Sludge drying intelligent control system and method based on Internet of Things

Abstract

The invention relates to the technical field of sludge treatment, in particular to a sludge drying intelligent control system and method based on the Internet of Things. The method comprises the following steps: constructing an Internet of Things monitoring network; the method comprises the following steps: performing periodic monitoring data acquisition by using an Internet of Things monitoring network, and performing multi-source data preprocessing to generate sludge state monitoring characteristic data; performing grid point moisture content value classification according to the sludge state monitoring characteristic data to generate sludge moisture content partition data; according to the sludge moisture content zoning data, zoning cooperative regulation and control treatment is carried out, and zoning cooperative regulation and control parameters are generated; and analyzing the risk that the odor exceeds the standard, and performing control instruction optimization on the zoned coordinated regulation and control parameters to obtain an odor linked intervention trigger signal so as to realize intelligent control of sludge drying. According to the invention, accurate regulation and control of water content and efficient treatment of odor in the sludge drying process are realized through an intelligent control technology, and the method can be widely applied to the drying treatment process in the fields of municipal sludge, industrial sludge and the like.

Description

Technical Field

[0001] The present invention relates to the technical field of sludge treatment, and in particular to an intelligent sludge drying control system and method based on the Internet of Things. Background Art

[0002] Sludge, a byproduct of wastewater treatment, is characterized by high moisture content, high organic matter content, and potential environmental pollution risks. Sludge drying, a key step in sludge treatment, aims to reduce sludge moisture content and volume for subsequent disposal or resource recovery. Traditional sludge drying technologies suffer from issues such as imprecise moisture control, high energy consumption, and incomplete odor treatment, hindering their large-scale application and resource utilization. Precisely controlling the moisture content during sludge drying to improve drying efficiency and reduce energy consumption, while also effectively treating odors (such as hydrogen sulfide, ammonia, and volatile organic compounds) generated during the drying process to avoid secondary pollution, are key technical challenges currently unresolved in the sludge drying process. By integrating the Internet of Things (IoT), big data analytics, artificial intelligence (AI), and automated control technologies, intelligent management of the entire sludge treatment process can be achieved, thereby improving treatment efficiency, reducing energy consumption, and minimizing manual intervention. While intelligent sludge treatment control systems have demonstrated significant advantages in improving efficiency, reducing energy consumption, and minimizing manual intervention, they still face several limitations and challenges in practical application. In particular, the management model of most sewage treatment plants is still at the traditional stage, making it difficult to fully utilize the advantages of intelligent systems. Intelligent systems can monitor odor concentrations and adjust treatment equipment, but their treatment effects on complex odor components (such as VOCs) are limited, and they still need to be coordinated with multiple technical means. In addition, the calculation time of intelligent algorithms is long and cannot meet the needs of real-time control. The existing system is difficult to flexibly expand to adapt to new treatment processes or equipment. Summary of the Invention

[0003] Based on this, the present invention provides an Internet of Things-based intelligent sludge drying control system and method to solve at least one of the above technical problems.

[0004] To achieve the above objectives, an intelligent control method for sludge drying based on the Internet of Things includes the following steps:

[0005] Step S1: deploying an IoT monitoring network for the sludge drying chamber to build an IoT monitoring network; utilizing the IoT monitoring network to periodically collect monitoring data from the sludge drying chamber, and preprocessing multi-source data to generate sludge status monitoring feature data;

[0006] Step S2: classifying the grid point moisture content values ​​according to the sludge state monitoring characteristic data to generate sludge moisture content partition data; performing drying correction factor processing on the sludge moisture content partition data to obtain the partition drying correction factor;

[0007] Step S3: Obtaining the sludge organic matter content and sludge density; analyzing the sludge surface evaporation rate based on the sludge organic matter content and sludge density, and deducing the change in moisture content of the sludge layer to generate moisture content change trend data; performing zoning coordinated control processing using the zoning drying correction factor and the moisture content change trend data to generate zoning coordinated control parameters; performing odor exceeding standard risk analysis based on the moisture content change trend data to generate odor exceeding standard risk data;

[0008] Step S4: Optimize the control instructions of the zone collaborative control parameters through the odor exceeding risk data to obtain the odor linkage intervention trigger signal; and realize the intelligent control of sludge drying according to the odor linkage intervention trigger signal.

[0009] Preferably, the present invention further provides an Internet of Things-based intelligent sludge drying control system, which implements the above-mentioned Internet of Things-based intelligent sludge drying control method. The Internet of Things-based intelligent sludge drying control system includes the following modules:

[0010] The sensor measurement module is equipped with a microwave moisture content sensor, an infrared temperature sensor, an electrochemical odor concentration sensor, a hot film air flow meter, and a differential pressure sensor, which is used to collect real-time monitoring and feedback of key parameters such as sludge moisture content, drying temperature, odor concentration, and ventilation volume;

[0011] The intelligent control module, with a PLC-based control core, is responsible for data processing and command issuance, and controls operating parameters based on monitoring data. The operating parameter control includes frequency converters, heaters, and feeding devices, which control the temperature, ventilation volume, and sludge feeding speed parameters during the sludge drying process.

[0012] The process unit control module is equipped with a drying temperature control unit, a ventilation volume control unit, and a sludge feeding speed control unit, including heat source supply volume adjustment, fan air volume adjustment, and feeding speed adjustment;

[0013] The data management and visualization module includes data storage and visualization interface, uses cloud platforms or local servers to store historical data, and uses mobile terminals to display key parameters such as sludge concentration, drying temperature, and ventilation volume in real time, while supporting remote monitoring.

[0014] The sensor measurement module of the present invention integrates multiple sensors to collect various parameters, including sludge moisture content, drying temperature, odor concentration, and ventilation rate. Microwave sensors are installed at key nodes in the sludge drying process (such as the feed inlet, dryer outlet, and return port). By leveraging the principle of signal attenuation when microwaves penetrate sludge, they measure sludge moisture content in real time, enabling full-process monitoring. Sensor data is transmitted to a central control system in real time via Internet of Things (IoT) technology. A temperature sensor senses ambient temperature using a thermistor and converts the temperature signal into an electrical output. This monitors the temperature of the drying equipment in real time to prevent excessively high or low temperatures from affecting drying efficiency. Gas sensors use chemical or physical methods to detect the concentration of specific gas components (such as hydrogen sulfide and ammonia) and monitor harmful gases generated during sludge treatment. A flow sensor measures air velocity or pressure differential to calculate ventilation rate, monitors ventilation rate in the drying equipment, ensures proper ventilation system operation, and optimizes drying efficiency. Sensor data is transmitted to the intelligent control core via wired or wireless communication protocols. Verification mechanisms (such as CRC) are incorporated into the transmission process to prevent data loss or errors. The intelligent control module is responsible for acquiring data from the sensor measurement module and performing real-time analysis and processing. Its core process includes data acquisition, data transmission, data processing, decision analysis, and control execution. By setting a reasonable sampling frequency, the system utilizes microwave sensors, temperature sensors, gas sensors, and flow sensors to collect real-time data such as sludge moisture content, drying temperature, odor concentration, and ventilation volume. The collected analog signals are converted into digital signals and output through a standard interface. Data anomalies are determined based on preset thresholds, process parameter changes are predicted, and control strategies are optimized. Based on the decision-making results, control instructions are sent to the process unit control module. In the process unit control module, the drying temperature control unit, the ventilation volume control unit and the sludge feed speed control unit are respectively responsible for the control of specific equipment. The drying temperature control unit controls the heat source supply by adjusting the power of the heater (such as adjusting the voltage and current) or the opening of the gas valve. When the temperature sensor detects that the drying temperature is lower than the set value, the control system increases the heat source supply; when the temperature is too high, the heat source supply is reduced or the cooling device is started; the ventilation volume control unit adjusts the fan speed through the frequency converter to control the air volume. When the odor concentration sensor detects that the odor concentration exceeds the standard, the control system increases the fan speed or opens the air valve to increase the ventilation volume; when the ventilation volume is too large, the fan speed is reduced or the air valve is closed; the sludge feed speed control unit is responsible for adjusting the sludge feed speed to ensure that the drying equipment processing capacity matches the feed volume. The sludge feed speed is controlled by adjusting the speed of the conveyor or pump through the frequency converter. When the drying equipment processing capacity is insufficient, the control system reduces the feed speed; when the equipment is idle, the feed speed is increased to improve the processing efficiency.Therefore, the present invention's intelligent control method for sludge drying based on the Internet of Things can be applied to the intelligent control technology for sludge dewatering and drying in sewage treatment plants. By optimizing the control strategy, energy consumption can be reduced, the drying efficiency can be effectively improved, and the purpose of energy saving, consumption reduction, and volume reduction can be achieved. The ability of multi-parameter collaborative monitoring is adopted to realize the real-time collection and transmission of different sensor data. At the same time, by combining traditional control methods and advanced algorithms, the control accuracy and response speed of the system are significantly improved through intelligent algorithms and optimized control strategies, achieving precise control. Through actuators such as frequency converters, heaters, and valves, precise adjustment of fan speed, heating power, sludge flow, etc. is achieved, and the actual operating status is fed back to the intelligent control system to form a closed-loop control, which significantly improves the stability and control effect of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a schematic flow chart of the steps of an intelligent control method for sludge drying based on the Internet of Things of the present invention;

[0016] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0017] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. It is obvious that the embodiments described are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts are within the scope of protection of the present invention.

[0018] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0019] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0020] To achieve this, please refer to Figure 1 The present invention provides an intelligent control method for sludge drying based on the Internet of Things, comprising the following steps:

[0021] Step S1: deploying an IoT monitoring network for the sludge drying chamber to build an IoT monitoring network; utilizing the IoT monitoring network to periodically collect monitoring data from the sludge drying chamber, and preprocessing multi-source data to generate sludge status monitoring feature data;

[0022] Step S2: classifying the grid point moisture content values ​​according to the sludge state monitoring characteristic data to generate sludge moisture content partition data; performing drying correction factor processing on the sludge moisture content partition data to obtain the partition drying correction factor;

[0023] Step S3: Obtaining the sludge organic matter content and sludge density; analyzing the sludge surface evaporation rate based on the sludge organic matter content and sludge density, and deducing the change in moisture content of the sludge layer to generate moisture content change trend data; performing zoning coordinated control processing using the zoning drying correction factor and the moisture content change trend data to generate zoning coordinated control parameters; performing odor exceeding standard risk analysis based on the moisture content change trend data to generate odor exceeding standard risk data;

[0024] Step S4: Optimize the control instructions of the zone collaborative control parameters through the odor exceeding risk data to obtain the odor linkage intervention trigger signal; and realize the intelligent control of sludge drying according to the odor linkage intervention trigger signal.

[0025] In an embodiment of the present invention, the sludge drying intelligent control method based on the Internet of Things includes the following steps:

[0026] Step S1: deploying an IoT monitoring network for the sludge drying chamber to build an IoT monitoring network; utilizing the IoT monitoring network to periodically collect monitoring data from the sludge drying chamber, and preprocessing multi-source data to generate sludge status monitoring feature data;

[0027] In an embodiment of the present invention, when deploying an Internet of Things monitoring network for a sludge drying chamber, a three-layer microwave moisture content sensor array is first set up in the chamber, with 5 measuring points in each layer, located at the sludge surface (0-5cm), middle layer (5-15cm) and bottom layer (15-30cm). At the same time, 16 infrared temperature sensors are arranged to form a 4×4 grid array, covering the entire surface of the drying chamber. Three electrochemical odor concentration sensors are installed at the top, middle and bottom of the chamber to monitor odor components such as hydrogen sulfide, ammonia and methyl mercaptan. A hot film air flow meter and a differential pressure sensor are installed at the inlet and outlet of the ventilation duct respectively. All sensors are connected to the edge computing controller via ZigBee, LoRa or WiFi wireless communication protocols to build an Internet of Things monitoring network. The basic sampling period is set to 5 minutes, which is dynamically adjusted according to the current drying stage (preheating, constant speed drying, speed reduction drying). The preheating stage is shortened to 2 minutes and the speed reduction drying stage is extended to 10 minutes. The collected data uses cubic spline interpolation to process missing values, moving median filtering to eliminate outliers, and extracts multiple characteristic parameters to generate sludge status monitoring characteristic data including stratified moisture content values, temperature field, odor concentration, etc.

