Solid waste coupled biological organic fertilizer performance prediction and process parameter inversion method

Abstract

The present application relates to a kind of solid waste coupling biological organic fertilizer performance prediction and process parameter inversion method, comprising: obtaining solid waste VOCs characteristic data, establishing auxiliary material particle size and diffusion resistance mapping table, screening carbon-based material and extracting carbon-nitrogen ratio balance coefficient;Using optimization algorithm to search auxiliary material ratio initial solution, according to the relationship between carbon-based material proportion and critical threshold, correct proportion and adjust humidity;According to the corresponding relationship of the modified proportion calculation adsorbent regeneration cycle and degradation period, extract temperature control node and mark switching time as temperature segmentation reference point;According to reference point, divide candidate time window, identify stable temperature platform in cooling stage to determine initial inoculation time;Recalculate metabolic activity peak time and adjust secondary inoculation time to constant temperature safety section midpoint accordingly;Generate equipment execution instruction set containing dosage, adding order and injection timing, package control message and transmit to field controller.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to a method for predicting the performance and retrieving process parameters of solid waste coupled with bio-organic fertilizer. Background Technology

[0002] Antibiotic residues and chemical sludge are hazardous solid wastes, and their harmless disposal is crucial to environmental safety and resource recycling. While traditional composting and fermentation technologies can handle ordinary organic waste, existing treatment methods often face a dilemma between safety and efficiency when dealing with these special solid wastes. Some methods excessively pursue high residue utilization rates and shortened fermentation cycles, neglecting the actual degradation process of antibiotic molecules within the compost pile, resulting in excessive residues in the finished product. Other methods employ extremely conservative process parameters, which, while ensuring safety compliance, result in extremely low residue utilization rates, poor economic efficiency, and fail to truly solve the problem of solid waste disposal.

[0003] The root of this dilemma lies in the temporal conflict between the antibiotic degradation process and the formation of compost fertility. Antibiotic molecules are stable and require sustained exposure to high temperatures for an extended period to fully decompose; for example, penicillin requires the core temperature of the compost pile to be maintained above 62 degrees Celsius for at least seven days. However, optimal composting conditions often require phased temperature fluctuations beyond this to promote the rotation of different functional microbial communities and nutrient conversion. When process parameters favor antibiotic degradation, microbial activity is inhibited, leading to incomplete composting; conversely, antibiotics may not be fully decomposed even when fertility targets are met.

[0004] Further complicating the situation is that the release patterns of volatile organic compounds (VOCs) in chemical sludge are not synchronized with the antibiotic degradation process. During the initial fermentation stage, as the fermentation pile heats up, VOCs are released in concentrated bursts, creating odor peaks. If forced ventilation is implemented at this point to accelerate antibiotic degradation, undegraded toxic gases will be directly released into the atmosphere. However, reducing ventilation during this stage to prolong the gas's residence time within the pile will inhibit aerobic microbial activity due to insufficient oxygen supply, thus delaying antibiotic biodegradation. This process contradiction between different pollutant removal pathways makes it difficult for traditional fixed-parameter formulations to simultaneously address multiple safety constraints.

[0005] Therefore, how to dynamically coordinate multiple process links such as temperature control, ventilation system, and material ratio to achieve the dual goals of high-proportion disposal of drug residues and quality of composted products, while ensuring complete degradation of antibiotics and no excessive emission of volatile organic compounds, has become a key issue that urgently needs to be addressed in this field. Summary of the Invention

[0006] The purpose of this invention is to provide a method for predicting the performance and retrieving process parameters of solid waste coupled with bio-organic fertilizer, in order to solve the technical problems of obstructed diffusion of volatile organic compounds and difficulty in accurately controlling the timing of microbial inoculation during solid waste treatment.

[0007] The technical solution of the present invention is as follows:

[0008] This invention provides a method for predicting the performance and retrieving process parameters of solid waste coupled with bio-organic fertilizer, mainly including:

[0009] Obtain volatile organic compound characteristic data of target solid waste, calculate the required specific surface area for gas-solid contact, establish a mapping table between auxiliary material particle size and diffusion resistance, screen carbon-based materials, and extract the dynamic equilibrium coefficient of carbon-nitrogen ratio of auxiliary material combination.

[0010] Based on the dynamic equilibrium coefficient of carbon-nitrogen ratio, an optimization algorithm is used to search for the initial solution of the excipient ratio that maximizes adsorption capacity. The initial solution is corrected according to the relationship between the proportion of carbon-based materials and the critical ratio threshold, and the humidity regulation capability of the excipients is adjusted to a set range to obtain the corrected excipient ratio scheme.

[0011] The relationship between the number of adsorbent regeneration cycles and the degradation cycle was calculated based on the revised excipient ratio scheme, and the temperature control node sequence was extracted. The switching time of the temperature gradient between the high temperature maintenance interval and the cooling interval was marked and its timestamp was obtained as the temperature segmentation benchmark point to divide the temperature control interval.

[0012] Candidate time windows for adding microbial agents are divided according to temperature segmentation benchmarks. Periods when the temperature exceeds the threshold are excluded. During the cooling phase, a stable temperature platform that meets the requirements for microbial colonization is identified, and the start time of this platform is taken as the initial inoculation time.

[0013] The peak time of bacterial metabolic activity is estimated based on the initial inoculation time. If the interval between the peak time and the end of the cooling phase is insufficient, the second inoculation time is adjusted to the midpoint of the subsequent constant temperature phase when the temperature is safe and the duration is sufficient.

[0014] Based on the secondary inoculation time and the revised excipient ratio scheme, a set of equipment execution instructions is generated, which includes the dosage, dosing sequence and injection timing. The temperature setpoint and humidity adjustment parameters are encapsulated into control messages and transmitted to the field controller.

[0015] The beneficial effects of this application are as follows: A method for predicting the performance and inverting process parameters of solid waste coupled with bio-organic fertilizer. In use, this invention first establishes a correlation mapping between auxiliary material particle size and diffusion resistance, and screens high-porosity carbon-based materials. Then, based on these materials, a particle swarm optimization algorithm is used to optimize the synergy between the carbon-nitrogen ratio balance coefficient and porosity, and the proportion of carbon-based materials is forcibly increased to above the critical threshold. This effectively constructs a porous media environment with high adsorption capacity, solving the problems of hindered diffusion of volatile organic compounds and insufficient adsorption capacity from the source. Simultaneously, based on the correspondence between the adsorbent regeneration cycle and the degradation cycle, the temperature switching moment is extracted as a segmentation benchmark. This benchmark is used to divide candidate time windows, from which... After eliminating the overheating period, a cooling zone is obtained. Then, a stable temperature plateau that meets the colonization time requirement is identified within the cooling zone to determine the timing of the initial inoculation. Furthermore, based on the peak metabolic activity time, the secondary inoculation is adjusted to the midpoint of the safe period during the isothermal maintenance phase, where the temperature is below the inactivation critical point and covers the survival rate increase period. This achieves decoupling and synergy between high-temperature degradation and low-temperature inoculation in terms of timing, ensuring sufficient degradation of antibiotics during the high-temperature period while avoiding heat shock inactivation of the bacterial agent and insufficient colonization time. Finally, the field equipment is driven to execute the process by encapsulating control messages containing the dosage, dosing sequence, and injection timing, which significantly improves the degradation efficiency of volatile organic compounds in solid waste and shortens the treatment cycle. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for predicting the performance and inverting process parameters of solid waste coupled with bio-organic fertilizer according to the present invention.

