Coal mine heat energy monitoring and cooling regulation and control method based on Internet of Things

Abstract

The invention discloses a coal mine heat energy monitoring and cooling regulation and control method based on the Internet of Things, and relates to the technical field of coal mine heat energy monitoring and cooling regulation and control. The collected temperature data are transmitted to a central monitoring system in real time through an Internet of Things gateway; redundant temperature sensors are arranged at key positions, and data of the temperature sensors in the same area are regularly compared and verified. According to the invention, the monitoring robustness is improved through a redundant temperature sensor and a data fusion algorithm, a monitoring blind area caused by a single-point fault is avoided, and the accident risk is reduced; a self-adaptive learning algorithm is adopted to dynamically optimize a detection threshold value and a compensation strategy, so that the anomaly detection accuracy is improved, and false alarms and missing alarms are reduced; the temperature trend is pre-judged through a predictive control algorithm, cooling is conducted in advance, balance between safety and energy efficiency is achieved, overheating accidents are prevented, and energy consumption is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of coal mine thermal energy monitoring and temperature reduction control, and in particular to a coal mine thermal energy monitoring and temperature reduction control method based on the Internet of Things. Background Art

[0002] Coal mine thermal energy monitoring and cooling control based on the Internet of Things is an intelligent system that aims to achieve temperature monitoring and automatic control in coal mine environments through the Internet of Things technology. Its basic principle is to collect thermal energy data such as ambient temperature, equipment heat dissipation, and air circulation in real time through sensors deployed at different locations in coal mines, and transmit these data to the central control system. The central system determines whether cooling measures are needed based on the set temperature range and real-time data through big data analysis and machine learning models. If necessary, the system will automatically control cooling equipment such as fans, cooling systems or water spray devices to reduce local temperatures to prevent equipment overheating, improve the working environment, and ensure the safety and production efficiency of miners. This system can not only improve the safety of coal mine operations, but also reduce energy consumption through intelligent management, and improve the accuracy and response speed of cooling measures.

[0003] The prior art has the following deficiencies:

[0004] In the coal mine thermal energy monitoring and cooling control system based on the Internet of Things, the failure of a single temperature sensor or the failure to detect data drift in time may have serious consequences. In the complex environment of coal mines, the system relies on the data fusion of multiple sensors to evaluate the thermal energy status. Once a sensor continues to output erroneous data (such as abnormally low temperatures or long-term unchanged temperatures), the system may misjudge environmental safety, resulting in delays or cancellations of cooling measures. In areas with high temperatures or dense mechanical equipment in summer, such delays may cause local overheating, leading to major accidents such as equipment failures, fires, and even gas explosions.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a method for coal mine thermal energy monitoring and cooling control based on the Internet of Things. By arranging redundant temperature sensors in key areas of coal mines and adopting data fusion algorithms, the robustness of monitoring is improved, ensuring that even if a single sensor fails, it can be compensated by redundant data to avoid monitoring blind spots and sudden accidents. The system adopts an adaptive learning algorithm to dynamically optimize thresholds and compensation strategies based on the historical data and status of the sensor, reduce false alarms and missed alarms, and improve maintenance efficiency. In terms of temperature control, the temperature trend is judged in advance through a predictive control algorithm, and the cooling equipment is actively triggered before exceeding the limit, so as to achieve a balance between safety and energy efficiency, avoid delays and energy waste caused by traditional passive cooling, and solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things, comprising the following steps:

[0008] Multiple temperature sensors are placed near key areas and equipment in coal mines, and the collected temperature data is transmitted to the central monitoring system in real time through the IoT gateway;

[0009] Redundant temperature sensors are placed at key locations, and the data of temperature sensors in the same area are regularly compared and verified to detect whether there is a risk of temperature sensor failure or data drift;

[0010] The anomaly detection algorithm analyzes the historical and current data of each temperature sensor to identify abnormal patterns. When an anomaly is detected, the temperature sensor is marked as a failure risk and an early warning is triggered.

[0011] Use data fusion algorithm to comprehensively analyze multiple temperature sensor data from the same area. If a temperature sensor data is abnormal, the data will be automatically discarded and compensated with redundant temperature sensor data to ensure the accuracy of the assessment of the environmental thermal energy status.

[0012] Based on the data fusion results, the thermal energy trend is judged in advance through the predictive control algorithm. When it is detected that the temperature is about to exceed the limit, the cooling equipment is automatically started to avoid delays;

[0013] Continuously monitor the status data of each temperature sensor, and continuously optimize the abnormal detection threshold and compensation strategy through adaptive learning algorithms to achieve self-diagnosis of the temperature sensor status and maintenance reminders, avoiding single point failures from going undetected for a long time.

[0014] Preferably, in the coal mine thermal energy monitoring and cooling control system, the key areas include: ventilation shaft, main tunnels and branch tunnels, mechanical equipment concentration area, coal seam mining face, coal bunker and conveyor belt area, and electrical control equipment room.