[0028] Step S2: classifying the grid point moisture content values ​​according to the sludge state monitoring characteristic data to generate sludge moisture content partition data; performing drying correction factor processing on the sludge moisture content partition data to obtain the partition drying correction factor;

[0029] In an embodiment of the present invention, when classifying the moisture content values ​​of grid points based on the characteristic data of sludge state monitoring, a three-dimensional Kriging interpolation method is used to construct an 8×8×6 grid array with a total of 384 grid nodes. Traversing each grid node, areas with a moisture content greater than 80% are marked as high moisture areas (code H) and assigned a value of 3; areas with a moisture content between 50% and 80% are marked as medium moisture areas (code M) and assigned a value of 2; areas with a moisture content less than 50% are marked as low moisture areas (code L) and assigned a value of 1. The six-connected component method is used to determine regional connectivity, and isolated areas with an area less than 5% of the total number of grids are merged into the adjacent largest area. Then, the three indicators of overall moisture content evaluation value, vertical moisture content gradient data, and moisture content unevenness are calculated to perform real-time sludge drying index evaluation. According to the drying index, the piecewise linear mapping function is applied to the different moisture content zones to calculate the drying correction factor: the correction factor FH for the high water content zone is 1.5-0.5DI (when DI≤0.6) or 1.8-1.0DI (when DI>0.6); the corresponding calculation formulas are used for the medium water content zone and the low water content zone respectively; the edge position area is additionally multiplied by the edge correction coefficient 1.1, and finally the zone drying correction factor data table is obtained.

[0030] Step S3: Obtaining the sludge organic matter content and sludge density; analyzing the sludge surface evaporation rate based on the sludge organic matter content and sludge density, and deducing the change in moisture content of the sludge layer to generate moisture content change trend data; performing zoning coordinated control processing using the zoning drying correction factor and the moisture content change trend data to generate zoning coordinated control parameters; performing odor exceeding standard risk analysis based on the moisture content change trend data to generate odor exceeding standard risk data;

[0031] In an embodiment of the present invention, the sludge organic matter content is obtained by expressing it through the ratio of volatile solids (VS) to total solids (TS), and the sludge density is measured by the bulk density method. Based on these parameters, the heat and mass transfer balance parameters are matched, including the effective thermal conductivity, specific heat capacity, mass diffusion coefficient and heat transfer coefficient. The surface water evaporation rate is calculated using the improved evaporation kinetic model, the heat energy received by the sludge is calculated by the heat balance analysis method, the internal water migration rate is calculated using the improved Fick diffusion model, and the layered heat balance method is used to deduce the trend of the moisture content change in the three layers within 24 hours. The zoned drying correction factor is used to perform drying uniformity target processing, evaluate the drying completion time, set the drying temperature of each area, and configure differentiated drying parameters. The sludge drying chamber is mapped into 5-8 independent physical control areas, the heat energy output ratio of the waste heat recovery and auxiliary heating device in each area is determined, and the fresh air mixing ratio and airflow switching cycle are set. At the same time, based on the drying stage path prediction, temperature change rate analysis and organic matter degradation characteristics, the odor release potential is evaluated, the odor concentration retention is calculated, the predicted odor concentration change curve is generated, and a multi-level threshold monitoring strategy is implemented to conduct odor exceeding standard risk analysis.

[0032] Step S4: Optimize the control instructions of the zone collaborative control parameters through the odor exceeding risk data to obtain the odor linkage intervention trigger signal; and realize the intelligent control of sludge drying according to the odor linkage intervention trigger signal.

[0033] In an embodiment of the present invention, when optimizing control instructions for zoned collaborative control parameters based on odor-exceeding risk data, peak detection and threshold analysis are first used to extract warning time nodes and calculate the risk intensity index for each warning point. A three-dimensional coordinated intervention strategy is implemented based on the warning time points: a temporary 15°C temperature reduction, a 70% increase in ventilation volume, and a pulsed odor capture system with a frequency of 3 minutes per session are implemented in high-risk areas. Intervention parameters are adjusted accordingly for medium- and low-risk areas. Simultaneously, false dry shell risk areas are identified from moisture content trend data, and the moisture content gradient between the surface and middle / bottom layers is calculated to determine whether the dry shell threshold of 15% is exceeded. If the threshold is exceeded, the depth of the sludge layer is determined based on its thickness, and the frequency of the sludge layer is set based on the depth and moisture content zones. The execution instructions for the sludge layer mechanism include six core parameters: zone coordinates, depth, speed, and angle, ensuring that the sludge uniformity index reaches above 0.85 after sludge layering. The odor linkage intervention trigger signal and the turnover mechanism execution instructions are integrated into a 48-hour rolling execution plan. Through the hierarchical control architecture, the heating system, ventilation system and turnover mechanism are driven to work together to achieve intelligent control of the entire sludge drying process.

[0034] Preferably, step S1 includes the following steps:

[0035] Step S11: Positioning key nodes of the sludge drying process in the sludge drying chamber and generating IoT monitoring node data;

[0036] Step S12: deploying an IoT monitoring network for the sludge drying chamber based on IoT monitoring node data to build an IoT monitoring network;

[0037] Step S13: setting a basic sampling period of 5 minutes, and using the Internet of Things monitoring network to periodically collect monitoring data from the sludge drying chamber based on the basic sampling period to generate multi-source original monitoring data;

[0038] Step S14: determining the current drying stage based on the multi-source original monitoring data, and adjusting the differential sampling frequency of the basic sampling period to generate sludge drying differential sampling data;

[0039] Step S15: Perform multi-source data preprocessing on the sludge drying differential sampling data to generate sludge status monitoring characteristic data; wherein, the sludge status monitoring characteristic data includes layered monitoring moisture content value, drying chamber monitoring temperature field, odor concentration monitoring data, monitoring air inlet velocity value and monitoring exhaust pressure difference value.

[0040] In an embodiment of the present invention, the key nodes of the sludge drying chamber process are located using the material-energy transfer gradient analysis method, dividing the drying chamber space into three-dimensional grid units of 10 cm × 10 cm × 10 cm. In each grid unit, initial temperature, humidity, and airflow test points are set, and initial data are collected using a thermocouple temperature sensor (measuring range 0-150°C, accuracy ±0.3°C), a capacitive humidity sensor (measuring range 0-100% RH, accuracy ±1.5% RH), and a hot ball wind speed sensor (measuring range 0-15m / s, accuracy ±0.05m / s). After 72 hours of continuous measurement, the heat transfer gradient value and moisture migration rate of each unit are calculated, and the locations where the heat transfer gradient change rate exceeds 4% / hour or the moisture migration rate exceeds 2% / hour are determined as key monitoring nodes. The spatial coordinates, key parameter type, parameter change rate, and priority score of each key node are recorded to form an IoT monitoring node dataset containing node number, precise positioning parameters, monitoring indicator type, and data transmission frequency requirements. Various sensors are fixedly installed at defined coordinates within the drying chamber. These include an array of PT100 platinum resistance temperature sensors (42 measurement points), an array of high-precision capacitive humidity sensors (36 measurement points), thermal gas velocity sensors (8 at the air inlet and 24 in the internal circulation area), micro-differential pressure sensors (12 in the exhaust system), and an array of electrochemical odor concentration sensors (6 in the exhaust system). All sensors are connected to a data acquisition unit via an RS485 bus. The acquisition unit utilizes a 24-bit high-precision analog-to-digital converter with a sampling rate of 100Hz. The data acquisition unit forms a ring-shaped redundant IoT network using industrial-grade switches, ensuring 99.99% network communication reliability. The central control unit utilizes a dual-machine hot backup architecture to receive, store, and process monitoring data in real time. It also implements sensor self-diagnosis and automatic isolation of failed nodes, forming a complete IoT monitoring network. During periodic monitoring data collection, a basic sampling period of 5 minutes is set, with synchronized sampling triggered across the entire network according to a unified clock. The temperature sensor uses a multiple sampling averaging method, sampling 12 times continuously within 5 seconds at each sampling point. The highest and lowest values ​​are removed and the arithmetic mean is taken. The humidity sensor samples the top, upper middle, lower middle, and bottom layers of the sludge, sampling each layer eight times continuously with a 2-second interval, and taking the weighted average. The airflow velocity sensor uses an orthogonal three-axis measurement method, sampling 10 times in each direction to synthesize a three-dimensional airflow velocity vector. The pressure difference sensor uses a pulsation compensation algorithm, sampling 50 times continuously at a frequency of 20Hz, and outputs after median filtering. The odor concentration sensor uses a temperature compensation circuit, sampling 15 times continuously, and outputs after temperature and humidity correction. All collected data is accompanied by sensor number, timestamp, spatial location identifier, measurement value, and unit information. It is transmitted in real time to a central database via industrial Ethernet, forming a multi-source raw monitoring data set containing temperature, humidity, airflow, pressure difference, and odor concentration fields.The determination of the drying stage adopts a multi-parameter comprehensive analysis method. First, calculate the moisture content change rate R1 = Δw / Δt (Δw is the difference in moisture content between two adjacent measurements, and Δt is the time interval), the temperature change rate R2 = ΔT / Δt (ΔT is the difference in temperature between two adjacent measurements), and the energy consumption index E = P / Δw (P is the input power per unit time, and Δw is the decrease in moisture content). When R1 > 3% / hour and R2 > 4°C / hour and E < 2.5 kWh / % are satisfied, it is determined as the high-speed dehydration stage, and the sampling period is adjusted to 2 minutes; when 1% / hour < R1 ≤ 3% / hour and 2°C / hour < R2 ≤ 4°C / hour and 2.5 kWh / % ≤ E < 4 kWh / % are satisfied, it is determined as the conventional dehydration stage, and the sampling period remains 5 minutes; when R1 ≤ 1% / hour and R2 ≤ 2°C / hour and E ≥ 4 kWh / % are satisfied, it is determined as the low-efficiency dehydration stage, and the sampling period is adjusted to 10 minutes. After the differential sampling frequency is adjusted, the monitoring data is re-collected according to the adjusted sampling period to form a sludge drying differential sampling data set with time-varying sampling interval marks. When performing multi-source data preprocessing on the sludge drying differential sampling data, the data cleaning, outlier correction, and feature extraction processes are executed. Cubic spline interpolation is used for data cleaning to handle missing values, and the mean value of the front and back data is used to replace the situation where more than three consecutive data points are missing. Moving median filtering is used for outlier correction, and the window width is 5 data points. If the deviation between the measured value and the median exceeds 3 times the standard deviation, it is determined as an outlier and replaced with the median value. Feature extraction includes: calculating the layered moisture content value, dividing the data of 24 moisture sensors into three layers of high, middle, and low according to the vertical height, and calculating the average value and standard deviation for each layer; constructing the temperature field of the drying chamber, generating a temperature field matrix of 10×10×3 grid points by three-dimensional Kriging interpolation using the data of 32 temperature monitoring points; reducing the dimension of the odor concentration data by principal component analysis, and extracting the first three principal components to characterize the odor characteristics; calculating the inlet air velocity value by weighted averaging the data of 12 wind speed sensors, and the weight coefficient is proportional to the air inlet area; calculating the maximum pressure difference, minimum pressure difference, and average pressure difference of the exhaust air pressure difference through the data of 8 pressure difference sensors. After all feature data is standardized, it constitutes the sludge status monitoring feature data set.

[0041] Preferably, the deployment of the Internet of Things monitoring network for the sludge drying chamber based on the data of the Internet of Things monitoring nodes includes:

[0042] Three layers of microwave moisture sensors are set in the sludge drying chamber, located at the sludge surface layer, the sludge middle layer, and the sludge bottom layer respectively. 5 measurement points are set for each layer. Among them, the measurement points on the sludge surface layer are 0 - 5 cm away from the surface, the measurement points on the sludge middle layer are 5 - 15 cm away from the surface, and the measurement points on the sludge bottom layer are 15 - 30 cm away from the surface;

[0043] 16 infrared temperature sensors are evenly arranged in the sludge drying chamber to form a 4×4 grid temperature measurement point array;

[0044] Three electrochemical odor concentration sensors are installed at the top, middle and bottom of the sludge drying chamber;

[0045] Install hot film air flow meter and differential pressure sensor at the inlet and outlet of ventilation duct respectively;

[0046] The microwave moisture content sensor, infrared temperature sensor, electrochemical odor concentration sensor, hot film air flow meter and differential pressure sensor are connected to the IoT edge computing controller located near the sludge drying chamber through wireless communication protocols to build an IoT monitoring network.