[0017] Figure 2 This is a flowchart of the auxiliary material ratio optimization and correction process of the method described in this invention;

[0018] Figure 3 This is a flowchart of the temperature segmentation and inoculation timing control sub-flowchart of the method described in this invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and are not intended to limit the invention; that is, the described embodiments are merely some embodiments of the invention, and not all embodiments. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0021] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0022] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0023] This invention provides a specific embodiment of a method for predicting the performance and inverting process parameters of solid waste coupled with bio-organic fertilizer, primarily focusing on two hazardous solid wastes: antibiotic residue and chemical sludge.

[0024] like Figures 1-3 The specific method for predicting the performance and inverting process parameters of solid waste coupled with bio-organic fertilizer in this embodiment may include:

[0025] In step S101, the initial concentration data of volatile organic compounds (VOCs) of the target solid waste (i.e., the first concentration value) and the gas-phase diffusion coefficient (i.e., the first diffusion rate) are obtained. The corresponding gas-solid contact specific surface area requirement (i.e., the first surface area) is calculated based on the initial VOC concentration data. A correlation mapping table between the particle size distribution of the auxiliary materials and the diffusion resistance coefficient of volatile substances is established by combining the gas-phase diffusion coefficient. Auxiliary material combination schemes with carbon-based material porosity greater than the preset porosity standard are selected from the correlation mapping table. The dynamic equilibrium coefficient of the carbon-nitrogen ratio of each component in the auxiliary material combination scheme is extracted.

[0026] Optionally, a first concentration value and a first diffusion rate of the target solid waste are obtained. A random forest regression algorithm is used to process the first concentration value and the first diffusion rate to obtain the first surface area.

[0027] In one possible implementation, the initial concentration of VOCs in the headspace phase of the sample is determined using gas chromatography-mass spectrometry (GC-MS) according to HJ 643—2013 standard. For example, the initial concentration was measured to be 2350 mg / m³ (using toluene as the characterization factor). The gas-phase diffusion coefficient is obtained by consulting a physicochemical handbook; for example, at a composting ambient temperature of 35°C, it is... The random forest regression model is trained using composting experiment data under different concentration and diffusion rates as the training set, with the measured optimal specific surface area as the output label. The model then predicts and outputs the first surface area, for example, 126 m² / kg.

[0028] Based on the first surface area, a match is retrieved from the associated mapping table, the corresponding first particle size value is extracted, and the corresponding diffusion resistance coefficient is directly read from the mapping table as the first resistance value.

[0029] The correlation mapping table is pre-defined based on the mass transfer resistance coefficient formula. The system calculates and records the effective porosity, tortuosity, and corresponding diffusion resistance coefficients of various excipients under different particle size distributions. Example: Using a primary surface area of ​​126 m² / kg as the primary key, a search in the mapping table yields a carbon-based excipient with an average particle size of 3.2 mm. Its primary resistance value (diffusion resistance coefficient) is directly read from the mapping table as 2.45 × 10⁻⁶. 5 s / m².

[0030] Obtain the first porosity corresponding to the first resistance value. If the first porosity is greater than the preset porosity standard, then the carbon-based material is determined as the first combination item, and combined with other types of auxiliary materials to form an auxiliary material combination scheme.

[0031] For example, the preset porosity standard is set at 55% according to the aerobic composting engineering standard (CJJ / T 52). The porosity of the above-mentioned carbon-based auxiliary material is 62% under the condition of an average particle size of 3.2 mm. Since 62% > 55%, the carbon-based material is determined to pass the initial screening and is used as the first combination item (carbon-based skeleton main material). Subsequently, structural auxiliary materials and nitrogen source auxiliary materials are added on the basis of this carbon-based material at a mass ratio of 7:2:1 to form a complete auxiliary material combination scheme.

[0032] The carbon-nitrogen ratio of each component is extracted from the auxiliary material combination scheme, consumption rate weights are introduced, and K-means clustering algorithm is used for clustering to determine the first degree of balance.

[0033] In one possible implementation, an elemental analyzer is used to determine the total carbon and total nitrogen content of each component in the excipient combination scheme. The calculated carbon-to-nitrogen ratios are 75.8, 60.7, and 11.4, respectively. A consumption rate weighting operator (carbon consumption factor of 0.85 and nitrogen consumption factor of 0.12) is introduced to weight and correct the original carbon-to-nitrogen ratios, yielding corrected eigenvalues ​​of 537.0, 430.0, and 80.8. These corrected eigenvalues ​​are then input into a K-means clustering algorithm, with the number of clusters set to K=3 (dividing the clusters into high-carbon, high-degradation, medium-carbon, stable, and low-carbon, fast-acting groups). After clustering, the overall coordination of the cluster center combinations is evaluated, and a first balance degree is calculated, for example, 0.78 (ranging from 0 to 1, with higher values ​​indicating stronger buffering capacity).

[0034] The first solid waste quantity is calculated based on the first balance degree. If the first solid waste quantity is less than the preset solid waste threshold, the first solid waste quantity is discarded to obtain the second solid waste quantity.

[0035] According to a specific embodiment of the present invention, firstly, for the currently selected auxiliary material combination scheme, the first solid waste quantity is calculated based on the product of the system's total daily processing design load and the first balance degree (dimensionless), wherein the first balance degree characterizes the buffering capacity of the auxiliary material combination to the odor of chemical sludge and carbon-nitrogen imbalance. The higher the balance degree value, the stronger the buffering capacity, allowing the system to operate with a higher target solid waste addition amount; the total daily processing design load is the daily processing capacity of the fermentation system under rated operating conditions (unit: tons / day). Subsequently, the calculated first solid waste volume is compared with a preset solid waste threshold, which is determined based on the plant's lowest daily economic operating cost or the equipment's lowest stable stirring load, for example, set to 10 tons / day. If the first solid waste volume is less than this threshold (e.g., the calculated value is only 6 tons), it indicates that the buffering capacity of the current auxiliary material combination is insufficient. To dispose of this small amount of sludge, expensive or large quantities of auxiliary materials are required, which is economically infeasible. In this case, the auxiliary material combination scheme is directly discarded, and no further correction or re-ratioing of the solid waste volume is made. Instead, the next set of candidate auxiliary material combinations is retrieved, the first solid waste volume is recalculated, and compared again until the first solid waste volume calculated by a certain set of auxiliary materials is greater than or equal to the threshold. At this point, the first solid waste volume that meets the threshold condition is directly retained as the second solid waste volume for subsequent actual addition control. Through the above threshold elimination mechanism, it is ensured that the final adopted second solid waste volume can meet the economic requirements of stable system operation while avoiding cost waste caused by inefficient auxiliary material schemes.

[0036] By analyzing the first volatilization amount of the second solid waste within a preset time period, the first contact surface corresponding to the first volatilization amount is calculated based on the gas-solid interface adsorption mass transfer equation.

[0037] In a specific embodiment, the preset time period refers to the initial heating phase of fermentation after the chemical sludge enters the composting tank, during which VOCs are released violently. This is typically set to the first 0-48 hours after fermentation begins. Within this time period, based on the determined amount of the second solid waste, the total mass of VOCs released in the first 48 hours is predicted and denoted as the first volatile matter. (Unit: g). To completely trap these volatile VOCs within the porous additive, the minimum required total gas-solid contact area needs to be calculated. (i.e., the first contact surface, unit: m²), specifically: first check the saturated adsorption capacity of the selected carbon-based porous excipient for the target VOCs. (Unit: g / m², representing the mass of target VOCs that can be adsorbed per unit area of ​​the auxiliary material surface), and set the adsorption efficiency constant. (A value of 0.85–0.95 is used as a safety redundancy coefficient to offset adsorption attenuation under actual operating conditions). Subsequently, a reverse derivation is performed using an external surface coverage model based on gas-solid interface adsorption mass transfer, i.e. The result of this calculation This value will serve as the core boundary constraint for subsequent auxiliary material formulation, ensuring that the actual adsorption capacity of the selected auxiliary material scheme is not lower than this value, thereby achieving effective interception of initial volatile VOCs in engineering.

