation mode by combining sliding time window analysis, to solve the problem of heat dissipation response lag caused by data isolation in traditional methods; the hierarchical decision module is based on a double time scale framework, the micro layer dynamically generates load migration path and priority weight according to the real-time correlation matrix, to realize instant balancing of computing power in heat-sensitive scenes, the macro layer optimizes the heat dissipation strategy by superimposing historical mode and real-time trend evolution, and periodically fuses environmental interference factors to correct instructions, to realize dynamic collaborative optimization of computing power and heat dissipation, to avoid local overheating and reduce energy consumption, to improve the energy efficiency and operation stability of the data center.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of grain storage, specifically a grain silo side wall temperature control system and method based on big data analysis. BACKGROUND

[0002] The grain silo side wall temperature control system based on big data analysis is a system that uses big data technology to analyze the temperature variation law of the grain pile and maintains the stability of the environment in the grain silo by intelligently regulating the cooling power of the silo wall. The goal is to ensure the safety of grain storage and reduce energy consumption.

[0003] However, this system still has certain challenges, such as how to accurately adjust the cooling power of the silo wall according to the temperature variation trend inside the grain pile to effectively solve the problem of insect infestation caused by local overheating.

[0004] The core of this problem lies in the need to monitor and predict the temperature distribution inside the grain pile, and accordingly achieve dynamic and accurate power distribution and regulation, which may be affected by factors such as data accuracy, algorithm prediction ability, and system response speed. SUMMARY

[0005] The present application aims to solve the problems raised in the background art by providing a grain silo side wall temperature control system based on big data analysis. The specific technical problems include: how to regulate the cooling power of the silo wall according to the temperature variation trend inside the grain pile to solve the problem of insect infestation caused by local overheating.

[0006] To achieve the above-mentioned purpose, the present application aims to provide a grain silo side wall temperature control system based on big data analysis, comprising: a data acquisition module for collecting temperature data through internal sensors of the grain pile; a big data analysis module in communication with the data acquisition module, using big data analysis algorithms to predict the temperature variation trend inside the grain pile; a temperature control area determination module for determining the target temperature control area and cooling demand according to the temperature variation trend; a power calculation module for calculating the power adjustment parameters of the side wall cooling equipment based on the cooling demand; and a device control module for applying the adjustment parameters to the side wall cooling equipment to dynamically adjust the cooling power. The big data analysis module is connected to the power calculation module; based on the formula determining whether the temperature variation trend exceeds the warning range, wherein To predict the temperature variation rate, K1 is the temperature sensitivity coefficient, T target is the expected target temperature, T current is the current collected temperature;

[0007] If not, the power calculation module sets P adjust = P initial × r1, otherwise increases the control sensitivity r2, and raises P adjust to P initial×r2, r1 is the initial power correction factor, and r2 is the power correction factor after the adjustment sensitivity is improved. The equipment control module controls the operation of the side wall cooling equipment accordingly.

[0008] The temperature control zone determination module includes a distance weight calculation unit, which calculates the distance weight W of the target temperature control zone. d =e λ×d d is the distance between the sensor position and the cooling sidewall, and λ is the spatial attenuation coefficient;

[0009] It also includes weighted average units, based on W d Weighted average of temperature sampling data;

[0010] The power calculation module utilizes P adjust ≥K2×ΔT×W d The power adjustment parameters are calibrated, K2 represents the power correction coefficient, ΔT is the target temperature offset, and the equipment control module sets differentiated cooling power adjustment for different distance areas.

[0011] The big data analytics module includes a temperature gradient calculation unit, which uses formulas... Determine the dynamic temperature gradient, G T T represents the temperature gradient value. top and T bottom These represent the temperatures at the upper and lower sampling points of the grain pile, respectively, with ΔH being the vertical height difference.

[0012] When G T Threshold G At that time, the equipment control module executes a zoned and layered temperature control strategy, increasing the local cooling power to P. zone =Base P +G T ×P factor And set the power limit for each partition; where Threshold G Base is the temperature gradient threshold. P Based on power, P factor This is the power influence factor.

[0013] The equipment control module includes a priority allocation unit, which prioritizes cooling needs for different zones and allocates priority to temperature control units N that meet temperature warning conditions. i If (T) max -T min K3×Tavg, prioritize adjusting the cooling module in the maximum power zone, and apply the special mode power Max(P) i ×α), α=5, Update And redistribute the remaining power to ensure energy balance;

[0014] Among them, T max and Tmin Temperature control unit N i The highest and lowest temperatures inside; K3 is the temperature difference coefficient; Tavg is the temperature control unit N. i Average temperature inside; P i Temperature control unit N i Cooling capacity; Max P The maximum power allowed by the system; i is the number of the temperature control unit, ranging from 1 to N; N is the total number of zones.

[0015] The big data analytics module includes a risk assessment unit, which uses a time series forecasting model to assess the overheating risk R. overheat The mean square error (MSE) is used to measure the goodness of fit of historical temperatures (t+1). When the MSE exceeds the preset level, the power calculation module... Adjust power parameters to optimize power consumption performance;

[0016] Power New The adjusted new power; Power Base Refers to the reference cooling power; K4 is an empirical calibration constant, ε =

[0017] 1×10 -6 Used to prevent the denominator from being zero; MSE t Let be the mean square error at time t.