[0047] In this embodiment of the present invention, the microwave moisture content sensor installed in the sludge drying chamber uses a 10.5 GHz microwave probe with a measurement range of 0-100% and an accuracy of ±0.5%. Five sensors are installed in the sludge surface layer, fixed at a height of 3 cm from the surface; five sensors are installed in the middle sludge layer, fixed at a height of 10 cm from the surface; and five sensors are installed in the bottom sludge layer, fixed at a height of 25 cm from the surface. Each layer of five sensors is arranged in a "M" pattern along the length of the drying chamber, ensuring coverage of the four corners and the center of the drying chamber. The sensors are secured in place by a 316L stainless steel corrosion-resistant sleeve with a diameter of 48 mm and a wall thickness of 3 mm, meeting IP67 protection rating. The microwave probe is connected to a signal conditioning module via a coaxial cable. The signal conditioning module calculates moisture content using the frequency change formula ΔF = k × ΔH, where ΔF is the frequency change, k is the device calibration factor (10 MHz / %), and ΔH is the moisture content change. Each sensor is equipped with an independent 24V DC power supply and an RS485 communication interface with a baud rate of 9600bps. The infrared temperature sensor utilizes a high-precision thermopile detector with a wavelength of 8-14μm, a temperature measurement range of 0-150°C, and an accuracy of ±0.5°C. The 16 infrared temperature sensors are arranged in a 4×4 grid array, with the horizontal spacing between sensors equal to 1 / 5 of the drying chamber length and the vertical spacing equal to 1 / 5 of the drying chamber width. The sensors are fixed to the top of the drying chamber at a uniform height of 50cm from the sludge surface, ensuring a circular measurement area with a diameter of 15cm. The sensor housing is made of anodized aluminum, weighs 120g, and measures 45mm×30mm×20mm. A dust filter is included to prevent dust from contaminating the sensing element. Each infrared temperature sensor uses the Stefani-Boltzmann law to calculate the target temperature: T = (E / σε)(1 / 4), where T is the target temperature (K), E is the radiation energy received by the detector, and σ is the Stefani constant (5.67×10^-8W / m 2 ·K 4), ε is the emissivity of the sludge surface (value is 0.95). The sensor outputs a 4-20mA standard current signal, which is converted into a digital signal through an A / D converter with a resolution of 12 bits. The electrochemical odor concentration sensor adopts a three-electrode structure and has the ability to detect three main odor components: hydrogen sulfide (H2S), ammonia (NH3) and methyl mercaptan (CH3SH). The detection ranges are 0-100ppm, 0-50ppm and 0-10ppm, respectively, with accuracies of ±0.1ppm, ±0.5ppm and ±0.05ppm, respectively. A sensor is installed at the top of the drying chamber 30cm away from the sludge surface; a sensor is installed in the middle, located at 1 / 2 of the height of the drying chamber side wall from the bottom; and a sensor is installed at the bottom, located 10cm from the bottom of the drying chamber. The size of each sensor is It weighs 150g and is made of 316L stainless steel. The surface is Teflon treated to prevent corrosion. The sensor has a built-in temperature compensation circuit. The operating temperature range is -10℃ to 60℃. The temperature compensation formula is Cactual = Cmeasured × [1+α(T-20)], where Cactual is the actual concentration value, Cmeasured is the measured concentration value, α is the temperature coefficient (0.015 / ℃), and T is the ambient temperature (℃). The sensor is powered by 12V DC, has a power consumption of 0.5W, and the signal output is an RS485 interface. It uses the Modbus-RTU protocol and updates the data every 5 seconds. A hot film air volume meter is installed at the inlet of the ventilation duct. It uses the constant temperature hot film principle and has a measurement range of 0-50m 3 / min, accuracy ±1%, response time 0.1s. The diameter of the air flow meter probe is 1 / 3 of the inner diameter of the ventilation duct, and it is inserted vertically into the center of the duct. The probe is made of 316L stainless steel and sprayed with an alumina ceramic layer to prevent dust accumulation. The air flow meter is fixed by a flange connection. The flange standard is DN100 and the pressure level is PN16. The air flow meter uses the power balance formula Calculate the air volume, where Q is the air volume (m 3 / min), k is the equipment calibration factor (25.3), P is the heating power (W), ρ is the air density (kg / m 3). A differential pressure sensor is installed at the outlet of the ventilation duct, with a range of 0-1000Pa, an accuracy of ±0.5%, and a sensitivity of 0.1Pa. The differential pressure sensor uses a silicon piezoresistive sensing element. The two pressure interfaces are respectively connected to the inside and outside of the pipe. The pressure is drawn through a stainless steel capillary with a capillary diameter of 1.5mm and a length of 15cm. The differential pressure sensor outputs a 4-20mA current signal and adopts a two-wire connection method. The cable uses a shielded twisted pair cable with a grounded shielding layer and strong anti-interference ability. The IoT edge computing controller is installed in a protective cabinet within 5 meters of the sludge drying chamber. The protective cabinet has a protection level of IP65 and dimensions of 600mm×800mm×300mm. The controller uses an industrial-grade ARM processor with a main frequency of 1.2GHz, 4GB of memory, 32GB of storage, and runs a real-time operating system. Each sensor is connected to the controller via wireless communication protocols. The microwave moisture content sensor and infrared temperature sensor use the ZigBee protocol, operating at a 2.4 GHz frequency, a transmission rate of 250 kbps, and a communication range of 50 meters. The odor concentration sensor uses the LoRa protocol, operating at a 433 MHz frequency, a transmission rate of 5 kbps, and a communication range of 100 meters. The air flow meter and differential pressure sensor use the WiFi protocol, operating at a 2.4 GHz frequency, a transmission rate of 11 Mbps, and a communication range of 30 meters. Each sensor is equipped with an independent wireless communication module, which consumes 150 mW and is powered by a 3.6 V lithium battery with a capacity of 3600 mAh and an operating time of 90 days. The controller uses the AES-128 encryption algorithm to ensure data transmission security. The data frame format consists of a device ID (2 bytes), a command code (1 byte), a data length (1 byte), data content (n bytes), and a CRC checksum (2 bytes). The controller checks the communication status of all sensors in the network every 30 seconds, recording signal strength and packet loss rate.

[0048] Preferably, step S2 includes the following steps:

[0049] Step S21: extracting the moisture content monitoring values ​​of the surface layer, middle layer and bottom layer based on the layered monitoring moisture content values ​​in the sludge state monitoring characteristic data;

[0050] Step S22: performing moisture content distribution grid processing according to the moisture content monitoring value to generate sludge vertical moisture content distribution data;

[0051] Step S23: Classifying the water content values ​​of the grid points based on the sludge vertical water content distribution data and performing connectivity processing on each area to generate sludge water content zoning data; wherein, sludge beds with a water content greater than 80% are marked as high water content areas, sludge beds with a water content between 50% and 80% are marked as medium water content areas, and sludge beds with a water content less than 50% are marked as low water content areas;

[0052] Step S24: Evaluate the sludge drying index based on the sludge vertical moisture distribution data and the sludge moisture content partition data to generate a real-time sludge drying index;

[0053] Step S25: performing drying correction factor processing on the sludge moisture content partition data using the real-time sludge drying index to obtain the partition drying correction factor.

[0054] In this embodiment of the present invention, real-time data from three layers of microwave moisture sensors is acquired from an Internet of Things (IoT) monitoring network. Data from five measurement points in the surface layer (0-5 cm) are validated to remove outliers. The outlier determination criterion is: a single point reading is considered an outlier if it deviates by more than 8 percentage points from the average of other measurement points in the same layer. The remaining valid data is processed using a median filter algorithm with a filter window size of 5. After filtering, the surface layer moisture content average (HT) is obtained. Similarly, data from five measurement points in the middle layer (5-15 cm) and five measurement points in the bottom layer (15-30 cm) are subjected to outlier removal and median filtering, respectively, to obtain the middle layer moisture content average (HM) and the bottom layer moisture content average (HB). The standard deviations (SD_T, SD_M, and SD_B) of the moisture content of each layer are calculated to characterize the moisture content uniformity of each layer. Finally, a moisture content monitoring matrix for the surface, middle, and bottom layers is output, containing the original value, average, and standard deviation of each measurement point. When processing the moisture distribution grid based on the moisture monitoring values, a three-dimensional Kriging interpolation method is used to construct a high-precision spatial distribution grid. A three-dimensional spatial coordinate system is established for the drying chamber, with the X-axis being the length direction of the drying chamber, the Y-axis being the width direction of the drying chamber, and the Z-axis being the height direction of the drying chamber. Based on the actual spatial positions and moisture content values ​​of the 15 measurement points, an 8×8×6 grid array is constructed, with a grid resolution of 25 cm in the length direction, 25 cm in the width direction, and 5 cm in the height direction. Kriging interpolation uses the semivariogram function γ(h)=C0+C[1.5(h / a)―0.5(h / a) 3 ], when h≤a; γ(h)=C0+C, when h>a. Where h is the spatial distance, C0 is the nugget effect value (taken as 0.01), C is the base value (taken as 1), and a is the range (taken as 20cm). Interpolation calculation is performed on each grid node, and the node moisture content value H(x, y, z)=∑λ i ·h(x i ,y i , z i ), where λ i is the weight coefficient of the i-th known point, h(x i ,y i , z i) is the moisture content value of the i-th known point. A three-dimensional data matrix with 384 grid nodes is generated to form the sludge vertical moisture content distribution data. Each element in the matrix represents the interpolated moisture content value at the corresponding position. Traverse the data of 384 grid nodes and perform zoning marking according to the moisture content threshold. Grid points with a moisture content greater than 80% are marked as high moisture content areas (area code H) and assigned a value of 3; grid points with a moisture content between 50% and 80% are marked as medium moisture content areas (area code M) and assigned a value of 2; grid points with a moisture content less than 50% are marked as low moisture content areas (area code L) and assigned a value of 1. After the preliminary classification is completed, perform regional connectivity processing and use the six-connected component method to judge the spatial regional connectivity. If the moisture content difference between adjacent grid points is less than 5 percentage points, it is determined to be in the same connected domain. For isolated areas with an area less than 5% of the total number of grids, use the region growing algorithm to merge them into the adjacent largest area, and the region growing algorithm uses the minimum Euclidean distance criterion. The connected domain marking is implemented using the depth-first search algorithm, with a search radius of 1 grid unit and a search depth of 6 layers. Finally, a sludge moisture content zoning data structure containing the region identification code, the proportion of the region volume, and the centroid coordinates of the region is generated. First, calculate the moisture content gradient index GH, GH = ∑|H(i, j, k)―H(i, j, k + 1)| / (N―1), where H(i, j, k) represents the moisture content value at the position (i, j, k), and N is the number of grid points in the vertical direction. Second, calculate the zoning ratio index PR, PR = (VL + 0.6VM + 0.2VH) / Vtotal, where VL, VM, and VH are the volumes of the low, medium, and high moisture content areas respectively, and Vtotal is the total volume of the sludge. Then calculate the drying uniformity index EU, EU = 1―σ / Haverage, where σ is the standard deviation of the moisture content of all grid points, and Haverage is the average moisture content. Finally, calculate the interlayer difference index LD, LD = (HT - HB) / HM, where HT, HM, and HB are the average moisture contents of the surface layer, middle layer, and bottom layer respectively. The real-time sludge drying index DI is obtained by weighted summation: DI = 0.3GH + 0.4PR + 0.2EU + 0.1LD, and the value range of DI is 0-1. The larger the value, the better the drying effect. A piecewise linear mapping function is used. For the high moisture content area (area code H), the drying correction factor FH = 1.5 - 0.5DI when DI ≤ 0.6; FH = 1.8 - 1.0DI when DI > 0.6. For the medium moisture content area (area code M), the drying correction factor FM = 1.0 when DI ≤ 0.3; FM = 1.0 + 0.5(DI - 0.3) when 0.3 < DI ≤ 0.7; FM = 1.2 when DI > 0.7. For the low moisture content area (area code L), the drying correction factor FL = 0.8 when DI ≤ 0.4; FL = 0.8 - 0.५(DI - 0.4) when 0.4 < DI ≤ 0.8; FL = 0.6 when DI > 0.8.For areas at the edge of the drying chamber (less than 10 cm from the chamber wall), an additional edge correction factor, FE = 1.1, is applied to correct for drying non-uniformity. For each zone, the drying energy allocation ratio, P = F × V / ∑ (F × V), is calculated, where F is the drying correction factor for each zone and V is the volume of each zone. A table of zone drying correction factors is generated, including zone number, zone location, correction factor value, and energy allocation ratio.

[0055] Preferably, evaluating the sludge drying index using the sludge vertical moisture distribution data and the sludge moisture content partition data includes:

[0056] Calculate the average sludge moisture content based on the sludge vertical moisture distribution data as the overall moisture content assessment value;

[0057] According to the vertical moisture distribution data of sludge, the moisture content difference of different depths of sludge layer is analyzed to obtain the vertical moisture gradient data;

[0058] Identify the uneven distribution of sludge moisture content based on the sludge moisture content partition data and generate the sludge moisture content unevenness;

[0059] A real-time sludge drying index is evaluated based on the overall moisture content assessment value, vertical moisture content gradient data, and sludge moisture content unevenness to generate a real-time sludge drying index.