[0038] In step S102, for the dynamic equilibrium coefficient of the carbon-nitrogen ratio, a particle swarm optimization algorithm is used to search for an initial solution of the excipient ratio that maximizes the adsorption capacity of volatile organic compounds (i.e., the first excipient ratio solution). If the proportion of carbon-based materials in the initial solution of the excipient ratio (i.e., the first carbon-based material proportion) is lower than the critical ratio threshold corresponding to the diffusion resistance coefficient of volatile substances (i.e., the first critical ratio threshold), then the mass fraction of carbon-based materials in the ratio is increased to a multiple of the critical ratio threshold. Simultaneously, the humidity regulation capability parameter of the excipients is adjusted to within the set range of the porosity saturation of the carbon-based materials to obtain the corrected excipient ratio scheme.

[0039] In this embodiment, the dynamic equilibrium coefficient of the carbon-nitrogen ratio is obtained. A particle swarm optimization algorithm is then used to search for the first excipient ratio solution that maximizes the adsorption capacity of volatile organic compounds, based on the dynamic equilibrium coefficient of the carbon-nitrogen ratio.

[0040] In one possible implementation, this embodiment provides a method for maximizing VOCs adsorption capacity by optimizing the ratio of the first excipient based on particle swarm optimization (PSO) algorithm, specifically:

[0041] (1) Initialization: Set the particle swarm size to 30, and encode each particle as a three-dimensional mass percentage vector. These correspond to the mass percentages of carbon-based materials, nitrogen source materials, and structural additives, respectively, with the following constraints: The initial values ​​for each dimension are randomly generated and normalized within the range of [0, 100%].

[0042] (2) Fitness function: With the goal of maximizing the theoretical total adsorption of VOCs, the fitness function F is constructed as follows:

[0043]

[0044] Where C / N is the dynamic equilibrium coefficient of the carbon-nitrogen ratio. The F value represents the porosity (%) of the material, and the weighting coefficients (0.6, 0.3, 0.1) were determined through previous orthogonal experiments. The higher the F value, the greater the theoretical adsorption capacity of the formulation.

[0045] (3) Iterative update and constraint handling: Particle velocity and position are updated according to the standard PSO formula:

[0046]

[0047]

[0048] Among them, inertia weight Learning factor , , A random number between [0,1]; This represents the optimal position in the history of an individual particle. This represents the globally best historical position. After each position update, the three-dimensional components of X are normalized to ensure that their sum equals 100%.

[0049] (4) Convergence conditions and output: When the fluctuation of the optimal fitness F of the population is less than 1% after 50 consecutive iterations, or when the total number of iterations reaches 500, the optimization will be terminated and the current global optimal position will be output as the final first auxiliary material ratio solution.

[0050] As an example, the typical optimal ratio obtained through the above steps is: 45% biochar, 35% livestock and poultry manure, and 20% straw (by mass). Under this ratio, the dynamic balance coefficient of carbon-nitrogen ratio and porosity are optimal in synergy, which can maximize the adsorption capacity of VOCs.

[0051] Based on the first auxiliary material ratio solution, the proportion of the first carbon-based material in the first auxiliary material ratio solution is extracted. That is, the percentage value of the corresponding carbon-based material in the ratio solution is directly used as this proportion. In this embodiment, the first auxiliary material ratio solution obtained by particle swarm optimization is a vector (biochar 45%, livestock manure 35%, straw 20%). The proportion of the first carbon-based material can be obtained by directly reading the mass percentage value of the carbon-based material item (i.e., biochar) from this ratio vector; since the total mass of the auxiliary materials in this ratio is normalized to 100%, the ratio of carbon-based materials to the total mass of the auxiliary materials is 45%.

[0052] For the proportion of the first carbon-based material, obtain the first critical ratio threshold corresponding to the diffusion resistance coefficient of volatile substances. If the proportion of the first carbon-based material is lower than the first critical ratio threshold, increase the proportion of the first carbon-based material to a multiple of the first critical ratio threshold to obtain the proportion of the second carbon-based material.

[0053] According to an embodiment of the present invention, regarding the proportion of the first carbon-based material, firstly, based on the diffusion resistance coefficient of volatile substances calculated in step S101, a first critical proportion threshold corresponding to the coefficient is retrieved from the threshold database built into the system. It should be noted that this threshold is pre-set to be positively correlated with the diffusion resistance coefficient, that is, the larger the resistance coefficient, the higher the corresponding critical proportion threshold. Subsequently, the proportion of the first carbon-based material (such as the aforementioned 45% biochar proportion) is compared with the threshold: if the first proportion is lower than the threshold, a compensation mechanism is triggered to increase the proportion of the first carbon-based material to 1.1 to 1.2 times (preferably 1.15 times) of the first critical proportion threshold, and the product value is used as the second carbon-based material proportion. This multiple value serves as a safety redundancy coefficient, aiming to ensure that the adjusted amount of carbon-based material added is absolutely greater than the critical safety line, thereby forcibly opening gas channels for the pile and providing sufficient adsorption sites, effectively avoiding the problem of odor accumulation caused by excessive diffusion resistance.

[0054] Based on the proportion of the second carbon-based material, the humidity regulation capability parameter of the first auxiliary material corresponding to the proportion of the second carbon-based material is obtained. Specifically, the saturated water holding capacity (i.e., the ratio of maximum water absorption mass to dry mass) of each individual auxiliary material (including carbon-based material, nitrogen source material, and structural auxiliary material) under standard conditions is determined in advance through experiments. Then, the saturated water holding capacity of each individual material is weighted and summed using the proportion of the second carbon-based material and the actual proportion of other components. The resulting weighted average value is the humidity regulation capability parameter of the first auxiliary material under this ratio. This parameter reflects the ability of the auxiliary material combination to absorb and lock in moisture under a given ratio.

[0055] Based on the humidity regulation capability parameter of the first auxiliary material, the porosity saturation setting range of the carbon-based material is obtained. The method for determining this range is as follows: First, the initial moisture content of the target solid waste material is measured. Then, based on the optimal moisture range for microbial activity during aerobic fermentation (in this embodiment, the mass moisture content is taken as 50% to 60%), the percentage of water volume in the pores of the carbon-based material relative to its total pore volume (i.e., porosity saturation, equivalent to saturation in soil mechanics) is calculated in reverse. The reasonable fluctuation range of humidity regulation capability is the benchmark range for determining whether the humidity regulation capability meets the fermentation requirements.

[0056] If the humidity regulation capability parameter of the first auxiliary material is not within the set range of the porosity saturation of the carbon-based material, then adjust the humidity regulation capability parameter of the first auxiliary material to the set range of the porosity saturation of the carbon-based material to obtain the humidity regulation capability parameter of the second auxiliary material.

[0057] In this embodiment, the calculated first auxiliary material humidity regulation capability parameter is compared with the aforementioned porosity saturation setting range. If the parameter is not within the range (e.g., below the lower limit or above the upper limit), it is forcibly set to the lower or upper limit of the range through a limiting adjustment method, so that it falls within the setting range. The adjusted value is then used as the second auxiliary material humidity regulation capability parameter. This adjustment only changes the humidity control target and does not involve changes in the proportion of carbon-based materials, thus avoiding iterative cycles.