[0018] The big data analytics module includes an anomaly detection unit that detects anomalies by measuring the slope of temperature changes. err <Threshold s ×log(σ), where σ is the standard deviation of the predicted fluctuation, the equipment control module freezes the current power value, and combines feedback closed-loop correction to adjust the cooling effect, according to Recovery. pows =min(Cap) Max C0 powers +β×σ slope Power recovery;

[0019] Among them, slope err Threshold is the slope error of the temperature change. s The slope error threshold is β; the power ramp speed is controlled by β, and Cap is used for the ramp speed. Max This is the upper limit of power; C0 powers The power value at the time of freezing; σ slope This represents the standard deviation of the slope of the temperature change.

[0020] The equipment control module includes a compensation mechanism unit and a power planning unit; the compensation mechanism unit addresses situations where localized overheating persists for an extended period without recovery, based on... To activate the compensation cooling module, press E.add =Comp Ratio ×P zone Increased energy consumption;

[0021] Among them, t stay The duration of the localized overheating state; T cycle γ is the system monitoring cycle; γ is the compensation start threshold; Comp Ratio To compensate for the power ratio, P zone This refers to the cooling power of a localized area.

[0022] The power planning unit pre-plans dynamic power curves to account for environmental fluctuations, using a piecewise polynomial representation. Each part corresponds to a different operating condition. The specific calculation logic is as follows: Each order coefficient corresponds to the influence of different control variables; finally, the power output at each moment is combined into a complete operating diagram; where, C k The calculated power value at time k; a k p represents the polynomial coefficients; t represents time; p represents the polynomial coefficients. k The order is the polynomial.

[0023] It also includes an efficiency evaluation module, which connects to the equipment control module and optimizes through cost-saving functions. The long-term economic efficiency of the comprehensive evaluation system is as follows: J energy (i,j) represents the actual electricity consumption record value of the j-th refrigeration device within the i-th time period; Total Eff η represents the overall system efficiency index, and η is the weight of energy-saving gain.

[0024] The method for a grain silo sidewall temperature control system based on big data analysis includes the following steps:

[0025] S1. Based on temperature data collected by sensors inside the grain pile, big data analysis is used to predict the temperature change trend inside the grain pile.

[0026] S2. Based on the predicted temperature change trend, determine the target temperature control area and the corresponding cooling demand;

[0027] S3. Based on the cooling demand, calculate the adjustment parameters of the side wall cooling equipment power.

[0028] S4. The adjustment parameters are applied to the sidewall cooling equipment to dynamically adjust the cooling power, suppress local overheating, and prevent pest infestation. Compared with the prior art, the beneficial effects of the present invention are:

[0029] The granary side wall temperature control system based on big data analysis can dynamically and accurately adjust the power of the warehouse wall refrigeration equipment according to the real-time and predicted temperature change trend of the grain pile, avoid local overheating, and create a stable temperature environment for grain storage. Precise temperature regulation controls the internal temperature of the grain pile within a range that is not suitable for insect growth and reproduction, reducing the risk of insect breeding caused by local overheating from the root, and ensuring the safety of grain storage. The efficient interaction of data and fast operation of each module of the system overcomes data processing delay, significantly improves the response speed to temperature changes in the grain pile, and quickly adjusts the refrigeration power of the equipment control module once the temperature of the grain pile fluctuates abnormally, timely intervening the temperature of the grain pile. With the help of the efficiency evaluation module to evaluate the economy of long-term operation of the system and the power planning unit to reasonably plan the power under different working conditions, the refrigeration power is allocated on demand, effectively reducing the overall energy consumption while ensuring the safety of grain storage, saving operating costs, and having both economic and environmental benefits. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 M flowchart of the present application;

[0031] Figure 2 M schematic diagram of the present application. DETAILED DESCRIPTION

[0032] In order to more clearly understand the purpose, technical scheme and advantages of the present application, the present application is described and explained below in conjunction with the drawings and examples.

[0033] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the same meaning as those commonly understood by a person of ordinary skill in the art to which the present application belongs. In the present application, the terms "one", "a", "an", "the", "these", and the like do not denote the number of quantities of the objects; they can be singular or plural. In the present application, the terms "include", "contain", "have" and any variants thereof are intended to cover non-exclusive inclusion; for example, a process, method and system, product or device containing a series of steps or modules (units) are not limited to the listed steps or modules (units), but can include steps or modules (units) not listed, or can include other steps or modules (units) inherent to the process, method, product or device. In the present application, the terms "connected", "connected", "coupled" and the like do not limit to physical or mechanical connection, but can include electrical connection, whether direct or indirect. In the present application, "multiple" means two or more. The term "and / or" describes the relationship between the associated objects, which means that there can be three relationships, for example, "A and / or B" can mean that A exists alone, A and B exist together, and B exists alone. In general, the character " / " represents the "or" relationship between the front and rear associated objects. In the present application, the terms "first", "second", "third" and the like are only used to distinguish similar objects, and do not represent a specific order of the objects.