[0060] In the embodiment of the present invention, when calculating the average moisture content of sludge based on the vertical moisture distribution data of sludge, the weighted average method is used to process the 384 node data in the 8×8×6 grid array. First, the grid data is divided into 6 layers according to the vertical height, each layer is 8×8 and has a total of 64 nodes. The arithmetic mean H is calculated for each layer of data. i =∑H(x, y, z i ) / (8×8), where H(x, y, z i ) is the coordinate (x, y, z i), i is the layer number (1-6). Then, considering the volume distribution characteristics of the sludge, the layer weight coefficient W_i is introduced. The weight of the surface layer (i=1) is 0.1, the weight of the subsurface layer (i=2) is 0.15, the weight of the upper middle layer (i=3) is 0.2, the weight of the lower middle layer (i=4) is 0.25, the weight of the upper bottom layer (i=5) is 0.2, and the weight of the bottom layer (i=6) is 0.1. The overall moisture content assessment value H_avg is calculated by weighted average: H_avg=∑(W_i×H_i), where ∑W_i=1. During the calculation process, if it is found that the data of a certain layer is missing by more than 50%, the weights of the adjacent layers are adjusted to compensate to ensure the representativeness of the assessment value. The final H_avg is accurate to one decimal place and is expressed in percentage. When analyzing the difference in moisture content at different depths based on the vertical moisture distribution data of the sludge, the interlayer gradient calculation method is used. First, calculate the moisture content difference between adjacent layers: ΔH_i = H_i - H_(i+1), where i = 1, 2, 3, 4, 5, where H_i is the average moisture content of layer i. Then, calculate the standardized gradient: G_i = ΔH_i / d_i, where d_i is the distance between the center points of the two layers, in centimeters. For example, d_1 = 5 cm between the surface and subsurface layers, d_2 = 5 cm between the subsurface and upper-middle layers, d_3 = 5 cm between the upper-middle and lower-middle layers, d_4 = 5 cm between the lower-middle and upper-bottom layers, and d_5 = ​​5 cm between the upper-bottom and bottom layers. Furthermore, calculate the absolute gradient sum: G_sum = ∑|G_i|, ​​which represents the overall vertical gradient strength. Calculate the gradient direction consistency index: C_d = |∑G_i| / G_sum, with a value range of [0, 1]. Values ​​closer to 1 indicate more consistent gradient direction. Calculate the gradient fluctuation index F_g = σ(G_i) / μ(G_i), where σ(G_i) is the gradient standard deviation and μ(G_i) is the gradient mean. Finally, the integrated parameters generate a vertical moisture gradient data matrix, which contains four sets of data: gradient values ​​of each layer, gradient sum, directional consistency and fluctuation index. Calculate the volume ratio of each partition, the volume ratio of the high water content area P_H = V_H / V_total, the volume ratio of the medium water content area P_M = V_M / V_total, and the volume ratio of the low water content area P_L = V_L / V_total, where V_H, V_M, and V_L are the volumes of the high, medium, and low water content areas, respectively, and V_total is the total sludge volume. Calculate the ideal uniform distribution deviation D_p = ∑|P_i-1 / 3|, with a value range of [0, 4 / 3]. Secondly, the spatial Gini coefficient G_s is used to calculate the degree of inequality in the moisture content distribution of the grid points: G_s = ∑∑|H_i-H_j| / (2n 2μ), where \(H_i\) and \(H_j\) are the moisture content values of any two points, \(n\) is the total number of grid points, which is 384, and \(\mu\) is the average moisture content. The dispersion degree \(S_d\) of the calculation area is \(N_c / N_t\), where \(N_c\) is the number of connected domains and \(N_t\) is the theoretical maximum number of connected domains, which is 3. The edge non-uniformity \(E_n=\sum(|H_e - H_c|) / n_e\), where \(H_e\) is the moisture content of the edge grid points, \(H_c\) is the average moisture content of the central area, and \(n_e\) is the number of edge grid points. The final non-uniformity \(U\) of the sludge moisture content is obtained by weighted summation: \(U = 0.3D_p+0.3G_s+0.2S_d+0.2E_n\), with a value range of \([0, 1]\). The overall moisture content evaluation value \(H_avg\) is normalized to the moisture content compliance index \(I_h=(H_0 - H_avg) / (H_0 - H_t)\), where \(H_0\) is the initial moisture content (taking 85%), and \(H_t\) is the target moisture content (taking 35%). When \(H_avg>H_0\), \(I_h = 0\); when \(H_avg < H_t\), \(I_h = 1\). Secondly, the vertical moisture content gradient data is converted into the gradient index \(I_g=(G_max - G_sum) / (G_max - G_min)\), where \(G_sum\) is the total gradient, \(G_max\) is the gradient upper limit (taking 5% / cm), and \(G_min\) is the gradient lower limit (taking 0.5% / cm). Then, the sludge moisture content non-uniformity \(U\) is converted into the uniformity index \(I_u = 1 - U\). Finally, the real-time sludge drying index \(DI\) is calculated by non-linear weighted summation: \(DI=w_h\times I_h+w_g\times I_g\times(1 - I_h)+w_u\times I_u\times I_h\), where \(w_h\), \(w_g\), and \(w_u\) are the moisture content index weight (0.6), the gradient index weight (0.25), and the uniformity index weight (0.15) respectively. The real-time sludge drying index \(DI\) has a value range of \([0, 1]\), and the higher the value, the better the drying effect, and the closer the sludge state is to the ideal drying target.

[0061] Preferably, in step S3, the sludge surface evaporation rate analysis is carried out based on the sludge organic matter content and the sludge density, and the deduction of the change in the moisture content of the sludge layer includes:

[0062] Extract the sludge surface moisture content from the stratified monitoring moisture content values in the sludge state monitoring characteristic data;

[0063] Use the monitored temperature field in the drying chamber, the monitored inlet air velocity value, and the monitored exhaust air pressure difference in the sludge state monitoring characteristic data to calculate the surface moisture evaporation rate of the sludge surface, and generate the sludge surface evaporation rate;

[0064] Match the heat and mass transfer equilibrium parameters based on the sludge organic matter content and the sludge density; [[ID=!14]]

[0065] Calculate the heat energy received by the sludge according to the monitored temperature field in the drying chamber in the sludge state monitoring characteristic data, and generate the sludge received heat energy value;

[0066] Based on the layered monitoring moisture content values ​​in the sludge status monitoring characteristic data, the moisture content difference between the inner and outer layers is analyzed, and the internal moisture migration rate is calculated based on the heat and mass transfer balance parameters and the sludge receiving heat energy value;

[0067] The heat required for water evaporation is estimated by the evaporation rate of the sludge surface and the internal water migration rate based on the heat energy value received by the sludge, and the change in moisture content of the sludge layer is deduced to generate moisture content change trend data.

[0068] In an embodiment of the present invention, the raw data of the five surface sensors are extracted, and the readings of each measuring point are subjected to a moving average process of six consecutive measurements with a sampling interval of 10 seconds to eliminate the influence of instantaneous fluctuations. The standard deviation filtering method is applied to remove abnormal points, and the standard deviation SD of the five measuring points is calculated. When the reading of a certain point deviates from the average value by more than 2.5×SD, the data of that point is eliminated. The remaining valid measuring point data are synthesized into the surface moisture content representative value Hs by the inverse distance weighted average method. The calculation formula is Hs = ∑(wi×Hi) / ∑wi, where Hi is the moisture content value of the i-th valid measuring point, wi is the weight coefficient of the i-th measuring point, and wi = 1 / di 2 , di is the horizontal distance from the measuring point to the center of the drying chamber. The final sludge surface moisture content is accurate to one decimal place and is expressed in percentage. When calculating the surface moisture evaporation rate of the sludge surface moisture content using the drying chamber monitoring temperature field, monitoring air inlet velocity value and monitoring exhaust pressure difference, the improved evaporation dynamics model is applied. First, extract the 16 temperature measurement point data at a height of 0-10cm from the monitoring temperature field, and calculate the average surface temperature Ts. Calculate the wind speed vector Vs from the monitoring air inlet velocity value, where Vs is the air inlet wind speed, in m / s. Calculate the pressure difference ΔP inside and outside the drying chamber from the monitoring exhaust pressure difference, in Pa. The evaporation rate is calculated using the empirical formula E=k×(Xs-X)×Vs^0.8, where E is the evaporation rate (kg / m 2 ·h), k is the evaporation coefficient (value is 1.2×10^-3), Xs is the saturated absolute humidity at the surface temperature, and X is the absolute humidity of the inlet air. Xs is calculated by looking up the Antoine equation through Ts: Xs=0.622×Ps / (P-Ps), where Ps is the saturated water vapor pressure at the surface temperature (Pa), and P is the atmospheric pressure (Pa). The calculation formula for Ps is: Ps=A-B / (C+Ts), A=16.26, B=3799.89, C=226.35, and Ts is in °C. The final sludge surface evaporation rate E is generated. A segmented fitting model is used. The sludge organic matter content is expressed by the ratio of volatile solids (VS) to total solids (TS). VS / TS is obtained from the sludge feed detection data, with a value range of 0.4-0.8. The sludge density ρs is determined by the bulk density method, with the unit of kg / m 3. Heat and mass transfer parameters include: effective thermal conductivity λe, specific heat capacity Cp, mass diffusion coefficient Dm, and heat transfer coefficient h. The calculation formula for the effective thermal conductivity λe is: λe=λw×Mw+λs×(1-Mw), where λw is the thermal conductivity of water (0.6W / m·K), λs is the thermal conductivity of dry sludge (0.2W / m·K), and Mw is the moisture content. The calculation formula for the specific heat capacity Cp is: Cp=4186×Mw+[1464+2093×(VS / TS)]×(1-Mw), unit J / kg·K. The calculation formula for the mass diffusion coefficient Dm is: Dm=D0×exp[―E / (R×T)]×(1―Mc) n , where D0 is the reference diffusion coefficient (5.77×10^-4m 2 / s), E is the diffusion activation energy (4200 J / mol), R is the gas constant (8.314 J / mol·K), T is the absolute temperature (K), Mc is the critical water content (value 0.3), and n is the empirical exponent (value 3.0). The heat transfer coefficient h is calculated as: h = 2.8 + 4.8 × Vs, unit W / m 2 ·K. Extract the data of 16 temperature sensors on the four walls of the drying chamber from the temperature field of the drying chamber, and calculate the average temperature Tw of the four walls. Extract the inlet air temperature Ti and the exhaust air temperature To from the temperature field of the drying chamber. Calculate the air volume Q from the monitored inlet air velocity value, Q = v × A, where v is the wind speed (m / s) and A is the cross-sectional area of ​​the air duct (m 2 The heat energy Qtotal received by sludge consists of three parts: convection heat transfer Qconvection, radiation heat transfer Qradiation and hot air transport Qwind. Convection heat transfer Qconvection = h×As×(Tw-Ts), where h is the heat transfer coefficient (W / m 2 ·K), As is the sludge surface area (m 2 ), Tw is the bulkhead temperature (K), Ts is the sludge surface temperature (K). Radiation heat transfer Q radiation = σ × ε × As × (Tw^4-Ts^4), where σ is the Stefan-Boltzmann constant (5.67×10^-8W / m 2 ·K^4), ε is the effective radiation coefficient (value is 0.85). Hot air transport Qwind=ρa×Cp,a×Q×(Ti-To), where ρa is the air density (kg / m 3 ), Cp,a is the specific heat capacity of air (1005J / kg·K), Q is the air volume (m 3 / s), Ti is the inlet air temperature (K), To is the exhaust air temperature (K). The thermal energy value received by the sludge Qtotal = Qconvection + Qradiation + Qwind, unit W. Calculate the average moisture content difference ΔMw of the surface layer (0-5cm), middle layer (5-15cm) and bottom layer (15-30cm), ΔMw = Mint-Msurf, where Mint is the average moisture content of the middle layer and the bottom layer, and Msurf is the moisture content of the surface layer. The internal moisture migration rate w is calculated using Fick's first law: w = -Dm×(dMw / dx), where w is the moisture migration rate (kg / m 2 ·h), Dm is the mass diffusion coefficient (m 2 / h), dMw / dx is the moisture content gradient (kg / kg·m). The Dm value depends on the sludge temperature T, moisture content Mw and organic matter content VS / TS, and is obtained by matching the diffusion coefficient formula in the heat and mass transfer balance parameters. The moisture content gradient dMw / dx is approximately ΔMw / Δx, where Δx is the effective distance from the surface to the interior (value 0.1m). The assessment of water migration flux needs to consider the migration resistance factor R, which increases as the moisture content decreases. R = exp[α×(M0-Mw)], where α is the empirical coefficient (value 5.0) and M0 is the initial moisture content. The corrected internal moisture migration rate wc = w / R, unit kg / m 2 h. Apply the layered heat balance method. First calculate the heat L required for evaporation of unit mass of water, L = 2500-2.4×Ts, unit kJ / kg, Ts is the surface temperature (℃). The heat consumed by surface evaporation Qevaporation = E×As×L, where E is the surface evaporation rate (kg / m 2 ·h), As is the surface area (m 2 ). Internal water migration consumes heat Q migration = wc × As × ΔH, where wc is the internal water migration rate (kg / m 2 h), ΔH is the internal water migration enthalpy change (valued at 40 kJ / kg). Effective thermal energy utilization rate η = Qeffective / Qtotal, where Qeffective = Qevaporation + Qmigration, and Qtotal is the thermal energy received by the sludge. Surface moisture content change rate dMsurf / dt = -(E-wc) / (ρs×δ×(1-Msurf)), where ρs is the sludge density (kg / m 3 ), where δ is the thickness of the surface layer (0.05 m). The rate of change of the moisture content in the middle layer is dMmid / dt = -wc / (ρs × δmid × (1-Mmid)), where δmid is the thickness of the middle layer (0.1 m). By recursively calculating with a time step of Δt = 1 h, the moisture content change trends of the three layers over the next 24 hours are determined, generating a moisture content change trend data matrix.