[0058] Based on the humidity regulation capability parameter of the second auxiliary material and the proportion of the second carbon-based material, the solution for the first auxiliary material ratio is updated to obtain the solution for the second auxiliary material ratio.

[0059] In one possible implementation, the update logic is as follows: First, the proportion of carbon-based materials in the first auxiliary material ratio solution is directly replaced with the proportion of carbon-based materials in the second solution; then, the proportions of the remaining nitrogen source materials and structural auxiliary materials are scaled proportionally according to their original ratios, so that their sum is normalized to 100% along with the sum of the carbon-based material proportions; simultaneously, the humidity regulation capability parameter of the second auxiliary material is written into the ratio solution as a background control variable for subsequent process control (such as water replenishment or humidity adjustment), thus completing the update of the ratio solution. In step S103, the correspondence between the number of adsorbent regeneration cycles and the degradation cycle is calculated based on the corrected auxiliary material ratio scheme. The temperature control node sequence required for the degradation process is extracted from the correspondence. The switching time between the boundary temperature of the high-temperature maintenance interval and the temperature gradient of the cooling interval is marked in the temperature control node sequence. The timestamp corresponding to the switching time is obtained as the temperature segmentation reference point.

[0060] Optionally, the adsorbent ratio is obtained according to the modified excipient ratio scheme. A linear interpolation algorithm is used to calculate the adsorbent regeneration cycle number and degradation period corresponding to the ratio, resulting in a table showing the correspondence between cycle number and degradation period.

[0061] In this embodiment, based on the revised excipient ratio scheme, the mass percentage of carbon-based materials (such as biochar) is directly extracted as the adsorbent ratio value, denoted as X. The system pre-stores a baseline process correlation table based on historical experimental data, which records multiple historical adsorbent ratio values ​​and their corresponding regeneration cycle numbers and degradation periods. When the current ratio value X is between two adjacent historical values ​​in the table... and between( When considering the number of regeneration cycles and the degradation period, known linear interpolation formulas are used respectively. Calculation, where and For the corresponding table and The target parameter value (number of cycles or degradation period) is Y, which is the target parameter value under the current ratio. The above interpolation calculation is performed independently for the number of regeneration cycles and the degradation period, and the results are entered into the corresponding table for the current batch. For example, if the carbon-based material accounts for 40% in the corrected ratio, and 30% in the baseline table corresponds to 5 cycles and 15 days of degradation, while 50% corresponds to 9 cycles and 10 days of degradation, then substituting into the formula yields the number of regeneration cycles as follows: The degradation period is... This generates a table showing the correspondence between the number of cycles and the degradation period for this formulation.

[0062] Extract the temperature control points for the degradation process from the corresponding table. Generate a temperature control node set based on these temperature control points. For each temperature control node set, obtain the high temperature value and cooling rate.

[0063] In this embodiment, after generating a correspondence table between cycle number and degradation period, the system matches and retrieves the standard temperature control curve corresponding to the degradation period from the system's preset process curve database according to the degradation period determined in the table. This curve is stored in the form of several temperature control points (each control point includes a time coordinate and a temperature setpoint). Extracting all control points in chronological order constitutes a temperature control node set. Subsequently, the highest temperature value is read from this temperature control node set as the high temperature value. At the same time, the temperature drop per unit time (e.g., a setpoint of -2℃ / hour) is directly extracted from the cooling segment of the curve as the cooling rate. This cooling rate is a fixed command value pre-calibrated in the process curve and is used for temperature gradient control in the subsequent aerobic fermentation process.

[0064] Determine the intersection state of the high temperature value and the cooling rate. If the high temperature value and the cooling rate intersect, mark the switching point corresponding to the intersection state. Obtain the corresponding system timestamp based on the switching point.

[0065] In one possible implementation, the system reads the current target temperature value and its target rate of change (i.e., the temperature change per unit time) in real time according to a set sampling period. When the target temperature value is detected to be continuously equal to the high temperature value (with an allowable fluctuation range of ±0.5℃) within a preset duration, and the target rate of change changes from zero to a negative value, it is determined that the high temperature value and the cooling rate intersect on the time axis. This intersection point is the switching point where the high temperature maintenance phase ends and the cooling phase begins. The system immediately marks this switching point and records the corresponding system timestamp as the starting reference for subsequent cooling process control. The sampling period can be set according to the thermal inertia of the fermentation system and the response speed of the temperature sensor, preferably 1 to 10 minutes, more preferably 5 minutes; the preset duration can be set according to the need to avoid misjudgments caused by instantaneous temperature fluctuations, preferably 1 to 4 hours, more preferably 2 hours. The above parameters can be scaled proportionally according to the actual scale of the fermentation tank (small-scale or engineering-scale), but the sampling period should not exceed one-tenth of the preset duration to ensure that enough sampling points are obtained within the judgment window, thus ensuring the statistical reliability and timeliness of the cross-judgment.

[0066] A decision tree algorithm is used to classify the system timestamps, resulting in a set of classified timestamps. The target timestamp is then extracted from this set and used as a temperature segmentation reference point. Based on this reference point, the temperature control intervals for the degradation period are divided, resulting in a first and a second temperature control interval.

[0067] In this embodiment, the system uses all candidate system timestamps recorded when the intersection state is marked as the input set, and then uses a rule-based decision tree algorithm to classify and filter the set step by step. The classification process is executed sequentially according to preset hard process constraint rules: the first level node determines whether the timestamp is within a preset working day or production shift, and removes timestamps from non-working periods; the second level node determines whether the core temperature of the reactor corresponding to the timestamp has been continuously greater than 62°C and maintained for seven days to ensure that the antibiotics fully meet the degradation requirements, and removes timestamps that do not meet the high temperature maintenance condition; the third level node determines whether the volatile gas release concentration at that moment has exceeded the odor peak and entered the decline stage, and removes timestamps that have not yet exceeded the peak. Through the above-mentioned layer-by-layer filtering of "yes / no" rules, a qualified timestamp set after classification is obtained, and the unique timestamp that meets all process constraint conditions is extracted as the target timestamp. Finally, the target timestamp is set as the temperature segmentation reference point. Using this reference point as the dividing line, the fermentation temperature control sequence is divided into two intervals: the interval before the reference point is defined as the first temperature control interval, which is used to maintain high temperature to achieve antibiotic degradation; the interval after the reference point is defined as the second temperature control interval, which is used to implement cooling strategies to adapt to the subsequent microbial inoculation process, thereby decoupling the process conflict between high-temperature degradation and low-temperature inoculation in terms of time sequence.

[0068] In step S104, candidate time windows for adding the inoculant are divided based on temperature segmentation benchmarks. If a period of temperature exceeding the inoculant's heat shock temperature threshold exists within a candidate time window, that period is excluded, and the candidate window is shifted forward to the stage where the temperature gradient begins to decrease in the cooling interval. Within this stage, a stable temperature plateau is identified where the temperature fluctuation buffer duration meets the requirements for colonization time. The start time of the stable temperature plateau is determined as the initial inoculation time node.

[0069] In this embodiment, temperature segmentation reference points are obtained. Candidate windows for adding inoculants are then divided based on these reference points to obtain an initial candidate window set. Real-time temperature sequences are then obtained from the initial candidate window set.