[0034] Next, please refer to Figure 1 , describes the present application based on big data analysis of grain silo side wall temperature control system and how to realize the regulation and control of side wall refrigeration power to solve the problem of local overheating leading to pest breeding. The whole system includes several steps: data acquisition and processing, trend prediction, target temperature control area and refrigeration demand determination, refrigeration equipment power parameter calculation, and dynamic adjustment of side wall refrigeration power. One of the purposes of the present embodiment is a kind of based on big data analysis of grain silo side wall temperature control system, including: data acquisition module, for collecting temperature data through grain pile internal sensor;First, the first step is to collect temperature data based on grain pile internal sensor and process. In this process, a large number of temperature sensors are arranged at different depths and different positions of the grain pile, and the temperature information of each node is recorded at regular intervals. The obtained data is stored in the database after cleaning and standardization. This multi-dimensional and real-time acquisition method can fully grasp the situation of grain storage environment. For example, a set of sensor array is set every 10 cubic meters in the grain warehouse, each set containing three different height temperature sensors at upper and lower horizontal planes. These sensors upload the measured data to the big data platform in the centralized controller through the wireless network, providing detailed and reliable basis for subsequent data analysis;

[0035] The big data analysis module is in communication connection with the data acquisition module, and predicts the temperature change trend inside the grain pile by using a big data analysis algorithm. Secondly, the temperature change trend inside the grain pile is predicted by using a big data technology, which is the second step. This operation depends on a machine learning algorithm such as a random forest regression or a long short-term memory (LSTM) network model. The system trains a precise trend prediction model by using a large amount of historical data sets, can estimate the temperature values that may be reached in each region in a short time in the future, and finds the risk area where an abnormal high-temperature point may appear. Specifically, a deep neural network structure with a time sequence label is used for fitting operation, so as to predict the trend of local heat aggregation in a short time. In this way, it can help the operation and maintenance personnel to predict the risk position in advance, and achieve active prevention rather than only passive response to sudden problems. In one embodiment, if the daily average temperature rise of the grain layer in a certain place exceeds 0.3℃ for three consecutive days, the system will automatically mark it as a key object for attention, and prepare to execute a more stringent cooling control plan;

[0036] The temperature control area determination module determines the target temperature control area and the refrigeration demand according to the temperature change trend. The power calculation module calculates the side wall refrigeration equipment power adjustment parameter based on the refrigeration demand. Then, the third step is to determine the target temperature control area and the corresponding refrigeration demand according to the prediction result. This step combines the properties of the grain itself and the physical conditions of the warehouse to determine which places indeed have a high possibility to generate pest risk. If some local area is predicted to reach the optimal temperature range for promoting the development and growth of pests, it is necessary to develop a target to reduce the actual operating temperature of these hot spot areas. For example, when the predicted temperature of a specific area is close to 15℃-20℃, which is the key temperature interval for the activity of many grain insects to enhance and start to multiply, at least three to five degrees Celsius should be reduced to block the development chain of the reproduction process. In addition, it is also necessary to quantify the required refrigeration power intensity to match the expected cooling effect;

[0037] The device control module applies the adjustment parameter to the side wall refrigeration equipment and dynamically adjusts the refrigeration power. Then, the fourth step is to calculate the specific side wall refrigeration equipment power adjustment value as the core operation instruction according to the refrigeration demand proposed in the last link. This involves accurate modeling simulation calculation and intelligent optimization solution method to generate the most suitable equipment power consumption adjustment suggestion for the current state change demand. For each target area to be processed, the influence of external meteorological conditions and the performance limitations of existing infrastructure and other multiple constraints are considered to find the best combination strategy to achieve the ideal energy consumption proportion while ensuring the performance meets the standard. For example, when a higher warning level occurs in a part close to the outer facade, the corresponding cold air pipe output is appropriately increased, but it will not be greatly overloaded to prevent energy waste or other unnecessary losses; at the same time, it ensures that the remaining parts that maintain a low temperature state are not excessively cooled;

[0038] Finally, after determining all necessary adjustments, the formal change settings are applied to the on-site physical hardware facilities, which is the fifth step of dynamically adapting the side wall refrigeration unit operating conditions. Through the pre-programmed automatic control module, real-time updated driving command is transmitted to the relevant associated devices to execute task change instructions. Finally, it ensures that the overall internal space maintains a reasonable and stable distribution of cool micro-environment to prevent local over-high temperature from causing bacteria or parasites to spread. For example, in the actual application scenario, if the northeast corner detects a significant deviation from the reference value and has been determined to have a high risk of harm and needs to be addressed, the small fan in that direction is immediately notified to increase its speed by one level and the associated heat sink plate opening angle is increased to enhance heat absorption and removal efficiency until it is confirmed that the balance has been restored.

[0039] Therefore, the above process fully demonstrates how to complete the scientific and fine control scheme design and implementation path for the grain warehouse based on the big data driving concept to solve the aforementioned problems: according to the temperature change trend of the grain pile, the refrigeration power of the warehouse wall is controlled, thereby successfully dealing with and eliminating the potential pest hazards caused by temperature rise.

[0040] Next, the grain warehouse side wall temperature control system based on big data analysis of the present application is described. The operation of the system is divided into the following steps: first, use the formula to judge whether the temperature change trend exceeds the warning range; second, set or adjust the refrigeration power parameter according to the judgment result; third, control the operation of the side wall refrigeration equipment according to the adjusted power.

[0041] Next, the grain warehouse side wall temperature control system based on big data analysis of the present application is described. The operation of the system is divided into the following steps: first, use the formula to judge whether the temperature change trend exceeds the warning range; second, set or adjust the refrigeration power parameter according to the judgment result; third, control the operation of the side wall refrigeration equipment according to the adjusted power.

[0042] In the first step, the formula is used to judge the temperature change trend. Among them, represents the predicted temperature change rate, which can be calculated by the data analysis module in the system based on historical and real-time temperature data. K1 is the temperature sensitivity coefficient, which determines the reaction sensitivity of the warning limit to the temperature difference. The value range is usually 0.05 to 0.2 (expressed as the inverse of the temperature change per unit time), and the preferred value is 0.1. The specific value is determined according to the control temperature accuracy requirement. target T is the target temperature, which is the ideal environmental temperature preset according to the characteristics of the stored grain. T current is the current actual temperature value, which is obtained from the temperature sensor installed on the side wall. By comparing the size of the left temperature change rate and the right threshold value, it is determined whether to trigger the control process.