[0069] Preferably, in step S3, using the partition drying correction factor and the moisture content change trend data to perform partition coordinated control processing includes:

[0070] Perform drying uniformity target processing according to the partition drying correction factor to generate drying uniformity target data;

[0071] Evaluate the drying completion time based on moisture content trend data;

[0072] The drying temperature is set based on the sludge moisture content partition data according to the drying completion time and drying uniformity target data to generate sludge drying temperature data;

[0073] Based on the drying uniformity target data and sludge drying temperature data, preliminary zone drying parameter configuration is performed to generate differentiated sludge drying parameters;

[0074] The differentiated sludge drying parameters are subjected to zone-by-zone collaborative regulation and control to generate zone-by-zone collaborative regulation parameters.

[0075] In the embodiment of the present invention, a uniformity objective function U=∑|Fi-F-| / n is constructed, where Fi is the drying correction factor of the i-th region, F- is the average of the drying correction factors of all regions, and n is the total number of regions. The smaller the objective function value, the more uniform the drying. The uniformity convergence threshold Uth=0.15 is set. When U>Uth, the adaptive adjustment algorithm of the partition drying correction factor is used to gradually adjust the correction factor of each region. The adjustment formula is: Where Fi' is the adjusted correction factor, and α is the convergence coefficient (valued at 0.6). An additional correction value δH = 0.2 is added to the correction factor for high water content areas (water content > 80%), and a correction value δL = 0.1 is reduced to the correction factor for low water content areas (water content < 50%). Through iterative calculation (maximum number of iterations 10 times), the uniformity objective function U converges to below Uth. Finally, the drying uniformity target data is generated, including the balanced correction factor for each region, the regional uniformity index and the convergence trajectory record. 24-hour forecast data points are extracted from the water content change trend data to construct the water content time series {t_i, H_i} for each region, where t_i is the time point (h) and H_i is the water content (%) at the corresponding moment. The data were fitted with an exponential decay function: H(t) = H_∞ + (H_0 - H_∞) × exp(-kt), where H_∞ is the equilibrium moisture content (30%), H_0 is the initial moisture content, and k is the drying rate constant (h^-1). The k value was determined using the least squares method, and the goodness of fit was measured using R 2 Value evaluation, requires R 2 >0.95. For poorly fitted data (R 2<0.95), a piecewise linear fitting method was used instead. A moisture content threshold for completion of drying, H_t = 40%, was defined. H(t) = H_t was substituted into the fitting equation to solve for the estimated time t_i for each zone to reach the target moisture content. The drying completion time, T_c, was taken as the longest drying time for each zone: T_c = max{t_i}. The standard deviation of drying times between zones, σ_t, was also calculated to assess drying uniformity. A drying process with σ_t < 6 hours was considered uniform. A zoned dynamic temperature control strategy was adopted. The sludge was divided into three zones: high moisture content, medium moisture content, and low moisture content, with corresponding benchmark temperatures, T_H = 90°C, T_M = 75°C, and T_L = 60°C, respectively. The temperature was adjusted based on the zone-by-zone equalization correction factor, F_i, in the drying uniformity target data: T_i = T_baseline × (1 + 0.15 × (F_i - 1)), where T_i is the actual set temperature for zone i and T_baseline is the benchmark temperature for that zone. The temperature compensation coefficient η is set based on the drying completion time T_c. When T_c > 24 hours, η = 1.1; when 12 hours ≤ T_c ≤ 24 hours, η = 1.0; and when T_c < 12 hours, η = 0.9. The final temperature of each zone is calculated using the formula: T_final = T_i × η × β, where β is a safety factor (set to 0.95) to ensure the temperature does not exceed 95°C. For edge zones (<15 cm from the bulkhead), an edge compensation temperature of 5°C is added to prevent heat loss. The generated sludge drying temperature data includes the zone number, zone type, set temperature, compensation coefficient, and final control temperature. The temperature control parameters for each zone are determined: the zone temperature set value T_set is the final control temperature in the sludge drying temperature data; the temperature fluctuation range ΔT = ±3°C; the ramp rate R_up = 5°C / min; and the ramp rate R_down = 2°C / min. Next, the airflow control parameters were determined: the airflow velocity V was graded based on regional moisture content: V_H = 2.5 m / s for the high moisture content zone, V_M = 2.0 m / s for the medium moisture content zone, and V_L = 1.5 m / s for the low moisture content zone. The airflow direction angle θ was adjusted based on the regional location: θ_top = 30° for the top zone, θ_mid = 0° for the middle zone, and θ_bottom = -30° for the bottom zone. The drying cycle parameters were then determined: the heating time t_heat to the heating pause time t_pause ratio was 4:1, with t_heat = 40 minutes and t_pause = 10 minutes for the high moisture content zone; t_heat = 30 minutes and t_pause = 8 minutes for the medium moisture content zone; and t_heat = 20 minutes and t_pause = 5 minutes for the low moisture content zone. Finally, the differentiated sludge drying parameters were generated, including the regional temperature control parameter matrix, the airflow control parameter matrix, and the drying cycle parameter matrix. The temperature parameter gradient between adjacent areas is restricted. When |ΔT_adjacent|>15℃, it is processed through the temperature transition zone: T_transition = w_1T_1+w_2T_2, where w_1 and w_2 are distance weighted coefficients, satisfying w_1+w_2=1.Next, zoned drying energy balance control is implemented. Total energy consumption is balanced by adjusting the drying intensity coefficients S_i for different zones, using the formula S_i = (T_i × V_i × t_heat, i) / (T_avg × V_avg × t_heat, avg). The energy allocation ratio is adjusted until the drying rates across zones converge. Key zone priority control is then implemented. Using a matrix priority assignment P, high-water-content zones P_H = 3, medium-water-content zones P_M = 2, and low-water-content zones P_L = 1. Zones with high priority receive a 1.2x coefficient in energy allocation. Finally, operating cycle collaborative optimization is performed, staggering the drying cycles of adjacent zones to avoid overlapping peak energy consumption. The staggered interval is calculated using the formula t_staggered = t_heat / n, where n is the number of zones. This final zoned collaborative control parameter is formed, including smooth transition parameters, energy balance parameters, priority parameters, and staggered scheduling parameters.

[0076] It is particularly important to coordinate and regulate the differentiated sludge drying parameters in different zones, including:

[0077] Based on the differentiated sludge drying parameters, the sludge drying chamber is mapped into independent physical control areas to obtain the drying chamber partition control areas;

[0078] According to the differentiated sludge drying parameters and the drying chamber partition control area, the heat energy output ratio of the waste heat recovery and auxiliary heating devices used in each control area during the operation phase is determined to obtain the initial heat source ratio of the partition;

[0079] Calculate the average and maximum concentration readings based on the monitored odor concentration in the sludge status monitoring characteristic data to generate odor concentration monitoring data;

[0080] The proportion of fresh air to total ventilation volume is set based on the real-time air inlet velocity value and the real-time exhaust pressure difference value for the odor concentration monitoring data to obtain the fresh air mixing ratio;

[0081] The airflow mode switching cycle is set for the differentiated sludge drying parameters through the fresh air mixing ratio to generate airflow switching cycle data;

[0082] The sludge differential drying parameters, the initial heat source ratio of the partition, the fresh air mixing ratio and the airflow switching cycle data are used as the partition coordinated control parameters.

