A compost constant oxygen aeration control method based on feedforward-feedback compound control

Abstract

This invention relates to the field of aerobic composting of organic solid waste, and particularly to a constant-oxygen aeration control method for composting based on a feedforward-feedback composite control. It includes oxygen concentration feedback control and delayed-compensation feedforward control. The purpose of delayed-compensation feedforward control is to provide a brief, homogenizing purge during the shutdown period of the feedback-controlled aeration system. Utilizing the forced convection effect generated by the purge, it promotes gas circulation within the device, rapidly mixing the gas inside the compost pile with the gas in the sensor area, shortening the measurement lag time, and enabling the sensor readings to more accurately reflect the true condition inside the compost pile. Simultaneously, its brief operation avoids impacting the normal aeration system and causing overshoot in the oxygen concentration control system. This invention employs an automatic control strategy of delayed-compensation feedforward-oxygen concentration feedback composite control, achieving high-precision control of oxygen concentration during the composting process.

Description

Technical Field

[0001] This invention relates to the field of aerobic composting of organic solid waste, and in particular to a constant oxygen aeration control method for composting based on feedforward-feedback composite control. Background Technology

[0002] In scientific research on aerobic composting, precise control of oxygen concentration is crucial for revealing the metabolic pathways of aerobic microorganisms and optimizing process parameters. Automatic aeration based on oxygen concentration feedback is a common technical approach to achieve this control. Under this approach, precise control requires the coordination of two aspects: first, data accuracy, meaning that the data obtained from sensor monitoring can accurately reflect the actual situation; and second, data comprehensiveness, with key factors including the integrity of the monitoring cycle, sampling frequency, and how to avoid interference caused by frequent sampling, thus balancing data accuracy and comprehensiveness. Through the synergistic combination of these two aspects and real-time feedback control, the oxygen concentration within the composting system can ultimately be stabilized within the target range, thereby achieving precise control.

[0003] However, due to the harsh environment inside the reactor, such as high temperature, high humidity, and the presence of corrosive gases like H2S, as well as the high-temperature operating conditions (such as those of zirconia sensors) or operating requirements (such as those of electrochemical sensors) of oxygen concentration sensors, they cannot be directly placed inside the reactor for contact-type in-situ online monitoring. This is because high temperatures would cause the microorganisms around the sensor to become inactive, while high humidity and corrosive gases would greatly shorten the sensor's lifespan.

[0004] Based on this practical problem, existing technical solutions can be divided into two directions: one is to sample and filter the gas from inside the composting unit to monitor its oxygen concentration. This method can accurately reflect the true oxygen concentration inside the composting unit; however, due to monitoring gaps or frequent sampling affecting the actual composting process, it cannot achieve comprehensive data. The second method is to place the sensor directly inside the composting device. To ensure the sensor's lifespan, it is usually placed in a gas phase space that is not in direct contact with the composting unit, such as near the gas outlet or inside the exhaust gas detection device. This method can achieve full-process monitoring without interfering with composting, but its data accuracy is questionable, directly reflected in the large fluctuations in actual oxygen concentration, exceeding the expected target range. Therefore, how to balance the accuracy and comprehensiveness of oxygen concentration data and achieve precise control throughout the entire process is a long-standing technical challenge in this field. Summary of the Invention

[0005] A discussion of existing technologies reveals that in-plant gas extraction sampling can ensure the accuracy of sampling data while also considering sensor lifespan, but it is difficult to guarantee the comprehensiveness of the data, often resulting in gaps in monitoring and control. Conventional sensor installation methods, such as placing them inside a closed composting device, near the gas outlet, or as a separate exhaust gas detection device, can ensure the comprehensiveness of sampling data. However, in order to balance sensor lifespan and avoid impacting the compost, the sensor is usually placed in a gas phase space that is not in direct contact with the compost, assuming that the oxygen concentration in this space represents the actual oxygen concentration in the compost.