[0015] Preferably, the specific steps of arranging redundant temperature sensors at key locations and regularly comparing and verifying the data of temperature sensors in the same area are as follows:

[0016] First, risk assessment and thermal analysis of the coal mine environment are carried out to determine the key locations where sensors need to be deployed;

[0017] Connect all deployed temperature sensors to the IoT gateway and upload the collected data to the central monitoring platform in real time via wireless or wired means;

[0018] Perform periodic comparison and verification based on multiple sensor data at the same location;

[0019] After detecting that a sensor fails, the system automatically enables the data of the redundant temperature sensor for compensation to ensure the continuity of temperature monitoring. If there is only one redundant temperature sensor in the area, the system will increase the collection frequency and combine the sensor data in the adjacent area to dynamically adjust the ambient temperature model and reduce the monitoring blind spots.

[0020] Preferably, the specific steps of analyzing the historical data and current data of each temperature sensor by an anomaly detection algorithm, identifying abnormal patterns and marking failure risks are as follows:

[0021] First, the historical data of each temperature sensor is collected over a long period of time through the IoT platform to form a temperature curve covering different time periods and environmental changes;

[0022] Compare the current data from the sensor with the historical model to analyze whether its changes over a given period of time conform to the expected pattern;

[0023] When abnormal temperature data is detected, the type of abnormality is determined through a pattern recognition algorithm;

[0024] Once a sensor is marked as at risk of failure, an early warning is immediately triggered and an alarm message is sent to the central monitoring platform.

[0025] Preferably, the specific steps of using a data fusion algorithm to comprehensively analyze the data of multiple temperature sensors in the same area and process abnormal data are as follows:

[0026] First, data is collected from multiple temperature sensors in the same area and preprocessed;

[0027] After time synchronization, data fusion algorithms are used to conduct comprehensive analysis of data from multiple sensors;

[0028] During the data fusion process, the system performs anomaly detection on the data of each sensor. If the standard deviation of the reading of a sensor deviates from the fusion result by more than the threshold, the sensor data is judged to be abnormal, and the abnormal data is automatically discarded, and the data of other sensors are preferentially used for compensation in the fusion model;

[0029] After discarding the abnormal data, the data compensation mechanism is activated, and the data of redundant temperature sensors are used to replace the abnormal sensor data to ensure the continuity of the monitoring results. If the number of redundant temperature sensors is insufficient, the data of sensors in adjacent areas are referenced, and reasonable temperature values ​​are estimated through the prediction model for compensation.

[0030] Preferably, based on the data fusion result, the specific steps of judging the thermal energy trend in advance and automatically starting the cooling device through the predictive control algorithm are as follows:

[0031] The system uses a data fusion algorithm The data X of n temperature sensors in the same area i (t) performs weighted average fusion, where w i is the weight of the ith sensor, and The weight w of the temperature sensor i Adaptive adjustment is made based on historical accuracy and error conditions, and is calculated using the following formula: in, represents the error variance of the i-th temperature sensor;

[0032] The ARIMA model is used to predict future temperature trends. The ARIMA model is used to analyze changes in time series data and predict the next temperature X. p (t+Δt), the prediction formula is as follows: X p (t+Δt)=φ 1 X f (t)+φ 2 X f (t-1)+…+θ 1 ∈(t-1), where φ 1 ,φ 2 is the autoregressive parameter, θ 1 is the moving average parameter, ∈(t-1) is the residual term;

[0033] The predicted temperature X p (t+Δt) and the set safety threshold T safe Compare and judge whether cooling measures are needed. The judgment expression is: ΔT=X p (t+Δt)-T safe If ΔT ≥ 0, an alarm is triggered and the cooling decision process is entered. In order to avoid false alarms, a hysteresis threshold T is used. hysteresisDouble threshold detection is performed, that is, only when ΔT>T hysteresis Continuously exceeds the specified time t c When the temperature drops, the cooling operation is triggered;

[0034] When it is detected that the temperature is about to exceed the limit, the startup strategy of the cooling equipment is optimized through the predictive control algorithm. The fan power of the cooling system is set to P f , the model of the change of ambient temperature over time is: t(t+1)=T(t)-αP f +β(T env -T(t)), where α is the cooling efficiency coefficient and β is the external environment temperature T env For the factors affecting regional temperature, the MPC algorithm optimizes the future period p f The value of can stabilize the system temperature within a safe range and minimize energy consumption: Where N is the prediction time window length and λ is the energy consumption weight factor.

[0035] Preferably, the specific steps of continuously monitoring the temperature sensor status data and optimizing the threshold and compensation strategy through the adaptive learning algorithm to achieve self-diagnosis and maintenance reminder are as follows:

[0036] Continuously connect to each temperature sensor through the IoT gateway to collect status data in real time;

[0037] The system uses an adaptive learning algorithm to continuously analyze the historical data and current status of sensors and dynamically adjust the threshold for anomaly detection;

[0038] When the temperature data or status data of a sensor exceeds the dynamic threshold range, the system immediately marks the sensor as "abnormal". In order to avoid inaccurate monitoring due to the failure of a single sensor, the system starts a real-time compensation strategy, removes the data of the sensor, and enables the data of the redundant temperature sensor for compensation. At the same time, the adaptive algorithm learns the difference between the redundant temperature sensor and the failed sensor, and gradually adjusts the compensation strategy;

[0039] By analyzing the state change trend of the sensor, the sensor can be self-diagnosed.

[0040] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0041] The present invention greatly improves the robustness of temperature monitoring by arranging redundant temperature sensors in key areas of coal mines and adopting data fusion algorithms. Even if a temperature sensor fails or data drifts, the system can compensate through the data of redundant temperature sensors to avoid monitoring blind spots caused by failure of a single sensor. This multi-level monitoring mechanism not only enhances the reliability of the system, but also reduces the probability of sudden accidents.