[0070] According to an embodiment of the present invention, taking the temperature segmentation reference point as the starting time and the preset end time of the degradation period as the ending time, several continuous time segments are divided within this time range according to fixed time intervals (e.g., every 12 hours). Each time segment serves as a candidate window for adding the inoculant, and all candidate windows together constitute an initial candidate window set. After obtaining the initial candidate window set, the system extracts temperature data segments corresponding to the time interval of each candidate window from the stored real-time temperature sensor historical data to obtain the real-time temperature sequence corresponding to each candidate window. Subsequent steps will evaluate whether each candidate window meets the temperature and time conditions for inoculation based on these temperature sequences, thereby selecting the optimal addition window. It should be noted that the candidate window is only used to specify the time range for data extraction and does not change or generate temperature values; the obtained temperature sequences all originate from the actual collected sensor data.

[0071] If any candidate window contains a real-time temperature sequence at any point in time that exceeds a preset heat shock temperature threshold, that candidate window is discarded entirely. After removing all candidate windows containing high-temperature points, the remaining set of candidate windows constitutes the cooling zone. For the real-time temperature sequence within the cooling zone, the sliding window method is used to calculate the rate of temperature change: a fixed window width and sliding step size are set. Within each window, with time as the independent variable and temperature as the dependent variable, a slope is obtained through linear fitting using the least squares method. This slope is the average cooling rate of that window. The slopes of all windows are arranged in chronological order to form a gradient value sequence.

[0072] In this embodiment, the preset heat shock temperature threshold is 50°C (this value is set based on the inactivation temperature of conventional compound microbial agents). For each candidate window in the initial candidate window set, the real-time temperature sequence corresponding to that window is retrieved. If the temperature value of any sampling point in the sequence is higher than 50°C, the entire candidate window is excluded; all candidate windows that are not excluded constitute the cooling zone. Subsequently, the temperature change gradient within the cooling zone is calculated using the sliding window method, specifically setting the window width to 4 hours and the sliding step size to 1 hour. Within each window, the least squares method is used to perform linear regression on all temperature data points within the window, and the fitting formula is:

[0073]

[0074] in, This represents the number of sampling points within the window. For the first The time for each sampling point For the corresponding temperature value, and These represent the average time and temperature within the window, and the slope, respectively. This represents the cooling rate of the window (a negative value indicates cooling). The windows are moved sequentially according to the sliding step size, and the slope of each window is calculated to obtain a gradient value sequence. This sequence reflects the continuous decreasing trend of temperature within the cooling zone, thus allowing for the selection of stable cooling periods suitable for adding microbial agents.

[0075] Identify the descending segment within the cooling zone based on the gradient value sequence. Obtain the temperature fluctuation within the descending segment. Calculate the buffer length for maintaining the temperature within a preset range based on the temperature fluctuation. If the buffer length exceeds the colonization period, the corresponding descending segment is designated as the stabilization platform. Obtain the starting point of the stabilization platform and designate it as the initial inoculation point.

[0076] In one possible implementation, the system identifies the descending segment within the cooling zone and determines the initial inoculation point through the following logic: First, the gradient value sequence is evaluated item by item. When multiple consecutive gradient values ​​(i.e., the cooling rate slope) are negative and their absolute values ​​are greater than a preset process control cooling rate threshold (e.g., not less than 0.5℃ / h, to distinguish them from the slight cooling caused by natural heat dissipation), this continuous period is marked as a descending segment. Second, the temperature fluctuation within this descending segment is obtained, specifically represented by the range (maximum value minus minimum value) or standard deviation. If the range or standard deviation is less than a preset allowable fluctuation value (e.g., range ≤ 2℃), it indicates that the pile temperature within this descending segment tends to be stable. Subsequently, it is determined whether the temperature value within this descending segment falls within a preset suitable temperature range for the microbial community (e.g., 35℃~40℃), and the length of time the temperature is continuously maintained within this range is counted, defined as the buffer length. If the temperature of this descending segment deviates from this range, it is not counted. When the buffer length exceeds the preset colonization period (e.g., 24 hours), the system determines that the descending segment has reached sufficient temperature stability and time leeway, and identifies it as a stable platform. Finally, the starting point of the stable platform on the time axis is extracted as the initial inoculation time node, which is used to trigger subsequent inoculum addition operations.

[0077] In step S105, the arrival time of the peak period of bacterial metabolic activity is calculated based on the initial inoculation time. If the interval between the arrival time and the end time of the temperature gradient in the cooling zone is less than the minimum colonization time of the bacterial community after inoculation, the second inoculation time is postponed to the isothermal maintenance phase after the end of the cooling zone. From the isothermal maintenance phase, a time period in which the temperature is below the critical temperature point for bacterial inactivation and the duration covers the rising period of the inoculation survival rate curve is selected. The midpoint of this time period is extracted as the adjusted second inoculation time.

[0078] Optionally, the initial vaccination time point and historical metabolic data can be obtained. The time series of historical metabolic data can be predicted using an autoregressive integral moving average model to obtain the peak metabolic activity arrival time.

[0079] In this embodiment, after acquiring historical metabolic data (i.e., time-series data of oxygen consumption rate or carbon dioxide release rate of previous batches), an autoregressive integral moving average model is constructed. The specific modeling steps are as follows: First, perform first-order or second-order differencing on the historical metabolic data (…). or To achieve sequence stabilization; subsequently, set and The search range is from order 0 to 5. By calculating the autocorrelation coefficient and partial autocorrelation coefficient for each order combination, and combining the Akaike Information Criterion (AIC) for automatic optimization, the order combination with the smallest AIC value is selected as the optimal model. and The model coefficients are solved using the well-known maximum likelihood estimation method, outputting a complete prediction equation. After modeling, the measured metabolic rate data sequence accumulated since the initial vaccination of the current batch is input into the model. A forward rolling prediction method is used (i.e., starting from the current measured point, after predicting the value at each time point forward, it is included in the historical sequence and the model input is updated, extrapolating point by point) to predict the metabolic rate change curves at each sampling time in the next 48 hours. Finally, the maximum point where the first derivative is zero and the second derivative is negative is found on the generated prediction curve. The timestamp corresponding to this maximum point is the time when the peak metabolic activity is reached.

[0080] The difference between the peak metabolic activity time and the preset end time of the cooling interval is calculated to obtain the interval between the peak and the end of the cooling interval. If the interval between the peak and the end of the cooling interval is less than the preset minimum colonization time, the temperature sequence data of the isothermal maintenance phase after the end of the cooling interval is extracted.

[0081] In this embodiment, the preferred example of the preset cooling interval termination time is 216 hours after fermentation starts (i.e., the end of day 9). Subtracting 216 from the number of hours corresponding to the peak arrival time gives the interval. If this interval is less than the preset minimum colony time (preferably 12 hours in this embodiment), temperature sequence data extraction is triggered. Specifically, the extraction method is as follows: using the cooling interval termination time as the starting point, a forward slice window (e.g., 72 hours) is set. A query is performed in the system database to extract all temperature data records (each record is a key-value pair of timestamp and temperature value) within the range from the starting point to 72 hours after the starting point. The query results are arranged in ascending order by time into a one-dimensional array, thus completing the temperature sequence data extraction for the isothermal maintenance phase after the cooling interval ends.

[0082] For temperature series data, a support vector regression model was used in conjunction with a pre-defined inactivation critical point for the inoculant to determine a safe isothermal range below the inactivation critical point. Specifically, timestamps were used as the basis for this determination. Using temperature values ​​at corresponding times as output labels, the radial basis function (RBF) is selected to map low-dimensional data to a high-dimensional space to capture subtle nonlinear fluctuations that may exist during the isothermal phase. An insensitive loss coefficient is introduced during model training. (Used to control the error tolerance of regression fitting) and penalty factor (To balance model complexity and fitting bias), using historical temperature fluctuation data from the later stages of similar fermentation processes as the training set, the regression function is obtained through quadratic programming. This function can smoothly fit future temperature change trends. Then, using a preset inactivation critical point for the microbial agent (42℃ in this example, for high-efficiency degradation microbial agents that are not heat-resistant at room temperature) as the upper limit of the safe temperature, the regression function is iterated along the time axis to solve the inequalities. For all continuous time intervals that satisfy this inequality Marked as a safe constant temperature section, among which and These are the start and end timestamps of the segment, respectively.