[0043] If the first step condition is not met, enter the initial control stage in the second step, adjust the power regulation parameter P adjust to P initial ×r1. Here P initial refers to the basic working power of the refrigeration device, and r1 is used as a correction factor to reduce the possibility of overreaction. Generally, the recommended value of r1 is between 0.75 and 0.9, so as to maintain stable refrigeration power output and avoid resource waste caused by frequent large-scale adjustment.

[0044] When the temperature exceeds the warning range, a higher sensitivity control mode is executed, and the control sensitivity is increased from the default value to a new proportion factor r2, and the adjusted power is set as P adjust = P initial ×r2. At this time, the value of r2 should be greater than r1, in the range of 1.2 to 1.5, for example 1.4, in order to quickly reduce the temperature and suppress the risk of high temperature. This step can quickly change the working state of the refrigeration side wall device to match the actual required intensity.

[0045] In one embodiment, assuming that the wheat seeds stored in the warehouse need to be kept at an ideal storage environment of 15°C for a long time, and the current detection finds that the temperature of a certain side wall has risen to 18°C, and the system estimates that it will further rise in the next few hours, then according to the above formula, the condition is true, and the refrigeration intensity of the device is automatically increased until the area returns to a balanced state. Specifically, if the preliminary assessment shows that the temperature will climb at +0.5°C per hour, and the estimated temperature difference exceeds the warning line, the refrigeration efficiency of the corresponding side wall area will be immediately increased to 1.4 times the basic value to deal with the emergency. Next, the distance weight calculation, temperature data weighted average target value acquisition, and power regulation parameter calibration method in the grain warehouse side wall temperature control system based on big data analysis of the present application are described.

[0046] First, calculate the distance weight W d of the target temperature control area. The formula is W d = e λ×d , where d is the physical distance between the sensor and the refrigeration side wall, which is a non-negative real number; λ is the spatial attenuation coefficient, usually selected in the interval [0.1, 1.0], and the optimal value is 0.5. The significance of this formula is to simulate the temperature influence weakening effect caused by distance in the heat transfer process. The farther the distance, the lower the weight value. Therefore, this formula can more scientifically evaluate the importance of temperature data.

[0047] Second, according to the obtained W dThe target value is obtained by weighted averaging of all temperature sampling data within the target temperature control area. This step multiplies the temperature values ​​recorded by all sensors by their corresponding distance weights and then performs a comprehensive calculation to obtain a more accurate target temperature reference value. For example, if multiple sensors are distributed within an area, and some sensors are close to the cooling sidewall and W... d If the value is larger, the measured temperature data will be given greater weight, thereby increasing its influence on temperature control strategies in key locations.

[0048] Third, in the power regulation parameter P adjust Apply constraint P to the above adjust ≥K2×ΔT×W d This is used to adjust the working intensity of the cooling equipment. K2 represents the power correction coefficient, and according to the test results, its reasonable value is within the range of [0.5, 2.0], with a preferred value of 1.0. The purpose of this step is to adjust the cooling output based on the difference ΔT between the current temperature and the desired temperature, and to clarify the minimum power requirement for a specific location by combining the previously calculated distance weights, so as to avoid resource waste or excessive energy consumption.

[0049] In one embodiment, assume five temperature sensors are installed on the side wall of the grain silo, numbered S1 to S5 from closest to furthest. Specifically, S1 is 0 meters away, and the others are spaced two meters apart. Setting λ = 0.5, the distance weight of each point can be obtained: e 0.5×0 =1.0 (for S1), e 0.5×2 ≈0.368 (for S2). Subsequently, if the instantaneous temperature data recorded by these sensors are {24, 23, 26, 28, 30}℃, the ideal target control temperature that accurately reflects the actual situation can be obtained by weighting and normalizing according to the aforementioned weights. Finally, the above constraints are used to configure the power value required for each corresponding refrigeration unit to ensure that different parts of the grain silo can efficiently maintain the ideal state without losing uniformity. Next, the method of determining the dynamic temperature gradient and implementing the zoned and layered temperature control strategy of the present invention is described. First, the steps are listed, including calculating the temperature gradient value, determining whether the temperature control strategy needs to be activated, and setting the local refrigeration power and distance threshold to define the zone, and the specific meaning of each step is explained in detail later.

[0050] The first step is to use the formula This is used to determine the temperature gradient between sampling points at the top and bottom of the grain pile. In this formula, T... top The real-time monitored temperature (in °C) represents the top area of ​​the grain pile, generally ranging from 0 °C to 40 °C, with the specific value depending on the type of stored grain and seasonal variations; T bottomtemperature measured at the bottom region of the grain pile (unit: ℃); and ΔH represents the vertical height difference between the two temperature measurement points (unit: meters). The role of this formula is to calculate the degree of temperature change in the vertical direction of the grain pile, i.e., the temperature gradient, so as to assess whether there is a significant temperature difference leading to uneven storage environment. The second step is to determine whether to implement zoned and layered temperature control according to the set threshold Threshold G . If the calculated temperature gradient value G T is greater than the set threshold Threshold G , it means that there is a risk of large vertical temperature difference in the current grain storehouse, and therefore corresponding zoned management measures need to be taken to further refine the optimization operation of each independent temperature control zone.

