An intelligent garbage pyrolysis oil system and its control method
Through intelligent garbage pyrolysis oiling system and advanced control algorithms, the problems of low heat transfer efficiency, catalyst carbon deactivation and insufficient energy utilization efficiency in garbage pyrolysis oiling system are solved, and a variety of garbage can be processed efficiently and produced The effect of high-quality oils.
Patent Information
- Application Number
- CN202411113724.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-14
AI Technical Summary
The existing garbage pyrolysis oiling system has problems such as low heat transfer efficiency, uneven product distribution, catalyst carbon deposition, low control accuracy and insufficient energy utilization efficiency, making it difficult to efficiently deal with a variety of garbage and produce high-quality oil products.
Adopt intelligent waste pyrolysis oiling system, including waste conveying unit, fluidized bed pyrolysis reactor, oil and gas separation unit, two-stage series catalytic cracking unit, distillation tower and generator set, combined with multi-scale time series decomposition prediction and dynamic Expert knowledge fusion and multi-objective optimization algorithms carry out intelligent control to achieve collaborative optimization of each unit.
It improves the pyrolysis efficiency, oil quality and catalyst life, enhances the system's adaptability and energy utilization efficiency, and ensures the stability of product quality and the overall efficiency of the system.
Smart Images

Figure CN118616463B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of garbage pyrolysis oil conversion system, and more specifically, to an intelligent garbage pyrolysis oil conversion system and a control method thereof. Background Art
[0002] The development of garbage pyrolysis oil technology has gone through several key stages. Early pyrolysis systems mainly used fixed bed or rotary kiln reactors, but these systems had problems such as low heat transfer efficiency and uneven product distribution. In order to improve these defects, researchers gradually turned their attention to fluidized bed pyrolysis technology. Fluidized bed reactors significantly improve pyrolysis efficiency and product uniformity by introducing fluidized media. However, it is still difficult to obtain high-quality oil products relying solely on the pyrolysis process. Pyrolysis oil usually has the characteristics of strong acidity, high oxygen content, and poor stability, which limits its direct application.
[0003] In order to improve the quality of oil products, some researchers began to introduce catalytic cracking technology into the pyrolysis system. The introduction of catalytic cracking has indeed improved the quality of oil products to a certain extent, but since most systems use single-stage catalytic cracking, the cracking effect on large molecules is not ideal and it is easy to cause carbon deposition and deactivation of the catalyst. This leads to the problem of frequent catalyst replacement in the long-term operation of the system, which greatly increases the operating cost.
[0004] In terms of control methods, traditional garbage pyrolysis oil systems mostly use PID control or simple fuzzy control. These control methods perform well when dealing with linear systems, but their adaptability and optimization capabilities are insufficient in the face of the high complexity and volatility of garbage components. Especially when dealing with different types of garbage, it is difficult for the system to achieve rapid adjustment and optimization, which affects the stability of product quality and the overall efficiency of the system.
[0005] In addition, the existing garbage pyrolysis oil system also has shortcomings in terms of energy utilization efficiency. Most systems fail to fully utilize the non-condensable gas produced during the pyrolysis process, resulting in energy waste. At the same time, due to the low control accuracy, the energy coordination between the various units of the system is not ideal, further reducing the overall energy utilization efficiency.
[0006] In the face of these challenges, it is particularly necessary to develop a garbage pyrolysis oil system that can efficiently process a variety of garbage, produce high-quality oil products, and has intelligent control capabilities. The present invention came into being in this context. Summary of the invention
[0007] The present invention is proposed in this context, and aims to address the limitations of the prior art and provide an intelligent garbage pyrolysis oil system and a control method thereof.
[0008] The present invention provides an intelligent garbage pyrolysis oil system, comprising:
[0009] The garbage conveying unit is used to pre-treat and transport the garbage; the fluidized bed pyrolysis reactor is used to receive the pre-treated garbage and perform pyrolysis to produce pyrolysis oil and non-condensable gas; the oil-gas separation unit is used to receive the pyrolysis products and separate them into pyrolysis oil and non-condensable gas; the two-stage series catalytic cracking unit is used to receive the separated pyrolysis oil and perform deep treatment to generate cracking products; the distillation tower is used to receive the cracking products and separate them into gasoline and diesel finished oil; the generator set is used to receive non-condensable gas and generate electricity; the control unit is used to collect system operation data, execute intelligent control algorithms, and coordinate the operation of each unit.