[0006] However, this assumption relies on two premises: First, the composting scale is based on a small-scale laboratory-scale composting device, intended for scientific research purposes, facilitating the control of variables and the reproduction of results. Therefore, the compost pile is relatively small, and the spatial differences in oxygen concentration at different locations are negligible. Thus, oxygen concentration control in large-scale factory composting is not within the scope of this invention. Second, the gas inside the device can be considered uniformly mixed. For non-direct contact measurements installed inside the composting device or in the exhaust gas measuring device, ignoring this premise can easily lead to measurement lag, compromising data accuracy and ultimately causing oxygen concentration fluctuations over a large range, making precise control impossible.

[0007] Explanation of measurement lag:

[0008] (1) When aeration begins, fresh air enters through the air inlet and passes through the pile. Microorganisms consume oxygen. At this time, due to the forced convection effect generated by positive or negative pressure, the gas phase space where the sensor is located is uniformly mixed with the gas inside the pile, and the oxygen concentration remains basically consistent. Therefore, the sensor data can accurately reflect the changes in oxygen concentration inside the pile.

[0009] (2) After aeration stops, microorganisms in the pile still consume oxygen. However, due to the lack of forced convection, the oxygen consumption in the gas phase space where the sensor is located is a relatively slow physical process that depends on the diffusion of the concentration gradient (higher concentration of oxygen in the gas phase space diffuses into the pile interior where the concentration is lower). This process cannot be synchronized with the oxygen consumption rate inside the pile, resulting in a diffusion delay. Therefore, the pile interior may already be oxygen-deficient, but the sensor measurement is delayed and the aeration system cannot be started. Only when the oxygen concentration in the gas phase space where the sensor is located is lower than the threshold does the aeration system start working, but the pile interior has been severely oxygen-deficient for a long time. This problem will be reflected in the large fluctuations in oxygen concentration: after aeration starts, the oxygen concentration will first drop sharply to an oxygen-deficient state, and then rise back to the predetermined threshold.

[0010] Therefore, this hysteresis effect causes the control system to oscillate repeatedly between overshoot and undershoot, resulting in large fluctuations in the actual oxygen concentration inside the compost pile. It cannot be stabilized within the predetermined oxygen concentration range, and there will still be periods of hypoxia and anaerobicity, which seriously affects composting efficiency, product quality and equipment energy consumption, and limits the accuracy of scientific research.

[0011] The specific technical solution provided by this invention is as follows:

[0012] A constant-oxygen aeration control method for composting based on feedforward-feedback composite control comprises two parts:

[0013] (a) Oxygen concentration feedback control

[0014] The oxygen measuring device can monitor the oxygen concentration in the composting device online and input the data into the control system. Depending on the research purpose, a target oxygen concentration range will be set in the control system. When the input real-time data is within the range, the aeration system will not operate and will maintain its original state.

[0015] When the input real-time data is lower than the lower limit of the range, the control system will turn on the aeration system, and then the oxygen concentration will gradually increase until the input real-time data exceeds the upper limit of the range, at which point the aeration system will be turned off. During this period, the aeration system will always remain on.

[0016] When the input real-time data exceeds the upper limit of the interval and the aeration system is shut down, the oxygen concentration in the pile will be gradually consumed and decrease under the action of microorganisms until the input real-time data is lower than the lower limit of the interval and the aeration system will be restarted. During this period, the measurement will be delayed due to diffusion delay. Therefore, feedforward control with delay compensation is introduced to intervene in this problem.

[0017] (ii) Delay-compensated feedforward control

[0018] The primary purpose of this feedforward control is not to provide oxygen to the reactor core, but rather to provide a brief period of homogenization purging during the shutdown of the feedback control aeration system. Utilizing the forced convection effect generated by this purging, the gas within the device circulates, rapidly mixing with the gas in the sensor area. This significantly reduces the measurement lag time, allowing sensor readings to more accurately reflect the true internal conditions of the reactor core. Simultaneously, its short-duration operation minimizes its impact on the normal aeration system, preventing overshooting of the oxygen concentration control system.

[0019] Feedforward control can be a separate loop independent of feedback control, or it can be a coupled control loop that is linked to feedback control.