[0051] When the condition G T > Threshold G is met, the third key stage of adjusting the parameters of the refrigeration system is entered. In this stage, the refrigeration power is gradually increased to effectively respond to the sudden local temperature rise by using the formula P zone = Base P + G T × P factor . Here, Base P refers to the basic power reference, which is set by default as a basic value suitable for most general grain storage modes (in the range of 1-5 kW), and the factor P factor represents a correction term introduced in response to different scene requirements, ensuring that the actual adjustment is flexible, efficient, and energy-saving, and the optimal choice is determined based on historical big data analysis.

[0052] The last part is to clearly define the standard for dividing zones: a fixed value distance threshold Threshold D = 5m is used to determine which physical spaces are classified into the same control category. Based on this, each independent unit is individually assigned a personalized maximum output power limit value, ensuring that each zone can accurately respond to the control target according to its own load condition, while maintaining maximum energy utilization efficiency performance.

[0053] In one embodiment, assume that a large intelligent grain storehouse using a big data analysis-based sidewall cooling scheme encounters an abnormal situation, for example, the monitoring shows that the two sensing probes located near the window report temperature data T top = 32℃ and T bottom = 18℃ in the morning, and the height difference ΔH = 4m. By substituting the formula, the result can be quickly obtained, i.e., there is a quite serious hot field deviation (G T= (32-18) / 4 = 3.5° / m). Since the alarm threshold has been calibrated in advance to Threshold_G = 3° / m, it is confirmed that the alarm line must be exceeded immediately to intervene in the processing. Then, combined with the preset algorithm, the additional power supply P zone = 3 + (3.5*0.8) = 6kW (where Base P = 3KW, P factor = 0.8 as the best balance point verified by experiments). At the same time, in the field of spatial structure segmentation, any interval less than five meters is considered as a continuous and unified processing block according to the rules. Finally, the entire complex process is successfully completed, so that the specific affected area returns to the stable working interval and achieves the maximum vision goal of energy saving and emission reduction.

[0054] Next, the priority allocation and power setting steps in the grain silo sidewall temperature control system based on big data analysis of the present application are described. First, the priority is allocated according to the cooling demand of each partition. This process determines the actual demand of different temperature control units according to the data obtained by big data analysis, and determines the high-risk areas that need to be processed in priority by comparing the regional temperature data with the preset warning threshold through algorithm. The higher the priority of the temperature control unit, the higher the weight it will be given in energy allocation to ensure that the key area is stable in the appropriate range.

[0055] Second, determine the maximum temperature difference condition to apply special mode power adjustment rules. The formula (T max -T min )>K3*Tavg means that when the difference between the highest temperature T max and the lowest temperature T min in the monitoring point exceeds the allowed temperature difference calculated by the coefficient K3 multiplied by the current average temperature Tavg, the special mode power adjustment logic is entered. For example, if it is detected that the temperature difference exceeds the standard, a special power factor α = 5 is enabled to adjust the maximum power of the corresponding area.

[0056] Third, update the power setting value according to the formula. For each temperature control unit i, the formula is used to update the base power value of the individual unit, where N represents the total number of all sub-temperature control units. Specifically, the purpose of this formula is to adjust the dynamic response speed of the area by introducing an exponential factor, and reasonably plan the energy consumption of the cold source equipment.

[0057] Fourth, ensure that the remaining available power can be redistributed to complete the overall energy efficiency optimization and balance goal of the system. This process dynamically evaluates and compensates the real-time power consumption of each area based on big data statistics.

[0058] In one embodiment, if the temperature of the second partition is abnormal and the special condition (T max- T min K3 x Tavg, where the optimal K3 value is set to 0.1, indicating that at this time the power boost ratio a = 5 is started to rapidly reduce the maximum temperature difference area, and in other areas, P i After assigning the base power value to the formula, the resource utilization efficiency is re-optimized to ensure that the entire system is in an efficient state.

[0059] Next, the optimization steps and working principles of the grain silo sidewall temperature control system based on big data analysis of the present application are described. Specifically, the following steps and meanings are included:

[0060] The first step is to collect and process historical data to build a time series prediction model to assess the risk of overheating R overheat This step requires detailed statistical analysis of the past stored temperature change curve. The time series model uses mean square error (MSE) as an indicator to measure the fitting accuracy of historical data.

[0061] The second step defines a dynamic adjustment of the cooling power Power New The formula is expressed as where Power Base is the initial reference power value of the refrigeration system, which is generally in the range of [10W, 100W], and its optimal value can be adjusted according to the actual load, which is 50W; K4 is an empirical calibration constant, which is usually selected in the range of [1, 10] to adapt to different environmental requirements, and the recommended optimal value is set to 5.6, and ε is set to a small amount of 1 x 10 -6 to ensure that the denominator does not tend to zero during calculation, thereby causing numerical fluctuations or non-convergence problems.

[0062] The third step determines whether the current temperature control strategy needs to be adjusted. That is, when the calculated MSE reaches or exceeds the predefined threshold, the power re-estimation process is triggered.

[0063] In one embodiment, for example, assume that the temperature prediction MSE(t+1) = 12 in the current monitoring period, and the power consumption optimization is performed using the aforementioned formula. Since the preset maximum tolerance error limit is set to 10 (i.e., if MSE > 10, the correction mechanism is started), the updated cooler running power The reason for this setting is that as the historical temperature fitting error increases, it reflects that the model's grasp of external factor changes has decreased, so the actual output power should be relatively weakened to save energy consumption.

[0064] The above method achieves precise control of the heat level in the grain storage environment while effectively reducing resource waste.