[0083] Candidate time periods with a duration longer than the preset survival rate curve rise period are extracted from the safe isothermal range. In this embodiment, after obtaining the set of safe isothermal ranges, the system calculates the duration of each safe isothermal range. Then, the pre-defined survival rate curve rise period (i.e., the shortest adaptation time required for the newly inoculated bacterial agent to undergo the logarithmic growth phase, determined by laboratory shake-flask culture kinetics experiments; in this example, it is 24 hours) is used as the screening threshold; all safe isothermal periods are traversed to determine whether each period meets the requirements. The timeframe is determined by the number of hours. If the condition is met, the entire segment is retained as a candidate time period; if not (e.g., a safe duration of only 8 hours, insufficient to cover the complete logarithmic growth phase of the inoculant), the segment is discarded. Ultimately, all retained safe isothermal segments with a duration greater than 24 hours are considered. Form a set of candidate time periods.

[0084] Obtain the start and end times of the candidate time period. Calculate the midpoint between the start and end times, and determine the midpoint as the adjusted time node for the second vaccination.

[0085] In step S106, a set of equipment execution instructions is generated based on the adjusted secondary inoculation time node and the revised auxiliary material ratio scheme, including the auxiliary material dosage, dosing sequence, and microbial agent injection timing. The temperature setpoint sequence and auxiliary material humidity regulation capability parameters from the set of equipment execution instructions are encapsulated into a control message. The control message is transmitted to the field controller of the solid waste treatment equipment.

[0086] In this implementation, the inoculation time and ratio are obtained. The inoculation time and ratio are processed using a random forest algorithm to obtain an initial instruction set. The dosage and order of administration are then extracted based on the initial instruction set.

[0087] The inoculation time includes the initial inoculation time node and the second inoculation time node determined in the aforementioned steps. The ratio value is directly obtained from the final output of the second auxiliary material ratio solution, which includes the final mass percentage of each component of carbon-based materials, nitrogen source materials, and structural auxiliary materials (e.g., biochar 42%, livestock and poultry manure 38%, straw 20%, with the sum normalized to 100%). Then, the inoculation time and ratio value are input into a pre-trained random forest algorithm model. This model adopts a multi-output regression framework and consists of multiple decision trees. Each tree determines node splitting based on input features (such as whether the proportion of carbon-based materials is greater than 40%, whether the second inoculation time is later than a preset threshold, etc.). The control parameters output by each tree are averaged to generate the final product. The initial instruction set is a data structure (such as JSON or XML format) containing the underlying operating logic of the equipment. Specifically, it includes parameters such as the motor speed of each auxiliary material feeding bin, the running time of the belt conveyor, the start-stop sequence of the mixer, and the valve opening control. Finally, the dosage and dosage sequence are extracted from the initial instruction set. The dosage is calculated by the model by multiplying the input ratio value by the total mass of the target solid waste in the current batch to obtain the absolute added mass of each auxiliary material (in tons or kilograms). The dosage sequence is the order in which each material enters the mixing tank according to the preset material mixing process specification, thus forming a complete equipment execution plan.

[0088] If the dosage exceeds the preset dosage threshold, the dosage sequence is rearranged to obtain an updated dosage sequence. The injection sequence corresponding to the updated dosage sequence is then obtained. Combining the injection sequence with the initial instruction set, the temperature setpoint is determined. The corresponding humidity value is extracted based on the temperature setpoint.

[0089] In this embodiment, the dosage threshold is determined by the maximum safe rated load of the solid waste treatment mixing equipment or the dosing silo (e.g., 5.0 tons). If the calculated dosage of a certain auxiliary material exceeds this threshold (e.g., 6 tons of carbon-based material are required, exceeding the 5-ton limit), the dosing sequence of the auxiliary material is rearranged in batches: that is, the excess auxiliary material is split into multiple batches and alternately interspersed between the dosing steps of other auxiliary materials. For example, the original dosing sequence is... (A represents excess carbon-based material), after rearrangement it becomes To avoid the risk of material blockage or uneven mixing caused by excessive addition in a single operation, an updated addition sequence is obtained. Then, based on the updated addition sequence, the system obtains a matching liquid microbial agent injection sequence. This injection sequence is the start-up time sequence of each microbial agent spraying device, and it has a strict causal alignment with the addition sequence. That is, if the time node of a certain material in the addition sequence is shifted due to rearrangement, the injection time node of the corresponding microbial agent is simultaneously shifted backward by the same amount to ensure that the microbial agent is sprayed onto the already uniformly mixed material. Finally, combining the injection sequence with the temperature range division information in the initial instruction set, the target temperature set value corresponding to each fermentation stage (e.g., 62℃ for the high-temperature degradation period and 38℃ for the inoculation plateau period) is mapped onto the current time axis as the instruction executed by the subsequent temperature control equipment.

[0090] The humidity value is processed by a support vector machine (SVR) model to obtain the regulating force. As a preferred embodiment of the present invention, the specific construction and training method of the SVR model is as follows: Multiple batches of historical fermentation data are collected, using the measured humidity value (%RH) of each batch of the pile as the input feature. The output label is the actual control intensity of the actuator required to maintain that humidity level (e.g., the pulse width modulation duty cycle of the humidifying spray pump, %). A training sample set of no less than 200 groups covering the humidity range of 30%RH to 90%RH was compiled; a Gaussian radial basis function was selected, and after previous cross-validation calibration, the preferred hyperparameter in this embodiment is the penalty factor. nuclear parameters Insensitive loss coefficient Using the aforementioned hyperparameters and all training samples, the regression function is solved using the sequence minimum optimization algorithm. The model is trained offline and stored in the system database. During online inference, the humidity value determined in the current batch is input into the model, and the model outputs the corresponding regulation force (i.e., the current required actuator control intensity).

[0091] Obtain the regulating force and temperature setpoints. Generate a control packet based on the regulating force and temperature setpoints. If the byte length of the control packet is less than a preset length threshold, send the control packet to the receiving end.

[0092] In this embodiment, the regulating force and the temperature setpoint determined simultaneously in step S107 (e.g., 62℃ for the high-temperature degradation period and 38℃ for the inoculation plateau period) are obtained and used as control parameters. These are then encapsulated into a standard data frame format according to a preset industrial fieldbus communication protocol (e.g., Modbus RTU) to generate a control packet. This control packet includes fields such as device address, function code, register start address, data bytes, and cyclic redundancy check (CRC) code. The byte length of the control packet is calculated and compared with a preset length threshold. The length threshold is determined by the maximum effective payload limit of a single frame in the fieldbus protocol; in this embodiment, a preferred example is 256 bytes. If the byte length of the current control packet is less than the threshold, the control packet is sent to the downstream actuator receiver via the communication interface to drive the humidification, ventilation, or heating equipment to perform control actions according to the target regulating force and temperature setpoint. If the control packet length exceeds the threshold, it is split into multiple sub-packets and sent sequentially until all instructions are transmitted.

[0093] Optionally, during the generation of the equipment's execution instruction set, the addition sequence parameters are set for different types of auxiliary materials. First, highly absorbent carbon-based materials (such as biochar, activated carbon, and waste wood chips) are added; second, nitrogen source materials that adjust the carbon-nitrogen ratio (such as livestock and poultry manure (pig manure / chicken manure), urea, and soybean meal) are added; and finally, structural auxiliary materials (such as crushed crop straw (corn / wheat straw), rice husks, and peanut shells) are added.