[0020] (1) In practical operation, it is simplified into a separate loop control independent of feedback control. The logic is simple, the hardware and design requirements are low, and the control effect is ideal. The main parameters include the dormancy period T and the working period t. T is the running time of feedback control. During this period, the feedforward control is in a dormant state. After the dormancy ends, it enters the working period and runs briefly for t. During this period, the aeration device starts to work. If it is in the "start aeration - oxygen concentration rise" stage of feedback control, the feedforward control will not affect the original state. If it is in the "stop aeration - oxygen concentration decrease" stage of feedback control, the aeration system will be turned on briefly to accelerate gas diffusion, make the gas in the composting device mix evenly, reduce the measurement lag time, and make the data measured by the sensor reflect the real situation in the compost pile in a timely manner.

[0021] (2) Coupled feedforward control with feedback control can further optimize control performance, improve control accuracy, and reduce energy consumption. However, it requires the addition of extra data processing and identification units, placing higher demands on the control system's storage, computation, logic control, and sensor accuracy. The specific method is as follows:

[0022] Method 1: Add the function of judging real-time data and historical trends in the feedback control to the feedforward control to further reduce unnecessary aeration: if the data is on an upward trend in the recent period (e.g., within 5 minutes), the feedforward control loop always remains in a dormant state; if the data is on a downward trend or basically unchanged in the recent period (e.g., within 5 minutes), the feedforward control loop maintains a normal working state, that is, different T and t parameters are set according to the composting stage.

[0023] Method 2: Predict key parameters using machine learning models.

[0024] a. First, the oxygen consumption rate is calculated based on the oxygen concentration and time to determine whether the oxygen concentration is increasing or decreasing. If the oxygen concentration is increasing, T is set to a relatively large value (e.g., 10000s, maintaining a dormant state). If the oxygen concentration is decreasing, a baseline value is initially set for both T and t, and this baseline value is adjusted according to the prediction rules. T is an initial reference value determined based on previous experimental data and empirical summarization, serving as a dynamically adjustable baseline anchor. Subsequent "prediction rules" will dynamically adjust this baseline value based on real-time monitored trends in oxygen concentration and other specific system state signals.

[0025] b. Prediction rule: When using automatic aeration control, by recording time and oxygen concentration, the oxygen concentration as a function of time, φ(O2), can be obtained. =C(t) represents the net change in oxygen concentration within the system, that is, the percentage of the total gas volume relative to the difference between the total amount of oxygen supplied to the system and the total amount of oxygen consumed by microorganisms from time 0 to time t. When the total amount of aeration supplied is greater than the total amount of oxygen consumed by microorganisms, the oxygen concentration increases, and vice versa.

[0026] Therefore, oxygen concentration will repeatedly fluctuate between rising and falling phases throughout the entire control process. If we define "oxygen concentration decrease – oxygen concentration increase" as a cycle, that is, the complete fluctuation process of oxygen concentration from one local maximum (or minimum) to the next local maximum (or minimum), and define the rate of change of oxygen concentration as φ'(O2), i.e.

[0027]

[0028] φ'(O2) will go through a phase of gradually increasing from a negative value to 0, gradually increasing from 0 to a positive value, and then gradually decreasing back to 0, where:

[0029] ① The value gradually increases from negative to 0 because during the process of oxygen concentration decrease, it falls below the lower limit of the target range. At this time, the aeration system starts to work. However, since the total aeration supply is less than the total consumption by microorganisms, the oxygen concentration still decreases and φ'(O2) is negative. However, the start of the aeration system will have a "braking effect" on the decrease in oxygen concentration. Therefore, the rate of decrease of φ(O2) gradually decreases, that is, the absolute value of φ'(O2) gradually decreases and φ'(O2) gradually increases.

[0030] ②When time t1 is reached,

[0031] This indicates that the total aeration supply equals the total microbial consumption at this point, with time t1 being a local minimum. Subsequently, the oxygen concentration transitions from a decreasing phase to an increasing phase.