[0065] Next, the grain silo sidewall temperature control system based on big data analysis of the present application is described. The key steps are as follows:

[0066] The first step is to detect abnormal points according to the temperature change slope to refine the local refrigeration regulation. This step is achieved by the formula slope err <Threshold s ×log(σ) to determine whether to freeze the current power value. Wherein, slope err represents the error between the temperature change slope and the expected value; Threshold s is the error tolerance threshold, which is a parameter derived from historical data and system experience, generally ranging from 0.1 to 1.5, and the optimal value is 0.5; σ is the standard deviation of the predicted fluctuation, indicating the uncertainty of the temperature prediction model. The purpose of this formula is to identify abnormal temperature change areas through dynamic assessment of temperature fluctuation characteristics and take timely measures. Only when the error exceeds the set range, the system will activate the frozen power control module. Freezing power can prevent further temperature imbalance. For example, the sensor feedback of a certain side of the silo shows that the slope error is significantly higher than the average standard deviation, that is, it enters the freeze mode to avoid excessive adjustment affecting temperature balance.

[0067] The second step is to combine feedback closed-loop correction to refrigeration effect and ensure at least three consecutive periods of stability before recovery. This is achieved by monitoring the temperature change curve and continuously fine-tuning the refrigeration machine output using a feedback mechanism. The consecutive period in this process is a time window length defined by the system, usually set to ten minutes or longer. This approach can effectively eliminate false judgments caused by temporary disturbances, such as temporary temperature rise disturbances caused by the start of ventilation equipment, which will not cause power out-of-control adjustment. In one embodiment, when the system is frozen, the signal value after three consecutive stable periods is automatically recorded to allow unlocking, thereby achieving more reliable regulation.

[0068] The third step involves determining the power recovery method from the frozen state to the normal operating state. This is achieved by the formula Recovery pows =min(Cap Max ,C0 powers +β×σ slope ). Specifically, parameter C0 powers refers to the initial power level of the refrigeration equipment in the frozen state, Cap Max is the upper limit of the refrigeration system's capacity (in kilowatts) to prevent overload; and β is a key factor controlling the power ramping speed, typically taking a value between 0.1 and 0.5, with an optimal value of 0.25; σ slopeis the predicted variance of temperature difference slope, which is used to quantify the adjustment risk caused by uncertain factors. Through such a setting, not only the safety of gradually restoring normal operation can be guaranteed, but also the impact on the entire cold chain environment can be effectively reduced. Specifically, during the period of local cooling failure in the grain depot, assuming that the initial refrigeration power C0=3kw and the predicted standard deviation is 4, the power recovery will slowly climb to a new level while not exceeding the upper limit of the capacity (such as 6kw). Therefore, the purpose of this formula is to smooth the switching process.

[0069] Next, the specific steps of setting a compensation mechanism for local area overheating in the grain depot side wall temperature control system based on big data analysis of the application are described. First, the residence time parameter and the determination formula are introduced; second, the calculation method of additional energy consumption and the setting rules of compensation power are defined; finally, the application process and its rationality in the actual scene of the grain depot are illustrated with examples. Specifically, the following steps are included: the first step is to monitor and record the abnormal temperature situation of the local area, and according to the residence time t stay determine whether the area has a sustained high temperature state. Here, t stay refers to the time length from the first occurrence of overheating to the recovery of the area to normal, which is generally in the range of [0, T total ](T total is the upper limit time of a single cooling cycle). The second step is to use the determination condition to decide whether to start the compensation refrigeration module. Among them, T cycle is the standard temperature control period, and γ is the threshold coefficient (generally in the range of 0.5-0.8), which is used to determine whether the overheating state is a short-term disturbance or a long-term uncontrollable situation. Through such condition setting, it can be distinguished whether the abnormal overheating is accidental or systematic problem.

[0070] The third step is to calculate the additional energy consumption according to the formula E add = Comp Ratio × P zone if the compensation mechanism is triggered, which is used to compensate for the consumption of insufficient refrigeration capacity in the local area. The parameter meanings here are as follows, Comp Ratio is the compensation power ratio factor, and the default value is set to 15 (the optimal value range is usually between 10 and 20, which can be adjusted according to historical data analysis); and P zone represents the basic refrigeration power demand of the specific area. In this way, the degree of energy consumption increase can be dynamically quantified, making the resource allocation more rational. The purpose of setting this formula is to minimize the impact of local problems and ensure the overall system performance while minimizing the excessive loss of total energy consumption.

[0071] For example, in one embodiment, assume that the temperature of a certain grain storage warehouse side wall is significantly increased due to external heat conduction, and exceeds the set safety limit for two consecutive days, with a residence time of 48 hours. At this time, by the formula (assuming γ = 0.6), it can be determined that additional refrigeration devices need to be activated immediately for intervention. The additional energy cost during compensation is described by E add = 15% * basic power consumption 100 kW = 15 kW. This case specifically demonstrates how to use the compensation mechanism to optimize system operation stability.

[0072] Next, the dynamic power curve planning method of the present application is described. This method aims to optimize the operation efficiency of the grain warehouse side wall temperature control system based on big data analysis, and adapt to environmental fluctuations within a certain time period.

[0073] Specifically, the following steps are included: determining the time period to be controlled and the corresponding environmental variable characteristics; obtaining the control logic expression under each working condition according to data modeling, which is described using piecewise polynomials; calculating and calibrating the polynomial parameter values for each stage; and finally integrating all the power outputs to form a complete operation graph for actual operation.