[0042] The present invention adopts an adaptive learning algorithm, and the system can dynamically adjust the detection thresholds and compensation strategies of different sensors to achieve accurate anomaly detection and maintenance optimization. The coal mine operating environment is complex and changeable, and the fixed threshold strategy is prone to false alarms or missed alarms. The adaptive algorithm continuously optimizes the threshold according to the historical data and real-time status of the sensor to avoid the interference of false alarms on production.

[0043] The present invention uses a predictive control algorithm to determine the ambient temperature trend in advance and automatically triggers the operation of the cooling device when necessary, achieving a high balance between safety and energy efficiency. Traditional passive cooling usually starts only after the temperature exceeds the critical value, while this system can predict the risk and intervene in advance before the temperature is about to exceed the limit, effectively avoiding overheating accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0045] Figure 1 The present invention is a method flow chart of a method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things. DETAILED DESCRIPTION

[0046] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0047] The present invention provides Figure 1 A method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things is shown, comprising the following steps:

[0048] Multiple temperature sensors are placed near key areas and equipment in coal mines, and the collected temperature data is transmitted to the central monitoring system in real time through the IoT gateway;

[0049] In the coal mine thermal energy monitoring and cooling control system, the selection of key areas is directly related to the comprehensiveness and effectiveness of system monitoring. These areas mainly include: ventilation wellhead, main tunnel and branch tunnel, mechanical equipment concentration area, coal seam mining face, coal bunker and conveyor belt area, and electrical control equipment room.

[0050] The ventilation shaft is the main air circulation entrance of the coal mine. Monitoring the temperature here helps to judge the ventilation efficiency and the impact of external high-temperature air.

[0051] Main and branch aisles are the main paths for air and personnel flow. Monitoring the temperature in these areas can help identify the risk of heat accumulation and ensure smooth air flow.

[0052] Areas where mechanical equipment is concentrated (such as motors, pumping stations, and compressors) will generate a lot of heat due to long-term operation of the equipment. Abnormal temperatures may indicate equipment failure or fire risk.

[0053] The coal seam mining face is the core area of ​​production operations. It is equipment-intensive and the working environment is closed, which makes it easy for heat to accumulate. Temperature monitoring is particularly critical.

[0054] If the temperature in the coal bunker and conveyor belt area is too high, it may cause spontaneous combustion of coal or failure of transportation equipment, so real-time monitoring is required to ensure safety.

[0055] The electrical control equipment room contains a large number of electrical equipment. Abnormal temperature will not only damage the equipment but may also cause power outages or fires. Therefore, it is a key monitoring area.

[0056] By placing sensors in these key areas, the system can fully capture the thermal energy status and ensure the accuracy and safety of monitoring.

[0057] Redundant temperature sensors are placed at key locations, and the data of temperature sensors in the same area are regularly compared and verified to detect whether there is a risk of temperature sensor failure or data drift;

[0058] The specific steps for arranging redundant temperature sensors at key locations and regularly comparing and verifying the data of temperature sensors in the same area are as follows:

[0059] The system first conducts risk assessment and thermal analysis of the coal mine environment to determine the key locations where sensors need to be deployed. These locations usually include equipment-intensive areas, ventilation system entrances and exits, areas prone to temperature rise in summer, and high-heat operation areas, such as substations, belt conveyors, ventilation shafts, and mining faces. In order to reduce the risk of single-point failures, the system deploys at least two sets of temperature sensors in each key area (i.e., redundant design) to ensure that even if a sensor fails, the data from the redundant temperature sensors can still ensure the continuity of monitoring. Redundant temperature sensors are produced by different models or different manufacturers to reduce the possibility of synchronous drift of similar equipment.

[0060] All deployed temperature sensors are connected to the IoT gateway and the collected data is uploaded to the central monitoring platform in real time via wireless or wired means. The system sets the collection frequency between 1 second and 1 minute, which is adjusted according to the safety requirements of the coal mine area. To ensure the timeliness of the data, all sensors upload timestamp data through a synchronization mechanism, so that the central system can accurately determine that the data of each sensor belongs to the same time period. The data at this time contains information such as the ambient temperature, sensor status, and transmission signal strength of each sensor for subsequent analysis.

[0061] The system periodically compares and verifies data from multiple sensors at the same location. For example, if most sensors in a group of sensors show a temperature of around 30°C, but one sensor shows 20°C or a constant temperature, the system determines that the sensor's data is abnormal through an anomaly detection algorithm and marks it as a potential failure or drift. To ensure accuracy, the system uses a sliding window algorithm to conduct a comprehensive analysis of sensor data fluctuations over a period of time. If a sensor's data shows significant deviations over multiple time periods, a system alarm is triggered and maintenance personnel are notified to conduct an investigation.

[0062] After detecting a sensor failure, the system will automatically enable the data of the redundant temperature sensor for compensation to ensure the continuity of temperature monitoring. If there is only one redundant temperature sensor in the area, the system will increase the acquisition frequency and dynamically adjust the ambient temperature model in combination with the sensor data in the adjacent area to reduce the monitoring blind area. During the compensation process, the system will also generate a sensor health report, indicating that the sensor may need to be replaced or recalibrated. If the same sensor is detected to be abnormal multiple times, the system will automatically issue a maintenance task and record the specific location and time in the alarm log to ensure that mine managers handle it in a timely manner to avoid safety hazards caused by sensor failure.