[0032] ③The reason why φ'(O2) gradually increases from 0 to a positive value is that during this period, the total aeration supply is greater than the total consumption by microorganisms, the oxygen concentration increases, and φ'(O2) becomes a positive value.

[0033] ④ When φ(O2) gradually increases and exceeds the upper limit of the target range, the aeration system stops working. However, due to the accumulation of air transported by aeration, the total aeration supply is greater than the total consumption by microorganisms. Therefore, the oxygen concentration still increases, and φ'(O2) is positive. The cessation of the aeration system, the continuous consumption by microorganisms, and the existence of diffusion delay will also have a "braking effect" on the increase in oxygen concentration. Therefore, the rate of increase of φ(O2) gradually decreases, that is, the gradual decrease of φ'(O2).

[0034] ⑤ When time t2 is reached,

[0035]

[0036] This means that the total aeration supply during this cycle equals the total microbial consumption. The oxygen concentration at time t2 is a local maximum, after which it transitions from an increasing phase to a decreasing phase.

[0037] ⑥ Assume there are N cycles of change throughout the process, and each cycle has its own times t1 and t2. Let T' = t2 - t 1， Then the set S = {T'} is related to T'. i For the i-th period, there exists T' = | i = 1, 2, 3, …, N}. i = t 2i -t 1i， This value represents the time taken from the beginning to the end of a single cycle for the net increase in oxygen concentration (the net value of oxygen provided by aeration and oxygen consumed by microorganisms), reflecting the intensity of oxygen consumption by microorganisms.

[0038] When microorganisms have a strong demand for oxygen, a longer aeration time is required to achieve a net increase in oxygen concentration and reach the target threshold; if the value of T' increases, that is, T' i > T' i-1 If the difference between the measured concentration decrease due to free diffusion of oxygen and the actual concentration decrease due to oxygen consumption by microorganisms is large, the hysteresis effect will be more obvious. Therefore, it is necessary to shorten the dormancy period T and increase the frequency of homogenization purging to reduce the influence of measurement hysteresis. Conversely, if the difference between the measured concentration decrease due to free diffusion of oxygen and the actual concentration decrease due to oxygen consumption by microorganisms is large, the hysteresis effect will be more obvious. Therefore, it is necessary to shorten the dormancy period T and increase the frequency of homogenization purging to reduce the influence of measurement hysteresis.

[0039] c. To avoid overshooting due to excessively long working cycles, the range of values ​​for t is relatively limited. Several preset values ​​can be used to meet the target requirements based on the pile volume, composting device volume, aeration rate, and aeration pipeline length, such as 30s, 60s, and 90s.

[0040] Furthermore, the control system can be constructed using other automatic control or Internet of Things (IoT) hardware such as PLCs and corresponding software.

[0041] Furthermore, in the feedforward control loop, in addition to setting T and t to control the opening and closing of the aeration system, the opening degree of the damper can also be used to couple the aeration rate and oxygen concentration, further optimizing the control accuracy. For T', when this value gradually increases, the aeration rate can be increased; when this value gradually decreases, the aeration rate can be decreased.

[0042] In addition to operating in a standalone high-precision composting oxygen concentration control system, this invention can also be integrated into a multi-parameter, multi-stage intelligent composting control system. For example, a multi-parameter intelligent control system can simultaneously monitor parameters such as oxygen concentration, temperature, and moisture content, with multiple parameters jointly regulating the composting system. However, the oxygen control system within it can still use the feedforward-feedback composite control system of this invention. Alternatively, in a multi-stage control system where oxygen concentration control is the primary condition in a certain stage, the feedforward-feedback composite control system of this invention can also be used.

[0043] The technical effects of the present invention are as follows:

[0044] (1) It can provide a high-precision control scheme for closed small composting devices and non-contact in-situ oxygen concentration monitoring and control, so as to achieve the basic control of the composting process within the range of ±1% of the target oxygen concentration.

[0045] (2) Improve composting efficiency and reduce energy consumption

[0046] Precise oxygen concentration control can provide the optimal oxygen concentration range for aerobic degradation by microorganisms, avoiding excessively high or low concentrations, thereby controlling energy consumption and reducing the generation of anaerobic conditions. At the same time, the composting process can be dynamically regulated according to actual needs to meet the oxygen requirements at different stages and optimize composting efficiency.