[0074] First, the time period and environmental fluctuation characteristics are preliminarily identified and classified. This step ensures that different time periods use corresponding control measures to more accurately match the effects caused by day-night or seasonal changes, such as significant temperature fluctuations. Through big data analysis, data for these time periods are collected in advance to establish judgment criteria.

[0075] Then, the control function is established, where C k represents the target power to be achieved within the kth time period, with t as the time base. Each order coefficient, such as a k , can be regarded as a measure of the strength of different control conditions; for any p k , it indicates the time weight relationship of the corresponding variable, and generally selects integer values from 1 to 5 as a reasonable order range, which can better balance the contradiction between precision requirements and implementation difficulty, and reduce unnecessary complexity. The purpose of this formula is to simulate the system behavior trend after considering the interactive effects of multiple factors and to design the optimization path. After that, the key parameters in the above formula are corrected and verified according to the actual situation to ensure that the actual application effect reaches the best state, and the process is adjusted repeatedly until the expected performance target is met.

[0076] For example, consider a scenario where daytime light enhancement causes rapid internal temperature rise, requiring rapid increase in the operation of the side wall cooling device, and at night, the response is reduced to avoid unnecessary energy consumption. Therefore, the daytime part of the curve can be set to resemble a binomial or three-way high-order form to reflect the requirement for rapid response; the corresponding low-amplitude fluctuation at night may only require a first-order approximation to express its simple and stable characteristics to meet the accuracy requirements.

[0077] In one embodiment, such as a particular summer month with significant increase in daylight hours, the distribution of higher than quadratic power components is increased during the day to enhance dynamic adjustment, while the low-order linear decreasing model is mainly set at night to maintain a stable energy-saving trend.

[0078] Next, the steps of the present application are described. The first step is to extend the refrigeration efficiency evaluation system to include long-term operation economic analysis. In this step, the existing refrigeration efficiency evaluation index needs to be improved, and the economy of the system is considered as an important consideration. This extension helps to comprehensively evaluate the performance of the grain silo side wall temperature control system in the entire life cycle, not just the short-term energy consumption.

[0079] The second step involves setting the form of the cost-saving optimization function. The formula is expressed as: where J energy represents the value of the recorded electric energy consumption per actual operation, which is usually determined by the power consumption and the unit price of electricity, and its value range can be dynamically adjusted according to the local power pricing policy; Total Eff refers to the effective refrigeration capacity or heat removal performance of the overall system, which takes a positive value and the larger the value, the more efficient the system; η is the energy saving gain weight, which is set according to the specific energy saving goal, and the optimal value depends on the balance between economic returns and environmental goals. By introducing the formula form, the energy use and energy saving results of the system are comprehensively balanced to ensure that cost minimization becomes the core driving force.

[0080] The third step requires actual benefit analysis through the above optimization framework. For example, in one embodiment, assume that a certain grain storage system needs uninterrupted cooling throughout the year to protect storage quality, and it is known that the average daily power consumption is 100 kWh and the electricity price is 0.5 yuan / kWh, then the direct electricity bill for a year reaches nearly 20,000 yuan. Specifically, in the comparison test before and after optimization, the use of big data model to adjust the air supply strategy can reduce the daily peak load by 30%, while maintaining temperature fluctuations of less than ±1°C, verifying that the guidance of the formula significantly improves economic benefits and environmental friendliness.

[0081] These steps enable the grain silo side wall temperature control system based on big data analysis not only to achieve precise temperature control function, but also to improve economic benefits and social responsibility in the entire operation cycle from a long-term perspective.

[0082] Please refer to Figure 2 The second purpose of the embodiment is to provide a grain silo side wall temperature control system based on big data analysis, which comprises the following method steps: the overall steps of the method start from data collection, and the temperature distribution and change trend data of each region inside the grain pile are obtained in real time through sensors arranged inside the grain pile. S1, based on the temperature data collected by the sensors inside the grain pile, the temperature change trend inside the grain pile is predicted by using big data analysis; these sensors can be densely distributed to ensure that the entire grain pile space is covered, providing detailed data support for subsequent analysis.

[0083] S2, according to the predicted temperature change trend, determine the target temperature control area and the corresponding cooling demand; then, the system enters the data analysis stage, by inputting the collected temperature data inside the grain pile into the pre-established big data analysis platform, using machine learning algorithm and other advanced data analysis techniques to predict the future possible temperature fluctuations and abnormal conditions inside the grain pile, especially the risk area of local overheating. The prediction result further guides the demarcation of the target temperature control area, and according to the potential high temperature risk assessment of the corresponding area, the cooling demand is required. This part is the important core of solving the problem, which determines which area can reach the environment suitable for pests breeding in a short time in the future according to the temperature change trend.

[0084] S3, based on the cooling demand, calculate the adjustment parameters of the side wall cooling equipment power; based on the above-mentioned predicted cooling demand information, the system will further calculate the specific adjustment parameters for accurate control of the power of the cooling equipment located on the wall. For example, when it is found that a certain layer or partition has a higher temperature rise probability, the system will automatically increase the running efficiency or power of the cooling device installed on the side wall near the place; on the contrary, the power is appropriately reduced to avoid energy waste and waste of resources caused by overcooling or influence on the quality of grain.

[0085] S4, apply the adjustment parameters to the side wall cooling equipment to dynamically adjust the cooling power, suppress local overheating and prevent pests breeding. In the final step, the control system outputs the adjustment instruction and sends the aforementioned adjustment parameters to the actual executing mechanism, i.e. different cooling modules or units installed on the four walls of the silo body, dynamically adjusts the working state of each related module, so that the side wall produces appropriate cooling capacity to suppress any adverse heat accumulation conditions that may cause pest problems, and effectively prevent the pest breeding threat induced by such local overheating phenomenon. This method realizes intelligent and accurate management, greatly improves the energy saving level and optimizes the quality and safety of stored grain.