[0083] In this embodiment of the present invention, a rectangular coordinate system (x, y, z) is established for the drying chamber, with the origin located at the lower left corner of the drying chamber. The x-axis represents length, the y-axis represents width, and the z-axis represents height. The x, y, and z dimensions are divided into 5, 4, and 3 segments, respectively, in a ratio of 5:4:3, forming a total of 60 basic control units. Units are merged based on the temperature control parameter matrix within the sludge differential drying parameters. Adjacent units with a temperature difference of less than 5°C are merged into the same control region. A region-growing algorithm is used for unit merging, with the seed point selected as the moisture content extreme point of each region. The number of merged control regions is controlled between 5 and 8, ensuring that each control region accounts for no less than 8% of the total drying chamber area. Each control region is assigned a unique identification code in the form of "H / M / L-number," where H / M / L represents the moisture content level and the number represents the region number. Finally, a drying chamber zoning control region map is generated, containing each region's 3D boundary coordinates, volume, moisture content characteristics, and number. A dual-heat source configuration model was constructed to determine the heat output ratio of each control zone based on the differentiated sludge drying parameters and the zoning of the drying chamber. The system is equipped with a waste heat recovery device (heat source A) and an auxiliary electric heating device (heat source B), with a total heat output controlled within 300 kW. The waste heat recovery device utilizes waste heat from the wastewater treatment plant's biogas boiler, with a thermal efficiency of 65% and a maximum thermal output of 200 kW. The auxiliary electric heating device utilizes a three-phase 380V electric heating tube array with a thermal efficiency of 98% and a maximum thermal output of 150 kW. The heat source ratio R is defined as the percentage of the waste heat recovery device's heat output to the total heat output. The initial heat source ratio RH for the high-water content zone (Class H) is calculated as: RH = 40 + 0.5 × TH, where TH is the set temperature for the zone and ranges from 0 to 100. The initial heat source ratio RM for the medium-water content zone (Class M) is = 60 + 0.25 × TM, where TM is the set temperature for the zone. The initial heat source ratio for the low-water zone (Class L) is fixed at RL = 80. The actual heat energy allocation to each zone is calculated using the following formulas: QA,i = Qi × Ri / 100, QB,i = Qi × (1-Ri / 100), where QA,i and QB,i represent the waste heat recovery and auxiliary heating energy allocated to zone i, respectively, and Qi represents the total heat demand for zone i, determined by the temperature parameter in the differentiated drying parameters. A table of initial heat source ratios for each zone is generated, containing the zone number, total heat demand, allocation of heat source A, allocation of heat source B, and the allocation percentage. Real-time concentration data for hydrogen sulfide (H2S), ammonia (NH3), and methyl mercaptan (CH3SH) are collected from three electrochemical odor concentration sensors located at the top, middle, and bottom of the drying chamber, with a sampling frequency of 30 seconds. Each sensor outputs three gas concentration values, totaling nine data points.The raw concentration data is exponentially smoothed using the formula C_t = αC_measured + (1-α)C_(t-1), where α is the smoothing coefficient (valued at 0.3), C_t is the smoothed concentration value at time t, C_measured is the measured concentration value, and C_(t-1) is the smoothed concentration value at the previous moment. The weighted average concentration of each gas is calculated, with the weight coefficient set according to the gas toxicity index: H2S has a weight of 5, NH3 has a weight of 1, and CH3SH has a weight of 10. The average concentration calculation formula is: C_average = (5C_H2S + 1C_NH3 + 10C_CH3SH) / 16, where C_H2S, C_NH3, and C_CH3SH are the average concentration values ​​(ppm) of the three gases, respectively. The maximum concentration reading uses the maximum weighted concentration value within the last 30 minutes. The final odor concentration monitoring data is generated, including instantaneous concentration, average concentration, maximum concentration, and concentration change rate. Calculate the normalized odor index (OI): OI = C_average / C_threshold, where C_average is the average odor concentration (ppm) and C_threshold is the odor control threshold (set at 10 ppm). Wind speed is then categorized by inlet air velocity (V) (m / s): V < 1.0 for weak wind; 1.0 ≤ V < 2.0 for moderate wind; and V ≥ 2.0 for strong wind. Ventilation effectiveness is also categorized by exhaust pressure difference (ΔP) (Pa): ΔP < 50 for weak wind; 50 ≤ ΔP < 100 for moderate wind; and ΔP ≥ 100 for strong wind. The formula for calculating the baseline fresh air mixing ratio (R_Fresh Air) is: R_Fresh Air = 40 + 30 × OI, with a range of [40% to 85%]. The reference value is adjusted based on wind speed and ventilation effectiveness: For strong wind with strong effect, R_Fresh Air = R_Fresh Air × 0.8; for strong wind with medium effect or medium wind with strong effect, R_Fresh Air = R_Fresh Air × 0.9; for weak wind with weak effect, R_Fresh Air = R_Fresh Air × 1.2; in other cases, R_Fresh Air remains unchanged. Air volume correction factor. Where V i is the real-time air inlet velocity, and ΔP is the real-time exhaust pressure difference. Ensure that the proportion of fresh air does not exceed 85%. When setting the airflow mode switching cycle for the differentiated sludge drying parameters based on the fresh air mixing ratio, a dual-mode alternating control mechanism is implemented. The drying system has two airflow modes: Mode A is a horizontal airflow (parallel to the sludge surface), and Mode B is a longitudinal airflow (perpendicular to the sludge surface). The benchmark switching cycle T0 is set to 60 minutes, that is, a complete cycle of Mode A → Mode B → Mode A is completed every 60 minutes. The relationship formula between the airflow switching cycle T and the fresh air mixing ratio R_Fresh Air is: T = T0 × (1.5-0.5 × R_Fresh Air / 100), when R_Fresh Air>60%; T = T0, when R_Fresh Air≤60%. The duration of airflow mode A TA = T × KA, the duration of airflow mode B TB = T × KB, where KA and KB are time distribution coefficients, satisfying KA + KB = 1. For high-water-content zones, KA = 0.3 and KB = 0.7; for medium-water-content zones, KA = 0.5 and KB = 0.5; and for low-water-content zones, KA = 0.7 and KB = 0.3. The switching start times for each control zone are staggered, with the difference in start times between adjacent zones being T / n, where n is the total number of zones. The smooth transition time for the mode switching process is set to 30 seconds, and linearly increasing / decreasing wind speed control is employed. A data table for the airflow switching cycle is generated, containing the cycle length, mode A duration, mode B duration, and switching start time for each zone. By integrating the differentiated sludge drying parameters, the initial heat source ratio for each zone, the fresh air mixing ratio, and the airflow switching cycle data into zone-by-zone coordinated control parameters, a multidimensional parameter matrix is ​​constructed. First, a zone index table is established, containing the unique identification codes, physical boundary coordinates, percentage of total volume, and corresponding sensor numbers for all control zones. Then, a real-time control parameter matrix is ​​constructed. The row vectors are the control areas, and the column vectors include: temperature setting value (℃), temperature fluctuation range (℃), heating rate (℃ / min), waste heat utilization ratio (%), auxiliary heating power (kW), fresh air mixing ratio (%), airflow mode code (A / B), mode switching time point (min), ventilation rate (m 3 / h). Matrix elements are updated every 5 seconds. A weight coefficient column is added to the parameter matrix to indicate the priority of parameter adjustment for each region. The high water content region has a weight of 1.5, the medium water content region has a weight of 1.0, and the low water content region has a weight of 0.5. The integrated parameters are finally adjusted through the regional collaborative optimization algorithm to ensure that the parameters of each region meet the global constraints: the total heat input does not exceed 300kW, and the total air volume does not exceed 5000m 3 / h, with the temperature difference between adjacent areas not exceeding 25°C. Ultimately, a partitioned collaborative control parameter data package containing 112 control variables is formed for real-time call by the execution layer PLC controller.

[0084] Preferably, the risk analysis of odor exceeding the standard based on the moisture content change trend data in step S3 includes:

[0085] Process the drying stage path based on the drying completion time and moisture content change trend data to generate predicted drying stage data;

[0086] Analyze the drying temperature change rate and sludge drying speed based on the predicted drying stage data;

[0087] Analyze odor components based on the organic matter content of sludge and generate odor component data;

[0088] Evaluate the odor release potential based on the drying temperature change rate and sludge drying speed using odor component data to generate odor release potential evaluation data;

[0089] Odor emission analysis is performed on the odor release potential assessment data through zoning coordinated control parameters, and odor concentration retention is calculated based on the predicted drying stage data to obtain the predicted odor concentration change curve;

[0090] Adaptive dry odor warning processing is performed based on the predicted odor concentration change curve to generate odor exceeding standard risk data.

[0091] In the embodiment of the present invention, the drying process is divided into three typical stages: preheating stage (I), constant rate drying stage (II) and speed-down drying stage (III). By analyzing the first-order derivative of moisture content dH / dt in the moisture content change trend data, the stage judgment rules are defined: when dH / dt>-0.2% / h, it is classified as the preheating stage; when -1.5% / h≤dH / dt≤-0.2% / h and the change rate fluctuation is less than 0.1% / h, it is classified as the constant rate drying stage; when dH / dt<-1.5% / h or the moisture content is less than 60% and the change rate continues to decrease, it is classified as the speed-down drying stage. A sliding window analysis (window width 4 hours) is performed on the moisture content change trend matrix to construct a drying stage conversion model. The stage duration calculation formula is: tpreheat = 5 + 0.2 × Hinitial, tconstant speed = (Hinitial - 60) / rconstant speed, and treduction speed = (60 - Hfinal) / rreduction speed, where Hinitial is the initial moisture content (%), Hfinal is the target moisture content (%), rconstant speed is the drying rate during the constant speed stage (% / h), and rreduction speed is the drying rate during the reduction stage (% / h). Time series mapping is used to generate complete predicted drying stage data, including the start and end times, duration, and corresponding moisture content ranges of each stage. The drying temperature field is constructed using real-time data from an infrared temperature sensor array, and the average temperature T and standard deviation σT are calculated. The drying temperature change rate dT / dt is calculated using the central difference method: dT / dt = (T(t+Δt)-T(t-Δt)) / (2Δt), where Δt is the time step (5 minutes). For the preheating stage (I), dT / dt shows an increasing trend, with a typical value of 3-5℃ / h; in the constant speed stage (II), dT / dt is close to constant, with a typical value of 0-1℃ / h; in the deceleration stage (III), dT / dt rises slowly, with a typical value of 1-2℃ / h. The sludge drying rate v is defined as the percentage reduction in moisture content per unit time, and the calculation formula is: v = -dH / dt, where dH / dt is the rate of change of moisture content (% / h). A correlation model between drying temperature and drying rate is established: v = a×(T-T0)×(1-e^(-b×(H-Hc))), where a is the heat transfer coefficient (0.05% / ℃·h), b is the moisture content influencing factor (0.03 / %), T0 is the starting temperature (30℃), and Hc is the critical moisture content (30%). The average drying rate is calculated for each stage. The temperature change rate dT / dt is used to form a stage characteristic matrix. The sludge organic matter content is characterized by the ratio of volatile solids (VS) to total solids (TS), and the potential odor release risk is analyzed in combination with the degradation characteristics of the organic matter. Sludge organic matter is first divided into three categories: readily degradable (proteins and sugars, accounting for 25-35% of VS), moderately degradable (cellulose and fats, accounting for 40-50% of VS), and refractory (lignin and humus, accounting for 20-30% of VS). The proportion of each category is determined based on the VS / TS value: when VS / TS>0.7, the proportion of readily degradable increases by 5%; when VS / TS<0.5, the proportion of refractory increases by 5%. A correspondence between organic matter and odor components is established: readily degradable species primarily produce hydrogen sulfide (H2S), ammonia (NH3), and amines; moderately degradable species primarily produce organic acids and aldehydes; and refractory species primarily produce methyl mercaptan (CH3SH) and olefins. The formula for predicting odor component percentages is as follows: Hydrogen sulfide = 15 + 20 × (VS / TS - 0.6), Ammonia = 10 + 15 × (VS / TS - 0.6), Methyl mercaptan = 5 + 8 × (VS / TS - 0.6), Organic acids = 25 - 10 × (VS / TS - 0.6), Aldehydes = 20 - 5 × (VS / TS - 0.6), and Others = 25 (in %). The resulting odor component data includes predicted concentrations of each component, odor thresholds, and hazardous concentration limits. The release temperature sensitivity coefficients (Si) for each odor component are calculated: Hydrogen sulfide (S) = 2.5, ammonia (S) = 1.8, methyl mercaptan (S) = 3.2, organic acids (S) = 1.2, and aldehydes (S) = 1.5. The relationship between odor release rate and temperature follows the modified Arrhenius equation: r = r0 × exp[E / (R × T0) - E / (R × T)], where r is the odor release rate (mg / kg·h), r0 is the reference release rate (taken as 20 mg / kg·h), E is the apparent activation energy (taken as 25 kJ / mol), R is the gas constant (8.314 J / mol·K), T0 is the reference temperature (303K), and T is the actual temperature (K). The drying rate influence coefficient K = 1 + 0.5 × v / v0, where v is the actual drying rate (% / h), and v0 is the reference drying rate (taken as 1% / h). The odor release potential calculation formula of each component is: Pi = Ci × Si × ri × K, where Pi is the release potential of component i (mg / m 3·h), Ci is the content percentage of component i, Si is the temperature sensitivity coefficient of component i, and ri is the release rate of component i. The total odor release potential Ptotal = ∑Pi×wi, where wi is the olfactory weight factor. According to different drying stages, the temperature sensitivity coefficient is adjusted: multiply by 1.5 in the preheating stage, multiply by 1.0 in the constant speed stage, and multiply by 0.7 in the speed reduction stage. Finally, the odor release potential evaluation data table is generated. The fresh air mixing ratio R_fresh air and airflow switching cycle data are extracted from the partition coordinated control parameters to calculate the effective air exchange rate λ = Q / (V×60), where Q is the ventilation volume (m 3 / h), V is the volume of the drying chamber (m 3 ), the unit is times / minute. The calculation formula for the actual odor emission rate E is: E = Ptotal × (1-e^(-λ×t)), where Ptotal is the total odor release potential and t is the cumulative time (min). The system retention coefficient Sr = 1-λ / (λ+k), where k is the natural attenuation coefficient of odor (taken as 0.01 / min). The calculation formula for the odor concentration retention Cr is: Cr = ∫(Ptotal × Sr)dt, the integration interval is [t1, t2], t1 and t2 are the start and end time of the drying stage respectively. The discrete calculation adopts the trapezoidal integration method with a step size of 10 minutes. The prediction formula for the odor concentration C(t) in the drying chamber is: C(t) = C0×e^(-λ×t)+Ptotal / λ×(1-e^(-λ×t)), where C0 is the initial concentration (mg / m 3 The odor concentrations at each drying stage are connected using piecewise cubic spline interpolation to ensure smooth and continuous curves. A predicted odor concentration change curve is generated, which includes the predicted odor concentration values ​​at 60 time points within 24 hours. Three levels of warning thresholds are set: Level 1 (low risk) threshold C1 = 10 mg / m 3 , Level 2 (medium risk) threshold C2 = 15 mg / m 3 , Level 3 (high risk) threshold C3 = 25 mg / m 3. First, calculate the risk duration, that is, the cumulative time T1, T2, and T3 when the predicted concentration exceeds the threshold value at each level. The risk intensity index I is calculated by the following formula: I = ∑wi × (Ci(t) - Cthreshold) × Δt, if and only if Ci(t) > Cthreshold, where wi is the risk weight coefficient (level 1 is 1, level 2 is 2, and level 3 is 4), Ci(t) is the odor concentration at time t, and Δt is the time step (10 minutes). The risk index of exceeding the standard R = I / Imax, where Imax is the historical maximum risk intensity index (initial setting is 1000). The warning level is determined according to the risk index: R < 0.3 is low risk, 0.3 ≤ R < 0.6 is medium risk, and R ≥ 0.6 is high risk. The risk time distribution characteristics are quantified by the kurtosis coefficient K and the skewness coefficient S, which are used to judge the risk concentration. The adaptive adjustment coefficient of the odor exceeding standard risk α = 1 + 0.2 × ln (1 + T2 / 60) is used to dynamically adjust the control parameters. Finally, the odor exceeding standard risk data is generated, including risk index, warning level, predicted exceeding standard time point, risk concentration and duration of each level of risk.