[0047] (3) A simplified method that is easy to operate in practice is proposed. The control logic is simple, the cost is low, and the control effect is ideal.

[0048] This invention focuses on cost control and practical ease of operation. Based on the high-precision control method, it proposes a simplified method with clearer and more straightforward control logic, which is easier to understand and operate, and has a lower cost. However, the control effect is still effective, and it can control the fluctuation within ±1% of the target oxygen concentration. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating the control principle of the present invention;

[0050] Figure 2 A diagram of the composting apparatus for an embodiment;

[0051] Figure 3 The oxygen concentration changes during intermittent aeration in this example (partial).

[0052] Figure 4 The example demonstrates the feedback of oxygen concentration changes using conventional oxygen concentration feedback.

[0053] Figure 5 The oxygen concentration variation of feedforward-feedback combined control aeration under low oxygen concentration (4-6%) is shown in the example.

[0054] Figure 6 The oxygen concentration variation of aeration under feedforward-feedback composite control in the example of medium oxygen concentration (7-9%);

[0055] Figure 7 The oxygen concentration variation of feedforward-feedback composite control aeration in the high oxygen concentration (10-12%) example is shown. Detailed Implementation

[0056] The specific technical solutions of the present invention will be described with reference to the embodiments.

[0057] In this embodiment, dewatered sludge from a wastewater treatment plant in Beijing (moisture content approximately 86%, dry basis carbon content approximately 28%, nitrogen content approximately 4%) and rice husks (moisture content approximately 7%, dry basis carbon content 39%, nitrogen content 0.6%) were mixed uniformly in a specific ratio to obtain approximately 6 kg of sludge-rice husk mixture (moisture content approximately 56%, carbon-nitrogen ratio approximately 27). This mixture was added to three identical composting devices as parallel controls, with approximately 1.85 kg of material added to each device for composting. The remainder accounted for losses during transfer and the original sample. The original sample volume (length * width * height) was approximately 18 * 16 * 40 cm.

[0058] like Figure 2 As shown, the complete composting system consists of a vertical composting device, a speed-adjustable air pump, a microcontroller (ESP32 with relay function), a host computer, an oxygen concentration measuring device (zirconia sensor), and a heat preservation device (temperature-adjustable constant temperature incubator).

[0059] (1) The vertical composting device uses bottom forced ventilation for aeration. There is an air inlet pipe at the bottom with evenly distributed air holes in the pipe. Air is pumped from bottom to top through the compost material and discharged from the device through the top exhaust pipe. The oxygen concentration is detected by the zirconia sensor near the exhaust hole during discharge. The exhaust gas is treated or collected for detection.

[0060] (2) There is a partition above the air inlet pipe. The space above the partition is used for the aerobic degradation of compost materials, and the space below is used for the separation and collection of leachate. There is a leachate drain valve at the bottom of the device to collect leachate for analysis and testing. A large number of small holes with a diameter of 5mm are distributed on the partition to allow air to pass through evenly and to prevent compost material particles from falling into the leachate collection chamber.

[0061] (3) The top cover of the device can be fully opened for feeding and discharging. The cover is at an angle of 15-30 degrees to the horizontal, which facilitates the flow of condensate on the cover to the lower end of the cover. A condensate guide plate is provided at this end to guide the condensate to the collection tank and prevent backflow into the pile. A drain valve is provided at the bottom of the collection tank to drain excess condensate.

[0062] (4) After adding the materials that have been mixed in proportion, the upper surface of the materials should be about 20-30cm away from the sensor.

[0063] (5) The device is placed in a temperature-adjustable constant-temperature incubator as a heat preservation condition. The temperature is dynamically adjusted so that the set temperature is always 3-5°C lower than the pile temperature to ensure the heat preservation effect while avoiding affecting the self-heating of the pile. The pile temperature is measured by a temperature sensor inserted into the center of the pile. After the raw material is put in, the temperature of the incubator is set to 35°C as the starting temperature to activate the pile and make it heat up rapidly. When the self-heating of the pile exceeds 35°C, dynamic temperature adjustment is performed again.