[0063] The anomaly detection algorithm analyzes the historical and current data of each temperature sensor to identify abnormal patterns (such as constant temperature and abnormal drift). When an anomaly is detected, the temperature sensor is marked as a failure risk and an early warning is triggered.

[0064] The specific steps to analyze the historical and current data of each temperature sensor through the anomaly detection algorithm, identify abnormal patterns and mark failure risks are as follows:

[0065] The system first collects historical data of each temperature sensor over a long period of time through the IoT platform to form a temperature curve covering different time periods and environmental changes. These historical data include multi-dimensional information such as different seasons, equipment operating status, day and night temperature differences, etc., which provide a reference for the training of anomaly detection algorithms. At the same time, the system models the normal fluctuation range of the sensor, such as the maximum and minimum thresholds and frequency of common temperature changes. In order to improve the accuracy of the model, the system uses machine learning methods (such as time series analysis and regression analysis) to enable the model to adapt to environmental changes. For example, when the temperature is high in summer, the baseline temperature of the system model will be automatically adjusted to avoid misjudgment.

[0066] The system compares the current data of the sensor with the historical model to analyze whether its changes in a given time period conform to the expected pattern. At this point, the algorithm checks the trend of the temperature data (such as whether the temperature rises or falls as expected), volatility (whether it fluctuates within a reasonable range), and stability within the time interval (whether it remains unchanged for a long time). If the system detects that the temperature value of a sensor remains constant or changes abnormally over a certain period of time, such as not conforming to the historical trend and the deviation exceeds the threshold, it is judged that data drift or sensor failure may occur. This process uses a sliding window algorithm, that is, rolling analysis of data over a period of time to ensure the accuracy of the results.

[0067] When an abnormal temperature data is detected, the system uses a pattern recognition algorithm to determine the type of anomaly, such as "abnormal temperature stability", "gradual temperature drift", or "short-term drastic fluctuations". Each type of anomaly corresponds to a different risk of failure. For example, a constant temperature may mean that the sensor is stuck, while a gradual drift may indicate that the sensor's accuracy is gradually decreasing. The system assigns a weight to each anomaly and generates a risk score that quantifies the health of the sensor. When the risk score exceeds the set threshold, the system marks the sensor as "at risk of failure". This evaluation mechanism also takes into account data from redundant temperature sensors to avoid misjudgments.

[0068] Once a sensor is marked as a risk of failure, the system will immediately trigger an early warning and send an alarm message to the central monitoring platform. The early warning information includes the location of the sensor, the type of anomaly, the comparison results of historical data and current data, and the risk score. To avoid delays, the system will automatically generate a work order and push it to the maintenance personnel's mobile device, requiring on-site investigation. In addition, when the early warning is triggered, the system will switch to the data source of the redundant temperature sensor and start a temporary monitoring strategy to increase the collection frequency in the area, or dynamically compensate based on the sensor data in the adjacent area to ensure the continuity and safety of monitoring.

[0069] Use data fusion algorithm to comprehensively analyze multiple temperature sensor data from the same area. If a temperature sensor data is abnormal, the data will be automatically discarded and compensated with redundant temperature sensor data to ensure the accuracy of the assessment of the environmental thermal energy status.

[0070] The specific steps of using the data fusion algorithm to comprehensively analyze the data of multiple temperature sensors in the same area and process abnormal data are as follows:

[0071] The system first collects data from multiple temperature sensors in the same area and performs preprocessing. Preprocessing includes removing null values, formatting data, and filtering out obvious error values ​​(such as negative temperatures or extreme values ​​that exceed a reasonable range). In order to ensure the comparability of data from each sensor at the same time point, the system uses a timestamp synchronization mechanism to align the data. This process can be achieved through the Network Time Protocol (NTP) or the synchronization mechanism of the IoT gateway to ensure that data from different sensors are synchronized within milliseconds. If the data upload delay of a sensor exceeds the set threshold, the data will be marked as "unreliable" and other sensor data will be used first.

[0072] After time synchronization, the system uses data fusion algorithms, such as weighted average algorithm, Kalman filter or Bayesian estimation, to comprehensively analyze the data of multiple sensors. The weighted average algorithm can assign different weights to each sensor data based on the historical performance of the sensor (such as failure rate and error rate). For example, sensor data with smaller errors have greater weights, while sensor data with larger errors have smaller weights. Kalman filtering predicts the temperature value at the next moment based on the current data, and continuously corrects the estimation results to improve the smoothness and accuracy of the data. Through these algorithms, the system can form a more reliable regional temperature model.

[0073] During the data fusion process, the system performs anomaly detection on the data of each sensor. If the standard deviation of the reading of a sensor deviates from the fusion result by more than a certain threshold, or the difference with the data of other sensors is too large, the sensor data is judged to be abnormal. The system will automatically discard the abnormal data and give priority to the data of other sensors for compensation in the fusion model. In addition, the algorithm will record the number and type of abnormalities of the sensor and generate a health status report. If a sensor has multiple abnormalities, the system will list it as a high-risk device and remind the staff to calibrate or replace it.