[0092] Preferably, step S4 includes the following steps:

[0093] Step S41: extracting warning time nodes based on odor exceeding risk data;

[0094] Step S42: Based on the warning time node and the odor exceeding risk data, the zone coordinated control parameters are optimized for temporary cooling, local ventilation, and pulse odor capture control instructions to obtain an odor linkage intervention trigger signal;

[0095] Step S43: identifying potential pseudo-dry shell areas based on the moisture content change trend data, marking risky drying locations, and generating pseudo-dry shell drying location data;

[0096] Step S44: performing intelligent sludge turning and throwing operation instruction processing on the pseudo-shell drying position data to generate a turning and throwing mechanism execution instruction;

[0097] Step S45: Based on the odor linkage intervention trigger signal and the turning mechanism execution instruction, the sludge drying intelligent control is realized.

[0098] In the embodiment of the present invention, extreme point detection is performed on the predicted odor concentration change curve in the odor exceeding risk data, and the sliding window method (window width is 30 minutes) is used to determine the local maximum point. When the concentration value in the window reaches the maximum and exceeds the first-level warning threshold (10mg / m 3 ), it is marked as a potential warning point. For each potential warning point, the risk intensity index I=C×T is calculated, where C is the predicted peak concentration (mg / m 3), where T is the predicted duration (minutes). The warning points are screened according to the risk intensity index: I > 500 is a high-risk warning point, 300 < I ≤ 500 is a medium-risk warning point, and 150 < I ≤ 300 is a low-risk warning point. The warning time node is set as the advance amount before the predicted concentration exceeds the threshold, and the calculation formula is: t warning = t peak - max(60, C × 4) minutes, where t peak is the time when the predicted concentration peak appears, and C is the predicted peak concentration value. The warning time node matrix includes four key parameters: warning time, warning level, risk intensity index, and predicted duration. When optimizing the control instructions for the partition collaborative control parameters through the odor over-standard risk data based on the warning time node, a three-dimensional linkage intervention strategy is implemented. For the temporary cooling intervention, the cooling amplitude ΔT is calculated according to the risk level: the high-risk warning point cools down by 15°C, the medium-risk warning point cools down by 10°C, and the low-risk warning point cools down by 5°C. The cooling duration calculation formula is: t cooling = 80 + 5 × C, where C is the predicted peak concentration. The cooling area is determined as the partition with the largest contribution to the odor concentration, and is identified through the odor release potential assessment data. The local increased ventilation volume intervention strategy determines the increment ratio according to the risk level: the high-risk area increases the air volume by 70%, the medium-risk area increases the air volume by 50%, and the low-risk area increases the air volume by 30%. The advance amount of the ventilation volume adjustment time t advance = 45 minutes, and the duration t increased air = t warning + t cooling. The pulsed odor capture control uses intermittent activated carbon filtration technology, with the pulse frequency in the high-risk area being 3 minutes / time, in the medium-risk area being 5 minutes / time, and in the low-risk area being 8 minutes / time. The capture duration t capture = t warning + t cooling + 30 minutes. The three intervention measures are integrated to generate an odor linkage intervention trigger signal, which includes the intervention timestamp, intervention type code, intervention area ID, intervention parameter value, and intervention duration. Calculate the moisture content difference between the surface layer (0 - 5 cm) and the middle layer (5 - 15 cm) ΔH surface-middle = H middle - H surface, and the moisture content difference between the middle layer and the bottom layer (15 - 30 cm) ΔH middle-bottom = H bottom - H middle, where H surface, H middle, and H bottom are the average moisture contents of the surface layer, middle layer, and bottom layer (%) respectively. The criterion for judging the formation of a false dry shell: ΔH surface-middle > 25% and H surface < 45%. The severity classification of the false dry shell: 35% ≤ ΔH surface-middle < 45% is mild, 45% ≤ ΔH surface-middle < 60% is moderate, and ΔH surface-middle ≥ 60% is severe. Use the thermal imaging differential processing technology to calculate the surface temperature anomaly index TI, TI = (T anomaly - T average) / T average, where T anomaly is the local abnormal temperature (°C), and T average is the regional average temperature (°C). Calculate the false dry shell risk index DSI = 0.7 × (ΔH surface-middle / 100) + 0.3 × TI by integrating the moisture content gradient and the temperature anomaly index. When DSI > 0.5, mark this area as a risk drying position, assign four attributes: area ID, coordinate range, false dry shell severity, and area proportion, and generate false dry shell drying position data.Determine the turning and throwing depth according to the severity of the false dry crust: the turning and throwing depth for mild false dry crust is 10 cm, for moderate false dry crust is 15 cm, and for severe false dry crust is 25 cm. The turning and throwing frequency is determined according to the false dry crust risk index DSI: once every 8 hours when 0.5 < DSI ≤ 0.6, once every 4 hours when 0.6 < DSI ≤ 0.8, and once every 2 hours when DSI > 0.8. The turning and throwing operation path planning adopts the snake-shaped coverage algorithm. The drying chamber is divided into an 8×6 grid matrix, with each grid size of 50 cm×50 cm. Mark the grids to be turned and thrown according to the false dry crust drying position data. The action sequence of the turning and throwing mechanism includes five steps: positioning, descending, turning and throwing, ascending, and moving. Execution parameters are assigned to each step, such as a descending speed of 5 cm / s and a turning and throwing rotational speed of 30 rpm. The priority order of the turning and throwing operation is based on the false dry crust risk index DSI, and is executed in descending order. The generated turning and throwing mechanism execution instructions include six parts of parameters: the turning and throwing area coordinates, turning and throwing depth, turning and throwing speed, turning and throwing angle, execution time, and priority. Build an event-driven execution queue, arrange the odor linkage intervention trigger signal and the turning and throwing mechanism execution instructions in chronological order, and generate a 48-hour rolling execution schedule. The execution plan adopts a hierarchical architecture: the top layer is the intervention decision-making layer, which receives the odor linkage intervention trigger signal and the turning and throwing mechanism execution instructions; the middle layer is the parameter conversion layer, which converts the intervention signal into specific control parameters; the bottom layer is the equipment execution layer, which drives the heating system, ventilation system, and turning and throwing mechanism to execute operations. The execution logic of the heating system: after receiving the temporary temperature reduction intervention signal, linearly reduce the heat source power by the specified temperature reduction amplitude ΔT within the t temperature reduction time, and control the temperature reduction rate at 0.25 °C / minute. The execution logic of the ventilation system: after receiving the local increased ventilation volume signal, increase the fan speed according to the trapezoidal curve and complete the air volume adjustment within 10 minutes. The execution logic of the odor capture system: periodically start and stop the activated carbon filtration device according to the pulse frequency, and the startup time ratio is 40%. The execution logic of the turning and throwing mechanism: execute the turning and throwing operation at the specified time according to the priority order in the execution instructions, and real-time feedback the turning and throwing completion status. Each execution action is transmitted to the sludge drying chamber control unit in real time through the control bus to form a closed-loop control system.

[0099] Particularly important is that the intelligent sludge turning and throwing operation instruction processing for the false dry crust drying position data includes: <o

[0100] Extract the maximum difference between the average moisture content of the sludge surface layer and the average moisture content of the middle / lower layer in the sludge state monitoring characteristic data corresponding to the false dry crust drying position data, and obtain the sludge layer moisture content gradient value;

[0101] Based on the sludge layer moisture content gradient value, judge whether it exceeds the preset dry crust threshold of 15%, then strengthen the turning and throwing requirement, and obtain the turning and throwing action trigger determination;

[0102] When the flipping action triggering judgment result is yes, the flipping operation depth of the flipping mechanism is determined according to the measured sludge layer thickness;

[0103] The frequency of the turning operation of the turning mechanism is set based on the turning operation depth and the sludge moisture content partition data;

[0104] Based on the depth and frequency of the turning operation, control instructions are generated to drive the turning mechanism to perform operations, and the turning mechanism execution instructions are obtained with the goal of achieving a sludge uniformity index of above 0.85 after turning.

[0105] In an embodiment of the present invention, a spatial index mapping table is established to map the regional coordinate range (x1, y1, x2, y2) in the pseudo-shell drying location data to the microwave moisture content sensor array. Real-time moisture content data for five measuring points in the corresponding area is extracted from the surface sensor (0-5 cm), and the arithmetic mean Hs is calculated. Data for five measuring points in each corresponding area is extracted from the middle sensor (5-15 cm) and the bottom sensor (15-30 cm), and the arithmetic mean Hm and Hb are calculated, respectively. The surface-middle layer difference ΔHs-m = Hm-Hs and the surface-bottom layer difference ΔHs-b = Hb-Hs are calculated. The sludge layer moisture content gradient value ΔHmax takes the maximum value of the two: ΔHmax = max(ΔHs-m, ΔHs-b). When multiple pseudo-shell regions exist, the moisture content gradient value is independently calculated for each region, generating a data matrix containing the region ID, region area, and corresponding moisture content gradient value. The system's default baseline dry crust threshold is 15%. This value, determined based on statistical analysis of laboratory drying tests, represents the initial sign of dry crust formation when the moisture content difference between the surface and deep layers reaches 15%. During actual judgment, a modified threshold calculation formula is used: THcorr = TH0 × (1 - 0.02 × (T - 60)), where THcorr is the modified dry crust threshold, TH0 is the baseline dry crust threshold of 15%, and T is the average drying chamber temperature (°C). When the sludge drying temperature exceeds 60°C, the risk of dry crust formation increases, and the threshold is lowered accordingly; conversely, the threshold is raised. The sludge layer moisture content gradient ΔHmax is compared with the modified dry crust threshold THcorr. If ΔHmax > THcorr, the decision to trigger dumping is "yes"; otherwise, it is "no". When ΔHmax > 30%, forced dumping is initiated, without considering temperature correction. The decision result is stored as a Boolean value (0 or 1) in the dumping decision matrix. The sludge layer thickness is measured using an ultrasonic level meter and an infrared rangefinder. The calculation formula is: H = (L0 - L1), where H is the sludge layer thickness (cm), L0 is the fixed distance from the drying chamber bottom to the sensor (cm), and L1 is the measured distance from the sludge surface to the sensor (cm). The sludge dumping depth D is related to the sludge layer thickness H and the moisture gradient ΔHmax. The calculation formula is: D = min(H × 0.8, H × 0.4 + 0.3 × ΔHmax). Ensure that the dumping depth does not exceed 80% of the total sludge layer thickness. The greater the gradient, the deeper the dumping. The dumping depth is limited to 5 cm; if the calculated value is less than 5 cm, it is taken as 5 cm. Sludge dumping depths are categorized based on its physical properties: light dumping (5-10 cm), moderate dumping (10-20 cm), and deep dumping (20-30 cm). The flipping mechanism adopts a spiral flipping device driven by a servo motor with an accuracy of ±0.5cm. The descending displacement of the servo motor is set according to the calculated flipping depth D.The benchmark turning cycle T0 is determined based on the turning depth D: when D < 10 cm, T0 = 360 minutes; when 10 cm ≤ D < 20 cm, T0 = 240 minutes; when D ≥ 20 cm, T0 = 120 minutes. Adjustment coefficients are applied based on the moisture content of the pseudo-dry shell area: k1 = 0.6 for high moisture content (>80%); k2 = 1.0 for medium moisture content (50%-80%); and k3 = 1.4 for low moisture content (<50%). The gradient adjustment coefficient (kg) is calculated based on the moisture content gradient ΔHmax: kg = 1 - 0.02 × (ΔHmax - 15) for ΔHmax > 15%; kg = 1 for ΔHmax ≤ 15%. The actual turning cycle T = T0 × kn × kg, where kn is the moisture content adjustment coefficient for the area. Turning frequency f = 1 / T, expressed in times / hour. The flipping mechanism control system sets the flipping time interval according to the calculated frequency f and establishes a flipping timer with an accuracy of minutes. A flipping execution instruction package is constructed, which contains 6 core parameters: flipping area coordinates (x1, y1, x2, y2), flipping depth D (cm), flipping speed V (rpm), flipping angle θ (°), flipping duration t (seconds) and flipping frequency f (times / hour). The flipping speed V is determined according to the depth: V = 20-0.5×D, ensuring that the greater the depth, the lower the speed, reducing mechanical stress. The flipping angle θ represents the installation angle of the spiral flipper and is set to 45°. The flipping duration t is calculated as follows: t = 30 + 2×S, where S is the area of ​​the flipping area (m). 2 The tossing uniformity index is calculated using the ratio of the standard deviation of moisture content: U = 1 - σ after tossing / σ before tossing, where σ is the standard deviation of regional moisture content. When U < 0.85, a supplementary tossing cycle is automatically triggered, increasing the tossing duration by 15 seconds per cycle until U ≥ 0.85 or the number of supplementary cycles reaches three. The execution command is transmitted to the tossing mechanism execution unit via the industrial fieldbus. The command format is: {region ID, D, V, θ, t, f, execution timestamp}.