[0064] Adopting such Figure 1 The control flow is shown.

[0065] After composting begins, an automatic aeration control program is activated. Oxygen concentration data is recorded throughout the composting process using an oxygen concentration measuring device, with a sampling interval of 6 seconds. Five control strategies are implemented: intermittent aeration without oxygen concentration feedback (12 minutes of aeration followed by 60 minutes of rest); conventional oxygen concentration feedback control aeration (target oxygen concentration of 5%, expected fluctuation range of 4-6%); and feedforward-feedback composite control aeration at different target oxygen concentrations (low, medium, and high oxygen concentrations, corresponding to target oxygen concentrations of 5%, 8%, and 11%, respectively, and an expected fluctuation range of ±1%). The aeration rate is set to 0.6 L / min / kg·DM, with the pump rate dynamically adjusted as the compost mass decreases. DM stands for Dry Matter.

[0066] The determination of T in feedforward control needs to be based on the actual state of the compost. During the heating period (the temperature at the center of the compost pile rises from room temperature to 50 degrees Celsius), the rate of microbial metabolism and oxygen consumption gradually increases, and T can be taken as 600-900s. When the temperature of the compost pile is above 50 degrees Celsius, the reaction enters the high-temperature maintenance period, and the rate of microbial metabolism gradually reaches its peak and maintains it for a period of time before slowly declining. T can be taken as 300-600s. When the temperature of the compost pile drops below 50 degrees Celsius, the easily degradable organic matter is gradually metabolized, and the rate of microbial metabolism gradually decreases. T can be taken as 900-3600s. The value of t is determined by the amount of material in the compost pile, the composting device, the aeration pipeline, and the state of the compost pile. In this case, the value can be 30-60s. In the early stage of composting, due to the high moisture content and small porosity of the compost pile, the gas convection effect will be weakened. Therefore, the value of t needs to be increased appropriately to achieve uniform purging. In the later stage of composting, the moisture content of the compost pile decreases, the porosity decreases, and the gas convection effect is better. The value of t can be decreased appropriately.

[0067] Oxygen concentration control effect as Figures 3 to 7 As shown.

[0068] The effects of different control strategies on oxygen concentration control are as follows: Figures 3 to 7 As shown, where Figure 3 , Figure 4 The oxygen concentration control effect of ordinary intermittent aeration without using the present invention and conventional oxygen concentration feedback control aeration is compared. Figures 5 to 7 To demonstrate the oxygen concentration control effect of this invention, target oxygen concentrations at different levels were used. The main fluctuation range was evaluated using the range of 90% of the data points. The oxygen concentration fluctuation range for intermittent aeration was 11.27%, experiencing significant fluctuations in each cycle and failing to maintain a stable oxygen concentration. The oxygen concentration fluctuation range for conventional oxygen concentration feedback control aeration was 4.36%, far exceeding the expected target range of 2%, also exhibiting significant oscillations within the same reaction cycle, with the system's oxygen concentration repeatedly switching between anoxic and aerobic states. The oxygen concentration fluctuation ranges for low, medium, and high-level feedforward-feedback composite control aeration were 1.55%, 1.04%, and 0.83%, respectively, all lower than the expected target range of 2%. Compared to conventional oxygen concentration feedback control, the control effect (percentage reduction in fluctuation range) was improved by 64%, 76%, and 81%, respectively, with an average improvement of 74%.

[0069] In laboratory-scale small-scale closed composting devices or when non-contact real-time oxygen concentration detection of exhaust gas, the gas diffusion delay phenomenon causes measurement lag. This invention adopts an automatic control strategy of delay compensation feedforward-oxygen concentration feedback composite control to achieve high-precision control of oxygen concentration during composting.