[0086] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only preferred examples of the present application and are not intended to limit the present application. Various changes and improvements can be made to the present application without departing from the spirit and scope of the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A big data analysis based temperature control system for grain silo side walls, characterized in that, The application relates to a grain storage temperature control system and method. The system comprises: a data acquisition module for collecting temperature data from internal sensors of the grain pile; a big data analysis module in communication connection with the data acquisition module, which uses big data analysis algorithms to predict the temperature change trend in the grain pile; a temperature control area determination module for determining a target temperature control area and refrigeration demand according to the temperature change trend; The temperature control region determination module comprises a distance weight calculation unit configured to calculate a distance weight of a target temperature control region , is a distance between the sensor position and the refrigeration side wall, is a spatial attenuation coefficient; and further comprises a weighted average unit configured to perform a weighted average on the temperature sampling data according to ​ The power calculation module utilizes calibrating the power regulation parameter, representing a power correction coefficient, is a target temperature offset, and the device control module sets differentiated refrigeration power regulation for different distance regions. The big data analysis module comprises a temperature gradient calculation unit, which is calculated by the formula determines the dynamic temperature gradient, is the temperature gradient value, and respectively are the temperatures of the sampling points at the upper and lower parts of the grain pile, is the vertical height difference; When , the device control module executes a partition layered temperature control strategy, increases the local refrigeration power to , and sets a partition power upper limit; wherein, is a temperature gradient threshold, is a basic power, is a power influence factor; The equipment control module includes a priority allocation unit, which prioritizes cooling needs for different zones and allocates priority to temperature control units that meet temperature warning conditions. ,like Prioritize adjusting the cooling module in the highest power zone and apply special power modes. , ,renew And redistribute the remaining power to ensure energy balance; wherein, and are the maximum temperature and the minimum temperature in the temperature control unit respectively; is the temperature difference coefficient; is the average temperature in the temperature control unit ; is the refrigeration power of the temperature control unit ; is the maximum power allowed by the system; i is the number of the temperature control unit, which ranges from 1 to N; N is the total number of areas. The big data analytics module includes a risk assessment unit, which uses a time series forecasting model to assess overheating risk. The mean square error (MSE) is used to measure the goodness of fit of historical temperatures (t+1). When the MSE exceeds the preset level, the power calculation module... Adjust power parameters to optimize power consumption performance; for the adjusted new power; denotes the reference cooling power; is an empirical calibration constant, for preventing the denominator from being zero; is the mean square error at the instant The big data analysis module comprises an anomaly detection unit, which detects an anomaly point through a temperature change slope, and if , is the standard deviation of the predicted fluctuation, the device control module freezes the current power value, combines feedback closed-loop correction to refrigeration effect, and restores power according to ; wherein, is the temperature change slope error, is the slope error threshold; control the power ramping speed, is the power upper limit; is the power value during freezing; is the standard deviation of the temperature change slope; a power calculation module for calculating side wall refrigeration equipment power adjustment parameters based on the refrigeration demand; The compensation mechanism unit determines to start the compensation refrigeration module based on the fact that the local overheating has not been recovered for a long time determines to start the compensation refrigeration module, and the compensation refrigeration module is started increases energy consumption wherein, is a duration of the local area overheating state; is a system monitoring period; is a compensation start threshold, is a compensation power ratio, is a refrigeration power of the local area; The power planning unit pre-plans the dynamic power curve for environmental fluctuations, adopts a segmented polynomial expression, each part corresponds to a working condition switching, the specific calculation logic is , each order coefficient corresponds to the influence of different control variables; finally, the power output at each time is combined into a complete operation diagram; wherein, is the power calculation value at ; is the polynomial coefficient; is the time; is the polynomial order; the efficiency evaluation module is connected with the device control module, and the cost saving optimization function is used to comprehensively evaluate the long-term operation economy of the system, wherein, represents the actual electricity record equivalent value of the jth refrigeration equipment in the ith time period; is the system comprehensive efficiency index, is the energy saving gain weight.

2. The big data analytics based temperature control system for grain silo side walls as claimed in claim 1 wherein, a device control module for applying the adjustment parameters to the side wall refrigeration equipment to dynamically adjust the refrigeration power; Based on the formula determining whether the temperature change trend exceeds a warning range, wherein to predict the temperature change rate, is a temperature sensitivity coefficient, is an expected target temperature, is a current collected temperature; If not, the power calculation module sets , otherwise increases the control sensitivity , and increases the power correction factor to , is the initial power correction factor, is the power correction factor after the control sensitivity is increased, and the equipment control module controls the sidewall refrigeration equipment accordingly.

3. A method of using a granary sidewall temperature control system comprising the big data analysis of any of claims 1-2, characterized in that, the device control module comprises a compensation mechanism unit and a power planning unit; the big data analysis module is connected with the power calculation module; the method comprises the following steps: S1. Using big data analysis to predict the temperature change trend in the grain pile based on the temperature data collected by the internal sensors of the grain pile; S2. Determining a target temperature control area and corresponding refrigeration demand according to the predicted temperature change trend; S3. Calculating the adjustment parameters of the side wall refrigeration equipment power based on the refrigeration demand; S4. Applying the adjustment parameters to the side wall refrigeration equipment to dynamically adjust the refrigeration power, inhibit local overheating and prevent insect breeding.

Images