[0074] After discarding the abnormal data, the system will start the data compensation mechanism and use the data of redundant temperature sensors to replace the abnormal sensor data to ensure the continuity of the monitoring results. If the number of redundant temperature sensors is insufficient, the system will also refer to the data of sensors in the adjacent area and estimate reasonable temperature values ​​for compensation through prediction models (such as time series models). At the same time, the data fusion algorithm will update the fusion model in real time to adapt to the dynamic changes in the environment and equipment status. All processing results will be stored in the log, and reports will be generated regularly to provide data support for subsequent analysis and troubleshooting. Through this dynamic update mechanism, the system ensures that the thermal energy status of the coal mine environment can still be accurately assessed when the sensor is abnormal.

[0075] Based on the data fusion results, the thermal energy trend is judged in advance through the predictive control algorithm. When it is detected that the temperature is about to exceed the limit, the cooling equipment (such as fans or cooling systems) is automatically started to avoid delays;

[0076] Based on the data fusion results, the specific steps of using the predictive control algorithm to predict the thermal energy trend in advance and automatically start the cooling equipment are as follows:

[0077] The system uses a data fusion algorithm The data X of n temperature sensors in the same area i (t) performs weighted average fusion, where w i is the weight of the ith sensor, and The weight w of the temperature sensor i Adaptive adjustment is made based on historical accuracy and error conditions, which can be calculated using the following formula: in, represents the error variance of the i-th temperature sensor;

[0078] The model integrates real-time data from multiple sensors, allowing the system to more accurately reflect the thermal energy status of the current area. f (t).

[0079] The ARIMA (Auto-Regressive Integrated Moving Average) model is used to predict future temperature trends. The ARIMA model is suitable for analyzing changes in time series data and predicting the next temperature X. p (t+Δt). The prediction formula is as follows: X p (t+Δt)=φ 1 X f (t)+φ 2 X f (t-1)+…+θ 1 ∈(t-1), where φ 1 ,φ 2 is the autoregressive parameter, θ1 is the moving average parameter, ∈(t-1) is the residual term;

[0080] By fitting the historical temperature data, the system predicts the temperature change trend in a short period of time. p (t+Δt), which is used to determine in advance whether the safety threshold will be exceeded.

[0081] The predicted temperature X p (t+Δt) and the set safety threshold T safe Compare and determine whether cooling measures are needed: ΔT = X p (t+Δt)-T safe If ΔT ≥ 0, the predicted temperature will exceed the threshold, and the system will trigger an alarm and enter the cooling decision process. In order to avoid false alarms, the system uses a hysteresis threshold T hysteresis Double threshold detection is performed, that is, only when ΔT>T hysteresis Continuously exceeds the specified time t c The cooling operation will be triggered only when

[0082] When the temperature is detected to be about to exceed the limit, the startup strategy of the cooling equipment is optimized through the predictive control algorithm (Model Predictive Control, MPC). Assume that the fan power of the cooling system is P f , the model of the change of ambient temperature over time is: T(t+1)=T(t)-αP f +β(T env -T(t)), where α is the cooling efficiency coefficient and β is the external environment temperature T env The MPC algorithm optimizes the future period P f The value of can stabilize the system temperature within a safe range and minimize energy consumption: Where N is the prediction time window length, λ is the energy consumption weight factor;

[0083] The MPC algorithm updates predictions based on new data during each control cycle, ensuring that the cooling equipment responds accurately and efficiently. The control signal is transmitted directly to the fan or cooling equipment, and the output power is continuously adjusted based on real-time temperature feedback to avoid overcooling or delayed response.

[0084] Through the above steps, the system realizes intelligent regulation based on data fusion and predictive control algorithm, accurately judges the thermal energy trend before the temperature exceeds the limit, and starts the cooling equipment in time to ensure the safe and efficient operation of the coal mine environment.

[0085] Continuously monitor the status data of each temperature sensor, and continuously optimize the abnormal detection threshold and compensation strategy through adaptive learning algorithms to achieve self-diagnosis of the temperature sensor status and maintenance reminders, thus avoiding single point failures that remain undetected for a long time;

[0086] The specific steps for continuously monitoring the temperature sensor status data and optimizing the threshold and compensation strategy through the adaptive learning algorithm to achieve self-diagnosis and maintenance reminders are as follows:

[0087] The system continuously connects to each temperature sensor through the IoT gateway and collects status data in real time, including key indicators such as temperature readings, data transmission delays, signal strength, and battery power (if it is a wireless sensor). These status data are stored in a central database, with the collection time and sensor location marked. Through the cumulative analysis of these data, the system can determine the health status of the sensor. For example, a sudden decrease in signal strength may indicate that the sensor has moved, and an increase in transmission delay may indicate an unstable network connection. In addition, the system also records the number of historical failures of each sensor to provide a basis for subsequent optimization.

[0088] The system uses an adaptive learning algorithm to continuously analyze the historical data and current status of sensors and dynamically adjust the threshold for anomaly detection. Since different sensors work differently in different environments, fixed thresholds are prone to false positives or false negatives. The adaptive algorithm develops personalized detection thresholds for each sensor by learning the data patterns of sensors in normal and abnormal conditions. For example, if the temperature in the area where a sensor is located fluctuates greatly, the system will automatically relax the fluctuation range to reduce the false alarm rate. The system will also automatically adjust the thresholds for different time periods based on weather changes or seasonal factors to ensure the accuracy of detection.