[0094] Regarding the timing of microbial agent injection, a thermotolerant microbial agent (e.g., thermophilic Bacillus steatophilus, thermophilic actinomycetes) is injected at the initial inoculation point. A room-temperature-efficient degrading microbial agent (e.g., white-rot fungi, Bacillus subtilis, yeast) is injected at the second inoculation point. Based on the time interval between the two injections, the optimal concentration ratio for synergistic effects of the microbial communities is calculated.

[0095] In one possible implementation, the system calculates the time interval between the initial vaccination and the second vaccination. Based on the pre-stored "time-viable cell count correspondence table" for thermophilic bacteria during the cooling phase (this table was determined by viable cell attenuation experiments at different laboratory temperatures), the residual surviving number of thermophilic bacteria during the second inoculation was obtained by referring to the table. Then, according to the preset optimal synergistic ratio of the two types of bacterial communities (in this embodiment, thermophilic bacteria: ambient bacteria = 1:3), with... Using this as a baseline, a reverse calculation was performed to obtain the target concentration of room-temperature bacteria required for the second inoculation. The optimal concentration ratio for synergistic effects of the microbial community was calculated. When encapsulating control messages, the temperature setpoint sequence was arranged chronologically to form temperature control curve data. The humidity regulation capability parameters of the excipients were matched with the target humidity range for each stage to generate a humidity control command sequence.

[0096] In one possible implementation, the system first divides the fermentation cycle into four stages: a heating period, a high-temperature period, a cooling period, and a constant-temperature aging period. A target humidity range is preset for each stage (e.g., 55%–60% for the high-temperature period and 45%–50% for the constant-temperature aging period). Then, the system reads the auxiliary material humidity regulation capability parameter (i.e., the material's own saturated water-holding limit, expressed as a percentage by mass) calculated in step S102. A segmented mapping method is used for matching: the target humidity range of each stage is compared with the auxiliary material humidity regulation capability parameter. If the upper limit of the target humidity for a stage is close to or reaches the water-holding limit (e.g., the difference is less than 5 percentage points), the control command generated for that stage is a continuous micro-spray command; if the lower limit of the target humidity for a stage is far below the water-holding limit (e.g., the difference is greater than 15 percentage points), the control command generated for that stage is an intermittent exhaust dehumidification command in conjunction with forced ventilation. All commands generated in each stage are arranged and combined in chronological order to form a complete humidity control command sequence.

[0097] Cyclic Redundancy Check (CRC-16) algorithm is used to verify the integrity of control messages to ensure the accuracy of data transmission. The specific verification steps are as follows: The sending end defines a 16-bit CRC register, with an initial value of 0xFFFF; the first byte of the control message is XORed with the lower 8 bits of the register; then the register is shifted right by one bit, and the least significant bit (LSB) shifted out is checked; if the shifted-out bit is 1, the register is XORed with a preset polynomial 0x8005; if the shifted-out bit is 0, no XOR operation is performed; the above right shift and XOR steps are repeated, and after processing all 8 bits of the current byte, the next byte is processed, until all bytes of the control message have been processed; the final value in the register is the checksum, which is appended to the end of the control message and sent to the receiving end; the receiving end uses the same CRC-16 algorithm to calculate the received message (excluding the checksum); if the result matches the received checksum, the message transmission is considered complete and error-free; otherwise, the message is considered erroneous and a retransmission is requested.

[0098] The verified control message is transmitted to the field controller of the solid waste treatment equipment via the industrial fieldbus protocol. Upon receiving the control message, the field controller parses the parameters within it. Based on the temperature setpoint sequence, it drives the heating and cooling devices to achieve precise temperature control. According to the dosing sequence parameters, it sequentially activates each auxiliary material dosing device, adding the auxiliary materials according to the preset dosage.

[0099] When the initial inoculation time is reached, the on-site controller triggers the microbial agent injection device to evenly spray the high-temperature resistant microbial agent onto the surface of the solid waste pile. When the second inoculation time is reached, the on-site controller triggers the microbial agent injection device again to inject a room-temperature high-efficiency degradation microbial agent.

[0100] Throughout the process, the on-site controller monitors key parameters such as temperature, humidity, and oxygen concentration inside the fermentation pile in real time. If the monitored data deviates from the set range, the control strategy is automatically adjusted based on the deviation to ensure that the treatment effect achieves the expected goal. As a preferred embodiment, the system collects pile temperature (thermal resistance, accuracy ±0.5℃), humidity (capacitive, accuracy ±2%RH), and oxygen concentration (electrochemical, accuracy ±1%) in real time at 5-minute intervals. Each fermentation stage (heating period, high temperature period, cooling period, and constant temperature aging period) has preset independent temperature, humidity, and oxygen concentration control ranges. The controller compares the measured values ​​with the upper and lower limits of the corresponding ranges. When the temperature deviation is ≥3℃, the fan frequency is adjusted; when the humidity deviation is ≥5%RH, the spraying is started or stopped or ventilation is adjusted; when the oxygen concentration deviation is ≥2%, the aeration duty cycle is changed. After adjustment, monitoring continues until the measured values ​​stabilize within the range for 30 minutes before determining that the stage is under control.

[0101] Through the above steps, precise control of the target solid waste treatment process is achieved. This method can optimize the ratio of auxiliary materials and the timing of microbial agent addition based on the volatile organic compound characteristics of solid waste, thereby improving the degradation efficiency of organic matter, shortening the treatment cycle, and reducing energy consumption. This achieves the technical effect of improving the solid waste treatment effect and solves the technical problems of low efficiency and high energy consumption in traditional solid waste treatment methods.

[0102] In practical applications, this method can be applied to various solid waste treatment scenarios, such as sludge treatment, kitchen waste treatment, and agricultural organic waste treatment. For different types of solid waste, by adjusting parameters such as the initial concentration of volatile organic compounds and the gas-phase diffusion coefficient, suitable auxiliary material ratio schemes and temperature control strategies can be generated.

[0103] For sludge-type solid waste with high moisture content, the proportion of carbon-based materials in the auxiliary material formulation is increased to improve adsorption capacity and aeration. For kitchen waste with high oil content, the temperature control strategy extends the high-temperature maintenance time to promote the full degradation of fatty substances.

[0104] This method can also be combined with IoT technology to achieve remote monitoring and intelligent scheduling of the solid waste treatment process. By installing sensors and communication modules on the solid waste treatment equipment, real-time monitoring data is uploaded to a cloud platform. Based on big data analysis results, the cloud platform dynamically optimizes the treatment parameters and sends the optimized control commands to the field controller, realizing intelligent management of the solid waste treatment process.

[0105] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. The scope of patent protection of the present invention shall be determined by the claims. Similarly, any equivalent structural changes made based on the description and drawings of the present invention shall also be included within the scope of protection of the present invention.