[0106] Preferably, the present invention further provides an Internet of Things-based intelligent sludge drying control system, which implements the above-mentioned Internet of Things-based intelligent sludge drying control method. The Internet of Things-based intelligent sludge drying control system includes the following modules:

[0107] The sensor measurement module is equipped with a microwave moisture content sensor, an infrared temperature sensor, an electrochemical odor concentration sensor, a hot film air flow meter, and a differential pressure sensor, which is used to collect real-time monitoring and feedback of key parameters such as sludge moisture content, drying temperature, odor concentration, and ventilation volume;

[0108] The intelligent control module, with a PLC-based control core, is responsible for data processing and command issuance, and controls operating parameters based on monitoring data. The operating parameter control includes frequency converters, heaters, and feeding devices, which control the temperature, ventilation volume, and sludge feeding speed parameters during the sludge drying process.

[0109] The process unit control module is equipped with a drying temperature control unit, a ventilation volume control unit, and a sludge feeding speed control unit, including heat source supply volume adjustment, fan air volume adjustment, and feeding speed adjustment;

[0110] The data management and visualization module includes data storage and visualization interface, uses cloud platforms or local servers to store historical data, and uses mobile terminals to display key parameters such as sludge concentration, drying temperature, and ventilation volume in real time, while supporting remote monitoring.

[0111] The present invention realizes real-time monitoring and feedback of key parameters such as moisture content, temperature, odor concentration and ventilation volume in the sludge drying chamber by constructing an Internet of Things monitoring network. Multi-source data preprocessing technology is used to generate accurate sludge status monitoring characteristic data, and based on this, grid point moisture content values ​​are classified to divide the sludge moisture content into zones. Further combined with the sludge organic matter content and density, the surface evaporation rate and moisture content change trend are analyzed to realize zone coordinated regulation and treatment, and generate optimized zone coordinated regulation parameters. At the same time, through the risk analysis of odor exceeding the standard, the odor linkage intervention mechanism is triggered, the control instructions are optimized, and the intelligent control of the sludge drying process is realized. This method not only improves the sludge drying efficiency and reduces energy consumption, but also effectively avoids odor emissions exceeding the standard, providing an efficient and environmentally friendly solution for drying treatment in the fields of municipal sludge, industrial sludge, etc.

[0112] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0113] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent control method for sludge drying based on the Internet of Things, characterized in that: The following steps are involved: Step S1: deploying an IoT monitoring network for the sludge drying chamber to build an IoT monitoring network; The IoT monitoring network is used to periodically collect monitoring data from the sludge drying chamber, and multi-source data is pre-processed to generate sludge status monitoring feature data. Step S2: classifying the water content values ​​of the grid points according to the sludge state monitoring characteristic data to generate sludge water content partition data; The drying correction factor is processed according to the sludge moisture content partition data to obtain the partition drying correction factor; Step S3: Obtaining sludge organic matter content and sludge density; Analyze the evaporation rate of the sludge surface based on the organic matter content and density of the sludge, deduce the change of the moisture content of the sludge layer, and generate the moisture content change trend data; Use the partition drying correction factor and moisture content change trend data to perform partition coordinated control processing and generate partition coordinated control parameters; Conduct odor exceeding standard risk analysis based on moisture content change trend data and generate odor exceeding standard risk data; Step S4: Optimize the control instructions of the zone collaborative control parameters through the odor exceeding risk data to obtain the odor linkage intervention trigger signal; and realize the intelligent control of sludge drying according to the odor linkage intervention trigger signal.

2. The intelligent control method for sludge drying based on the Internet of Things according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: Positioning key nodes of the sludge drying process in the sludge drying chamber and generating IoT monitoring node data; Step S12: deploying an IoT monitoring network for the sludge drying chamber based on IoT monitoring node data to build an IoT monitoring network; Step S13: setting a basic sampling period of 5 minutes, and using the Internet of Things monitoring network to periodically collect monitoring data from the sludge drying chamber based on the basic sampling period to generate multi-source original monitoring data; Step S14: determining the current drying stage based on the multi-source original monitoring data, and adjusting the differential sampling frequency of the basic sampling period to generate sludge drying differential sampling data; Step S15: Perform multi-source data preprocessing on the sludge drying differential sampling data to generate sludge status monitoring characteristic data; wherein, the sludge status monitoring characteristic data includes layered monitoring moisture content value, drying chamber monitoring temperature field, odor concentration monitoring data, monitoring air inlet velocity value and monitoring exhaust pressure difference value.

3. The intelligent control method for sludge drying based on the Internet of Things according to claim 2 is characterized in that: The deployment of IoT monitoring network for sludge drying chamber based on IoT monitoring node data includes: Three layers of microwave moisture content sensors are set up in the sludge drying chamber, located at the sludge surface layer, sludge middle layer and sludge bottom layer respectively. Five measuring points are set up in each layer. Among them, the measuring points of the sludge surface layer are 0-5cm away from the surface, the measuring points of the sludge middle layer are 5-15cm away from the surface, and the measuring points of the sludge bottom layer are 15-30cm away from the surface. 16 infrared temperature sensors are evenly arranged in the sludge drying chamber to form a 4×4 grid array of temperature measurement points; Three electrochemical odor concentration sensors are installed at the top, middle and bottom of the sludge drying chamber; Install hot film air flow meter and differential pressure sensor at the inlet and outlet of ventilation duct respectively; The microwave moisture content sensor, infrared temperature sensor, electrochemical odor concentration sensor, hot film air flow meter and differential pressure sensor are connected to the IoT edge computing controller located near the sludge drying chamber through wireless communication protocols to build an IoT monitoring network.

4. The intelligent control method for sludge drying based on the Internet of Things according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: extracting the moisture content monitoring values ​​of the surface layer, middle layer and bottom layer based on the layered monitoring moisture content values ​​in the sludge state monitoring characteristic data; Step S22: performing moisture content distribution grid processing according to the moisture content monitoring value to generate sludge vertical moisture content distribution data; Step S23: Classifying the water content values ​​of the grid points based on the sludge vertical water content distribution data and performing connectivity processing on each area to generate sludge water content zoning data; wherein, sludge beds with a water content greater than 80% are marked as high water content areas, sludge beds with a water content between 50% and 80% are marked as medium water content areas, and sludge beds with a water content less than 50% are marked as low water content areas; Step S24: Evaluate the sludge drying index based on the sludge vertical moisture distribution data and the sludge moisture content partition data to generate a real-time sludge drying index; Step S25: performing drying correction factor processing on the sludge moisture content partition data using the real-time sludge drying index to obtain the partition drying correction factor.

5. The intelligent control method for sludge drying based on the Internet of Things according to claim 4 is characterized in that: The sludge drying index evaluation based on the sludge vertical moisture distribution data and the sludge moisture content zone data includes: Calculate the average sludge moisture content based on the sludge vertical moisture distribution data as the overall moisture content assessment value; According to the vertical moisture distribution data of sludge, the moisture content difference of different depths of sludge layer is analyzed to obtain the vertical moisture gradient data; Identify the uneven distribution of sludge moisture content based on the sludge moisture content partition data and generate the sludge moisture content unevenness; A real-time sludge drying index is evaluated based on the overall moisture content assessment value, vertical moisture content gradient data, and sludge moisture content unevenness to generate a real-time sludge drying index.

6. The intelligent control method for sludge drying based on the Internet of Things according to claim 1 is characterized in that: In step S3, the evaporation rate of the sludge surface is analyzed based on the organic matter content and the sludge density, and the change in the moisture content of the sludge layer is deduced, including: Extracting the sludge surface moisture content according to the layered monitoring moisture content value in the sludge state monitoring characteristic data; The surface moisture evaporation rate of the sludge surface moisture content is calculated using the drying chamber monitoring temperature field, the monitoring air inlet velocity value and the monitoring exhaust pressure difference in the sludge state monitoring characteristic data to generate the sludge surface evaporation rate; Matching heat and mass transfer balance parameters based on sludge organic matter content and sludge density; Calculate the sludge receiving heat energy based on the drying chamber monitoring temperature field in the sludge state monitoring characteristic data to generate the sludge receiving heat energy value; Based on the layered monitoring moisture content values ​​in the sludge status monitoring characteristic data, the moisture content difference between the inner and outer layers is analyzed, and the internal moisture migration rate is calculated based on the heat and mass transfer balance parameters and the sludge receiving heat energy value; The heat required for water evaporation is estimated by the evaporation rate of the sludge surface and the internal water migration rate based on the heat energy value received by the sludge, and the change in moisture content of the sludge layer is deduced to generate moisture content change trend data.

7. The intelligent control method for sludge drying based on the Internet of Things according to claim 1 is characterized in that: In step S3, the partition-by-partition coordinated control process using the partition-by-partition drying correction factor and the moisture content change trend data includes: Perform drying uniformity target processing according to the partition drying correction factor to generate drying uniformity target data; Evaluate the drying completion time based on moisture content trend data; The drying temperature is set based on the sludge moisture content partition data according to the drying completion time and drying uniformity target data to generate sludge drying temperature data; Based on the drying uniformity target data and sludge drying temperature data, preliminary zone drying parameter configuration is performed to generate differentiated sludge drying parameters; The differentiated sludge drying parameters are subjected to zone-by-zone collaborative regulation and control to generate zone-by-zone collaborative regulation parameters.

8. The intelligent control method for sludge drying based on the Internet of Things according to claim 7 is characterized in that: The risk analysis of odor exceeding the standard based on the moisture content change trend data in step S3 includes: Process the drying stage path based on the drying completion time and moisture content change trend data to generate predicted drying stage data; Analyze the drying temperature change rate and sludge drying speed based on the predicted drying stage data; Analyze odor components based on the organic matter content of sludge and generate odor component data; Evaluate the odor release potential based on the drying temperature change rate and sludge drying speed using odor component data to generate odor release potential evaluation data; Odor emission analysis is performed on the odor release potential assessment data through zoning coordinated control parameters, and odor concentration retention is calculated based on the predicted drying stage data to obtain the predicted odor concentration change curve; Adaptive dry odor warning processing is performed based on the predicted odor concentration change curve to generate odor exceeding standard risk data.

9. The intelligent control method for sludge drying based on the Internet of Things according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: extracting warning time nodes based on odor exceeding risk data; Step S42: Based on the warning time node and the odor exceeding risk data, the zone coordinated control parameters are optimized for temporary cooling, local ventilation, and pulse odor capture control instructions to obtain an odor linkage intervention trigger signal; Step S43: identifying potential pseudo-dry shell areas based on the moisture content change trend data, marking risky drying locations, and generating pseudo-dry shell drying location data; Step S44: performing intelligent sludge turning and throwing operation instruction processing on the pseudo-shell drying position data to generate a turning and throwing mechanism execution instruction; Step S45: Based on the odor linkage intervention trigger signal and the turning mechanism execution instruction, the sludge drying intelligent control is realized.

10. An intelligent control system for sludge drying based on the Internet of Things, characterized in that: The method for intelligently controlling sludge drying based on the Internet of Things according to claim 1 is used to execute the method, and the intelligent control system for sludge drying based on the Internet of Things includes the following modules: The sensor measurement module is equipped with a microwave moisture content sensor, an infrared temperature sensor, an electrochemical odor concentration sensor, a hot film air flow meter, and a differential pressure sensor, which is used to collect real-time monitoring and feedback of key parameters such as sludge moisture content, drying temperature, odor concentration, and ventilation volume; The intelligent control module, with a PLC-based control core, is responsible for data processing and command issuance, and controls operating parameters based on monitoring data. The operating parameter control includes frequency converters, heaters, and feeding devices, which control the temperature, ventilation volume, and sludge feeding speed parameters during the sludge drying process. The process unit control module is equipped with a drying temperature control unit, a ventilation volume control unit, and a sludge feeding speed control unit, including heat source supply volume adjustment, fan air volume adjustment, and feeding speed adjustment; The data management and visualization module includes data storage and visualization interface, uses cloud platforms or local servers to store historical data, and uses mobile terminals to display key parameters such as sludge concentration, drying temperature, and ventilation volume in real time, while supporting remote monitoring.

Images