[0089] When the temperature data or status data of a sensor exceeds the dynamic threshold range, the system immediately marks the sensor as "abnormal state". In order to avoid inaccurate monitoring due to the failure of a single sensor, the system starts a real-time compensation strategy, removes the data of the sensor, and enables the data of the redundant temperature sensor for compensation. At the same time, the adaptive algorithm learns the difference between the redundant temperature sensor and the failed sensor, and gradually adjusts the compensation strategy. For example, in the same environment, if the data of multiple redundant temperature sensors are relatively consistent, the system will determine the failure of the main sensor more quickly and reduce the judgment time. In addition, the system will adjust the collection frequency and data calibration cycle of the sensors in the monitoring area according to the detected abnormal frequency.

[0090] The system can perform self-diagnosis of sensors by analyzing the state change trend of sensors. For example, if the sensor data drift gradually increases, the system will predict its possible failure time and generate maintenance reminders in advance. Maintenance reminders will be pushed to relevant technicians to remind them to calibrate or replace the equipment to avoid long-term undetected failures. The system will also establish a maintenance cycle model based on historical data and generate a personalized maintenance plan for each sensor. If a sensor has abnormalities for many times but has not been replaced, the system will upgrade the reminder level and issue a high-priority warning on the management platform to ensure timely processing. In addition, the system will simultaneously record maintenance history and fault data to provide a reference for future analysis.

[0091] Specific implementation method 1: redundant temperature sensor arrangement and multi-level monitoring;

[0092] In the complex and high-risk environment of coal mines, the accuracy and continuity of temperature monitoring are crucial, so the system uses redundant temperature sensors in key areas and ensures the reliability of data collection and transmission through a multi-level network. Taking substations, belt conveyor systems, and mining faces as examples, each area is not only equipped with a main temperature sensor, but also with at least one redundant temperature sensor. When a sensor fails or data drifts, the redundant device can take over immediately to ensure uninterrupted temperature monitoring. To reduce the risk of synchronous drift of sensors of the same type, redundant temperature sensors give priority to devices of different models or manufacturers. This heterogeneous redundant design can significantly reduce the risk of false alarms caused by the same sensor type.

[0093] In addition, the system adopts a multi-level data collection and transmission method. The sensor data is first transmitted to the IoT gateway. Each gateway is responsible for covering a local area, such as a belt conveyor or a vent. Each gateway is interconnected with the central control platform through wireless communication. When a gateway fails, the adjacent gateway will automatically expand the coverage to ensure the integrity of the monitoring area. The central system compares the data of the redundant temperature sensors every few minutes. If the temperature difference exceeds the predetermined threshold (such as ±2°C), the system will immediately mark the main sensor as "data abnormal". If the system confirms that the abnormal data comes from a sensor failure, the data of the main sensor will be eliminated, and the data of the redundant temperature sensor will be enabled for compensation. This mechanism not only improves the robustness of the system, but also avoids monitoring blind spots caused by the failure of a single device.

[0094] Specific implementation method 2: Adaptive learning algorithm is used for dynamic adjustment and maintenance of thresholds;

[0095] Temperature sensors have been running in complex coal mine environments for a long time, and fixed threshold strategies are difficult to adapt to changes on site. Therefore, the system introduces an adaptive learning algorithm to dynamically adjust the detection threshold of each sensor based on historical data and environmental changes. The system generates a set of time series-based data models for each sensor, including the temperature difference pattern between day and night, the temperature rise caused by equipment operation, and temperature fluctuations in different seasons. For example, during the peak operating period in summer, the system recognizes that the temperature often exceeds 30°C, but this is normal. Therefore, the system automatically relaxes the detection threshold and adjusts it to 35°C to avoid unnecessary false alarms. In winter, the system tightens the threshold to ensure more sensitive detection of abnormalities at lower temperatures.

[0096] In addition, sensor status data (such as signal strength, response time, and battery status) are also continuously monitored. When the system detects that the battery power of a sensor is below the set 20% threshold, a maintenance reminder is automatically generated and pushed to the mobile device of the field technician. If the signal strength fluctuates or the frequency of data upload decreases, the system will determine that the sensor may be in an edge area or the connection is unstable, thereby reminding the technician to recalibrate or adjust the sensor position. The system's adaptive algorithm also optimizes the maintenance cycle of each sensor based on maintenance history. For example, for sensors that frequently fail, the system will shorten their maintenance intervals and generate detailed equipment health reports to ensure that all equipment is always in optimal working condition.

[0097] Specific implementation method three: predictive control and automatic cooling system linkage;

[0098] Based on the data fusion results of multiple sensors, the system uses the ARIMA model and predictive control algorithm to analyze and predict the ambient temperature trend. When the system predicts that the temperature in a certain area will exceed the set safety threshold (such as 40°C) within the next 10 minutes, it will trigger the start command of the cooling equipment in advance to avoid equipment damage or safety accidents caused by excessive temperature. In high-temperature areas, such as mining faces and substations, the system can intelligently adjust the power of the fan or cooling system based on the real-time temperature and prediction results. For example, when it is detected that the temperature is about to rise but has not yet reached a dangerous level, the system will run the fan at medium and low power to ensure that energy consumption is minimized. When the temperature approaches the critical value further, the system will gradually increase the power of the equipment to achieve layer-by-layer cooling.