Claims

1. A method for predicting the performance and inverting process parameters of solid waste coupled with bio-organic fertilizer, characterized in that, The method includes: Obtain volatile organic compound characteristic data of target solid waste, calculate the required specific surface area for gas-solid contact, establish a mapping table between auxiliary material particle size and diffusion resistance, screen carbon-based materials, and extract the dynamic equilibrium coefficient of carbon-nitrogen ratio of auxiliary material combination. Based on the dynamic equilibrium coefficient of carbon-nitrogen ratio, an optimization algorithm is used to search for the initial solution of the excipient ratio that maximizes adsorption capacity. The initial solution is corrected according to the relationship between the proportion of carbon-based materials and the critical ratio threshold, and the humidity regulation capability of the excipients is adjusted to a set range to obtain the corrected excipient ratio scheme. The relationship between the number of adsorbent regeneration cycles and the degradation cycle was calculated based on the revised excipient ratio scheme, and the temperature control node sequence was extracted. The switching time of the temperature gradient between the high temperature maintenance interval and the cooling interval was marked and its timestamp was obtained as the temperature segmentation benchmark point to divide the temperature control interval. Candidate time windows for adding microbial agents are divided according to temperature segmentation benchmarks. Periods when the temperature exceeds the threshold are excluded. During the cooling phase, a stable temperature platform that meets the requirements for microbial colonization is identified, and the start time of this platform is taken as the initial inoculation time. The peak time of bacterial metabolic activity is estimated based on the initial inoculation time. If the interval between the peak time and the end of the cooling phase is insufficient, the second inoculation time is adjusted to the midpoint of the subsequent constant temperature phase when the temperature is safe and the duration is sufficient. Based on the secondary inoculation time and the revised excipient ratio scheme, a set of equipment execution instructions is generated, which includes the dosage, dosing sequence and injection timing. The temperature setpoint and humidity adjustment parameters are encapsulated into control messages and transmitted to the field controller.

2. The method according to claim 1, characterized in that, The process of acquiring volatile organic compound (VOC) characteristic data of the target solid waste, calculating the required gas-solid contact surface area, establishing a mapping table between auxiliary material particle size and diffusion resistance, screening carbon-based materials, and extracting the dynamic equilibrium coefficient of the carbon-nitrogen ratio of the auxiliary material combination includes: The initial concentration of VOCs in the target solid waste is used as the first concentration value, and the gas phase diffusion coefficient is used as the first diffusion rate. The first concentration value and the first diffusion rate are processed by the random forest algorithm to obtain the required gas-solid contact surface area. Based on the required specific surface area, match the corresponding particle size of the auxiliary material from the mapping table and read its diffusion resistance coefficient; The porosity of the excipient is obtained. If the porosity is greater than a preset standard, the excipient is used as the first combination item and combined with other excipients to form a combination scheme. The carbon-nitrogen ratio of each component in the combined scheme is determined, consumption rate weights are introduced, and cluster analysis is performed to determine the first degree of balance.

3. The method according to claim 1, characterized in that, The process involves using an optimization algorithm based on the carbon-nitrogen ratio dynamic equilibrium coefficient to search for an initial solution of the excipient ratio that maximizes adsorption capacity. This initial solution is then corrected according to the relationship between the proportion of carbon-based materials and the critical ratio threshold. The humidity regulation capability of the excipients is adjusted to a set range to obtain the corrected excipient ratio scheme, which includes: Using the carbon-nitrogen ratio dynamic equilibrium coefficient as input, the particle swarm optimization algorithm is used to search for the first excipient ratio solution that maximizes adsorption capacity; Extract the carbon-based material ratio from the first auxiliary material ratio solution. If the ratio is lower than the critical ratio threshold corresponding to the diffusion resistance coefficient, increase it to a multiple of the threshold to obtain the second carbon-based material ratio. The corresponding auxiliary material humidity adjustment capability parameter is determined based on the proportion of the second carbon-based material. If the parameter is not within the set range of the porosity saturation of the carbon-based material, it is adjusted to be within that range. The first auxiliary material ratio solution is updated based on the adjusted humidity regulation capability parameter and the proportion of the second carbon-based material to obtain the second auxiliary material ratio solution.

4. The method according to claim 1, characterized in that, The process involves calculating the correspondence between the number of adsorbent regeneration cycles and the degradation cycle based on the revised excipient ratio scheme, extracting the temperature control node sequence, marking the switching times of the temperature gradient between the high-temperature maintenance zone and the cooling zone, and obtaining their timestamps as temperature segmentation reference points to divide the temperature control zones. This includes: The adsorbent ratio is obtained according to the modified excipient ratio scheme. The corresponding number of regeneration cycles and degradation period are calculated using a linear interpolation algorithm, and a table of correspondence between the number of cycles and degradation period is generated. Extract temperature control points from the corresponding table and generate a temperature control node set; obtain high temperature value and cooling rate from the node set. Determine the intersection state of the high temperature value and the cooling rate. If an intersection occurs, mark the switching point and record the corresponding system timestamp. The system timestamps are classified and filtered using a decision tree algorithm, and target timestamps that meet the process constraints are extracted as temperature segmentation reference points. Using the aforementioned benchmark point as the boundary, the temperature control range of the degradation period is divided to obtain the first temperature control range and the second temperature control range.

5. The method according to claim 1, characterized in that, The process of dividing candidate time windows for inoculant addition based on temperature segmentation benchmarks, excluding periods when temperatures exceed thresholds, and identifying stable temperature platforms that meet the requirements for microbial colonization during the cooling phase, with the start time of this platform as the initial inoculation time, includes: The temperature segmentation reference point is used as the starting point to divide several candidate time windows, and the real-time temperature sequence corresponding to each candidate window is obtained. Candidate windows with temperatures higher than a preset heat shock threshold in the sequence are removed, and the remaining candidate windows constitute the cooling zone. For the temperature sequence in the cooling zone, the sliding window method is used to calculate the rate of temperature change, and a gradient value sequence is obtained. Identify the descending segment within the cooling zone based on the gradient value sequence, obtain the temperature fluctuation of the descending segment, calculate the buffer time for maintaining the temperature within a preset range, and if the buffer time is greater than the colonization period of the microbial community, then the descending segment is determined as a stable platform. The start time of the stable platform is extracted as the initial inoculation time node.

6. The method according to claim 1, characterized in that... The peak time of bacterial metabolic activity is calculated based on the initial inoculation time. If the interval between the peak time and the end of the cooling phase is insufficient, the second inoculation time is adjusted to the midpoint of a period in the subsequent isothermal phase where the temperature is safe and the duration is sufficient. This includes: The time of initial vaccination and historical metabolic data were obtained, and the time of peak metabolic activity was obtained by using an autoregressive integral moving average model for time series prediction. The difference between the peak arrival time and the preset cooling phase termination time is calculated as the interval period; If the interval is less than the preset minimum colonization time, then the temperature sequence data of the constant temperature maintenance phase after the cooling phase is extracted. The temperature sequence was fitted using a support vector regression model combined with a preset inactivation critical temperature of the inoculant to determine a safe constant temperature range below the inactivation critical point. Candidate segments with a duration longer than the preset survival rate curve rising period are selected from the safe constant temperature segment, and the midpoint between the start and end times of the candidate segment is taken as the adjusted second vaccination time node.

7. The method according to claim 1, characterized in that, The process involves generating a set of equipment execution instructions, including dosage, dosing sequence, and injection timing, based on the secondary inoculation time and the revised excipient ratio. The temperature setpoint and humidity adjustment parameters are then encapsulated into control messages and transmitted to the field controller. The vaccination time and ratio values ​​are obtained, and an initial instruction set is generated using the random forest algorithm. The dosage and order of administration are then extracted from the initial instruction set. If the dosage exceeds a preset threshold, the dosage sequence is rearranged in batches, and the injection sequence corresponding to the rearranged dosage sequence is obtained. The temperature setpoint is determined by combining the injection sequence and the initial instruction set. The corresponding humidity value is extracted based on the temperature setpoint. The humidity value is then processed using a support vector machine model to obtain the adjustment force. A control packet is generated based on the adjustment force and the temperature setpoint. If the byte length of the control packet is less than a preset threshold, the control packet is sent to the receiving end.

Images