Claims

1. A method for controlling constant-oxygen aeration in composting based on feedforward-feedback composite control, characterized in that, It consists of two parts: (a) Oxygen concentration feedback control The oxygen measuring device monitors the oxygen concentration in the composting device online and inputs the data into the control system. According to the preset purpose, the target oxygen concentration range is set in the control system. When the input real-time data is within the range, the aeration system does not operate and maintains the original state. When the input real-time data is lower than the lower limit of the range, the control system turns on the aeration system, and then the oxygen concentration gradually increases until the input real-time data exceeds the upper limit of the range, at which point the aeration system is turned off. During this period, the aeration system remains on. When the input real-time data exceeds the upper limit of the interval and the aeration system is shut down, the oxygen concentration in the pile is gradually consumed and decreases under the action of microorganisms until the input real-time data is lower than the lower limit of the interval and the aeration system is restarted. During this period, the measurement will be delayed due to diffusion delay. Feedforward control with delay compensation is used to intervene in this problem. (ii) Delay-compensated feedforward control The purpose of delayed compensation feedforward control is to provide a brief period of homogenization purging during the shutdown of the feedback control aeration system. By utilizing the forced convection effect generated by the purging, the gas circulation within the device is promoted, allowing the gas inside the pile to mix rapidly with the gas in the sensor area. This shortens the measurement lag time, enabling the sensor readings to more accurately reflect the true condition inside the pile. At the same time, its short-term operation avoids its impact on the normal aeration system and prevents overshooting of the oxygen concentration control system.

2. The composting constant-oxygen aeration control method based on feedforward-feedback composite control according to claim 1, characterized in that, The aforementioned delay compensation feedforward control employs a separate loop independent of the feedback control, or a coupled control loop linked with the feedback control.

3. The composting constant-oxygen aeration control method based on feedforward-feedback composite control according to claim 2, characterized in that, The method of using a separate loop control independent of feedback control is as follows: the parameters include the sleep period T and the working period t, where T is the running time of the feedback control. During this period, the feedforward control is in a sleep state. After the sleep period ends, it enters the working period and runs briefly for t. During this period, the aeration device starts to work. If the current situation is in the "aeration on - oxygen concentration rise" stage of feedback control, the feedforward control will not affect the original state. If the current situation is in the "aeration off - oxygen concentration decrease" stage of feedback control, the aeration system will be turned on briefly to accelerate gas diffusion, make the gas in the composting device mix evenly, reduce the measurement lag time, and enable the data measured by the sensor to reflect the real situation in the compost pile in a timely manner.

4. The composting constant-oxygen aeration control method based on feedforward-feedback composite control according to claim 3, characterized in that, The method of using a coupled control loop with feedback control linkage is as follows: Add the function of judging real-time data and historical trends in feedback control to the feedforward control to further reduce unnecessary aeration: if the data has been on an upward trend in a recent period, the feedforward control loop will always remain in a dormant state; if the data has been on a downward trend in a recent period or has remained basically unchanged, the feedforward control loop will maintain a normal working state, that is, different T and t parameters are set according to the composting stage.

5. The composting constant-oxygen aeration control method based on feedforward-feedback composite control according to claim 3, characterized in that, The method employs a coupled control loop with feedback control linkage, specifically by using a machine learning model to predict key parameters: The oxygen concentration consumption rate is calculated based on the oxygen concentration and time to determine whether the oxygen concentration is increasing or decreasing. If the oxygen concentration is increasing, T is a large value; if the oxygen concentration is decreasing, the values ​​of T and t are first set as a baseline, and then adjusted according to the prediction rules. Prediction rule: Record the time T' during which the rate of change of oxygen consumption rate changes from a positive value to 0 during the oxygen concentration increase phase. This value represents the state from a net increase in oxygen concentration to 0, reflecting the intensity of oxygen consumption by microorganisms. When microorganisms have a strong demand for oxygen, they need to be aerated for a long time to achieve a net increase in oxygen concentration, and aeration will stop after exceeding the target threshold. If the value of T' gradually increases, the value of T should be gradually decreased, that is, the dormancy period should be reduced, and vice versa. The value of t is preset to several fixed values.

Images