[0099] The system uses a feedback adjustment mechanism to ensure that the operation of the cooling equipment matches the actual situation on site. After the cooling equipment is running, the system will continue to monitor the temperature change trend. If the temperature drops rapidly in a short period of time, the system will automatically reduce the fan speed to avoid excessive cooling. In addition, the system will determine whether the cooling effect meets expectations based on the feedback data from sensors in adjacent areas. For example, if the temperature in a certain area does not drop as expected, it may be due to a fan failure or a cooling pipe blockage. The system will immediately switch to the backup cooling equipment and generate a maintenance work order to remind the technician to conduct an on-site inspection.

[0100] The system also records the start and stop frequency and operating time of the cooling equipment to provide a reference for subsequent maintenance and optimization. If a certain device is started frequently, the system will analyze its operating history and identify possible hidden dangers, such as fan blade wear or motor overload. Through this intelligent linkage mechanism, the system not only ensures the temperature safety of the mine area, but also achieves a high degree of automation and optimization in energy saving and maintenance.

[0101] The present invention greatly improves the robustness of temperature monitoring by arranging redundant temperature sensors in key areas of coal mines and adopting data fusion algorithms. Even if a temperature sensor fails or data drifts, the system can compensate through the data of the redundant temperature sensor to avoid monitoring blind spots caused by the failure of a single sensor. This multi-level monitoring mechanism not only enhances the reliability of the system, but also reduces the probability of sudden accidents. For example, if the main sensor in a high-temperature area fails, the system can quickly switch to the redundant temperature sensor to ensure continuous monitoring of the thermal energy status. In addition, the real-time monitoring and automatic alarm functions of the sensor health status can remind technicians to perform maintenance in time before a failure occurs, effectively preventing major safety accidents. This design significantly reduces the risk of production interruption caused by equipment failure and provides a guarantee for the safe operation of coal mine operations.

[0102] The present invention adopts an adaptive learning algorithm, and the system can dynamically adjust the detection thresholds and compensation strategies of different sensors to achieve accurate anomaly detection and maintenance optimization. The coal mine operating environment is complex and changeable, and the fixed threshold strategy is prone to false alarms or missed alarms. The adaptive algorithm continuously optimizes the threshold according to the historical data and real-time status of the sensor to avoid the interference of false alarms on production. For example, in high temperature areas in summer, the system will automatically relax the threshold range of temperature fluctuations, and increase the detection sensitivity in lower temperature environments. This flexible threshold adjustment ensures the accuracy of detection. In addition, the system will generate a personalized maintenance plan based on the operating status of the sensor, automatically reminding technicians to calibrate or replace it, avoiding long-term failure of equipment without being discovered, thereby increasing the service life of the sensor and the overall stability of the system.

[0103] The present invention uses a predictive control algorithm to judge the trend of ambient temperature in advance, and automatically triggers the operation of the cooling equipment when necessary, thereby achieving a high balance between safety and energy efficiency. Traditional passive cooling is usually activated only after the temperature exceeds the critical value, while the present system can predict the risk and intervene in advance before the temperature is about to exceed the limit, effectively avoiding the occurrence of overheating accidents. For example, when it is detected that the temperature in the equipment area is gradually rising and approaching the threshold, the system will adjust the operating power of the fan in advance to reduce the risk of equipment overload. At the same time, the system adjusts the output power of the cooling equipment through real-time feedback to avoid energy waste caused by excessive cooling. Predictive control not only improves the safety of the coal mine environment, but also reduces the operating frequency and energy consumption of the equipment, ensuring that the system achieves the best balance between safety and energy saving.

[0104] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things, characterized in that: The following steps are involved: Multiple temperature sensors are placed near key areas and equipment in coal mines, and the collected temperature data is transmitted to the central monitoring system in real time through the IoT gateway; Redundant temperature sensors are placed at key locations, and the data of temperature sensors in the same area are regularly compared and verified to detect whether there is a risk of temperature sensor failure or data drift; The anomaly detection algorithm analyzes the historical and current data of each temperature sensor to identify abnormal patterns. When an anomaly is detected, the temperature sensor is marked as a failure risk and an early warning is triggered. Use data fusion algorithm to comprehensively analyze multiple temperature sensor data from the same area. If a temperature sensor data is abnormal, the data will be automatically discarded and compensated with redundant temperature sensor data to ensure the accuracy of the assessment of the environmental thermal energy status. Based on the data fusion results, the thermal energy trend is judged in advance through the predictive control algorithm. When it is detected that the temperature is about to exceed the limit, the cooling equipment is automatically started to avoid delays; Continuously monitor the status data of each temperature sensor, and continuously optimize the abnormal detection threshold and compensation strategy through adaptive learning algorithms to achieve self-diagnosis of the temperature sensor status and maintenance reminders, avoiding single point failures from going undetected for a long time.

2. The method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things according to claim 1 is characterized in that: In the coal mine thermal energy monitoring and cooling control system, the key areas include: ventilation shaft, main tunnels and branch tunnels, mechanical equipment concentration area, coal seam mining face, coal bunker and conveyor belt area, and electrical control equipment room.

3. The method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things according to claim 1 is characterized in that: The specific steps for arranging redundant temperature sensors at key locations and regularly comparing and verifying the data of temperature sensors in the same area are as follows: First, risk assessment and thermal analysis of the coal mine environment are carried out to determine the key locations where sensors need to be deployed; Connect all deployed temperature sensors to the IoT gateway and upload the collected data to the central monitoring platform in real time via wireless or wired means; Perform periodic comparison and verification based on multiple sensor data at the same location; After detecting that a sensor fails, the system automatically enables the data of the redundant temperature sensor for compensation to ensure the continuity of temperature monitoring. If there is only one redundant temperature sensor in the area, the system will increase the collection frequency and combine the sensor data in the adjacent area to dynamically adjust the ambient temperature model and reduce the monitoring blind spots.

4. The method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things according to claim 1 is characterized in that: The specific steps to analyze the historical and current data of each temperature sensor through the anomaly detection algorithm, identify abnormal patterns and mark failure risks are as follows: First, the historical data of each temperature sensor is collected over a long period of time through the IoT platform to form a temperature curve covering different time periods and environmental changes; Compare the current data from the sensor with the historical model to analyze whether its changes over a given period of time conform to the expected pattern; When abnormal temperature data is detected, the type of abnormality is determined through a pattern recognition algorithm; Once a sensor is marked as at risk of failure, an early warning is immediately triggered and an alarm message is sent to the central monitoring platform.

5. The method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things according to claim 1 is characterized in that: The specific steps of using the data fusion algorithm to comprehensively analyze the data of multiple temperature sensors in the same area and process abnormal data are as follows: First, data is collected from multiple temperature sensors in the same area and preprocessed; After time synchronization, data fusion algorithms are used to conduct comprehensive analysis of data from multiple sensors; During the data fusion process, the system performs anomaly detection on the data of each sensor. If the standard deviation of the reading of a sensor deviates from the fusion result by more than the threshold, the sensor data is judged to be abnormal, and the abnormal data is automatically discarded, and the data of other sensors are preferentially used for compensation in the fusion model; After discarding the abnormal data, the data compensation mechanism is activated, and the data of redundant temperature sensors are used to replace the abnormal sensor data to ensure the continuity of the monitoring results. If the number of redundant temperature sensors is insufficient, the data of sensors in adjacent areas are referenced, and reasonable temperature values ​​are estimated through the prediction model for compensation.

6. The method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things according to claim 1 is characterized in that: Based on the data fusion results, the specific steps of using the predictive control algorithm to predict the thermal energy trend in advance and automatically start the cooling equipment are as follows: The system uses a data fusion algorithm The data X of n temperature sensors in the same area i (t) performs weighted average fusion, where w i is the weight of the ith sensor, and The weight w of the temperature sensor i Adaptive adjustment is made based on historical accuracy and error conditions, and is calculated using the following formula: in, represents the error variance of the i-th temperature sensor; The ARIMA model is used to predict future temperature trends. The ARIMA model is used to analyze changes in time series data and predict the next temperature X. p (t+Δt), the prediction formula is as follows: X p (t+Δt)=φ1X f (t)+φ2X f (t-1)+…+θ1∈(t-1), where φ1, φ2 are autoregressive parameters, θ1 is the moving average parameter, and ∈(t-1) is the residual term; The predicted temperature X p (t+Δt) and the set safety threshold T safe Compare and judge whether cooling measures are needed. The judgment expression is: ΔT=X p (t+Δt)-T safe If ΔT ≥ 0, an alarm is triggered and the cooling decision process is entered. In order to avoid false alarms, a hysteresis threshold T is used. hysteresis Double threshold detection is performed, that is, only when ΔT>T hysteresis Continuously exceeds the specified time t c When the temperature drops, the cooling operation is triggered; When it is detected that the temperature is about to exceed the limit, the startup strategy of the cooling equipment is optimized through the predictive control algorithm. The fan power of the cooling system is set to P f , the model of the change of ambient temperature over time is: T(t+1)=T(t)-αP f +β(T env -T(t)), where α is the cooling efficiency coefficient and β is the external environment temperature T env The MPC algorithm optimizes the factors affecting regional temperature by f The value of can stabilize the system temperature within a safe range and minimize energy consumption: Where N is the prediction time window length and λ is the energy consumption weight factor.

7. The method for coal mine thermal energy monitoring and temperature reduction control based on the Internet of Things according to claim 1 is characterized in that: The specific steps for continuously monitoring the temperature sensor status data and optimizing the threshold and compensation strategy through the adaptive learning algorithm to achieve self-diagnosis and maintenance reminders are as follows: Continuously connect to each temperature sensor through the IoT gateway to collect status data in real time; The system uses an adaptive learning algorithm to continuously analyze the historical data and current status of sensors and dynamically adjust the threshold for anomaly detection; When the temperature data or status data of a sensor exceeds the dynamic threshold range, the system immediately marks the sensor as "abnormal". In order to avoid monitoring inaccuracy due to the failure of a single sensor, the system starts a real-time compensation strategy, removes the data of the sensor, and enables the data of the redundant temperature sensor for compensation. At the same time, the adaptive algorithm learns the difference between the redundant temperature sensor and the failed sensor, and gradually adjusts the compensation strategy; By analyzing the state change trend of the sensor, the sensor can be self-diagnosed.

Images