[0010] Specifically, the garbage conveying unit includes: a garbage receiving bucket equipped with a weighing sensor; a multi-stage crusher, including two stages of coarse crushing and fine crushing, which automatically adjusts the crushing force according to the type of garbage; a dryer using hot air circulation drying, equipped with a humidity sensor; a sealed conveyor belt equipped with a speed-adjustable variable frequency motor; wherein the garbage receiving bucket, crusher, dryer and conveyor belt are connected in sequence, and the leakage of pollutants is prevented by a sealed connection, and the conveyor belt is connected to the feed port of the fluidized bed pyrolysis reactor.
[0011] Specifically, the fluidized bed pyrolysis reactor includes: a feed port provided with an air lock device, connected to the conveyor belt; a gasification zone located at the bottom of the reactor, used to introduce an appropriate amount of air to achieve gasification of part of the garbage; a pyrolysis zone located in the upper middle part of the reactor, used to carry out pyrolysis reaction under anaerobic or oxygen-deficient conditions; a discharge port provided with a screw conveyor, used to continuously discharge pyrolysis residues; a temperature control system with multi-point temperature measurement and zone control, used to ensure accurate temperature distribution of 500-800°C; a fluidized medium circulation system including a cyclone separator and a conveying pipeline; wherein the pyrolysis product enters the oil-gas separation unit from the top outlet of the reactor, and the fluidized medium is collected by the cyclone separator and re-injected into the bottom of the reactor to form a closed-loop circulation.
[0012] Specifically, the oil-gas separation unit includes: a rapid condenser with a spiral tube bundle structure and a low-temperature ethylene glycol solution as a cooling medium; an oil-gas separator with a multi-stage centrifugal separation technology; an oil mist collector for capturing tiny oil droplets; and an oil storage tank provided with a level gauge and a temperature sensor; wherein the rapid condenser is connected to the top outlet of the fluidized bed pyrolysis reactor, the oil-gas separator, the oil mist collector and the oil storage tank are connected in sequence, the oil storage tank is connected to a two-stage series catalytic cracking unit, and the non-condensable gas outlet is connected to a generator set.
[0013] Specifically, the two-stage series catalytic cracking unit includes: a primary catalytic cracking reactor with a fixed bed design for preliminary cracking of macromolecules; a secondary catalytic cracking reactor with a fluidized bed design for deep cracking and product quality optimization; a catalyst regeneration system including a regeneration tower and an air blower; and a product rapid cooling system for preventing secondary reactions; wherein the primary and secondary catalytic cracking reactors are connected in series, the catalyst regeneration system is connected to the two reactors to form a catalyst circulation, and the product rapid cooling system is connected to the outlet of the secondary catalytic cracking reactor and to the feed port of the distillation tower.
[0014] Specifically, the distillation tower includes: a preheating section using heat from the bottom of the tower; a distillation section using high-efficiency separation fillers; a condensation section using graded condensation; a side-line extraction system for extracting intermediate distillate products; and a reflux system including a reflux tank and a reflux pump; wherein the preheating section is connected to the outlet of two-stage catalytic cracking units connected in series, the top product of the condensation section is gasoline, and the bottom product is diesel, and the side-line extraction system is used to extract the intermediate distillate products.
[0015] The control method of the intelligent garbage pyrolysis oil system includes the following steps: collecting system operation data; performing multi-scale time series decomposition and prediction on the operation data; dynamically fusing the prediction results with expert knowledge; performing multi-objective optimization based on the fusion results; generating and executing control instructions; and evaluating system performance and performing feedback optimization.
[0016] Specifically, the multi-scale time series decomposition and prediction step includes:
[0017] The preprocessed time series data is decomposed using the variational mode decomposition algorithm:
[0018]
[0019] in, is the original signal, For the The intrinsic mode functions, is the residual, To decompose the scale;
[0020] Construct a long short-term memory network for each intrinsic mode function for prediction:
[0021]
[0022] in, For the The predicted value of each mode, is the time step;
[0023] Adopt adaptive weighted reconstruction algorithm to fuse the prediction results of each modality:
[0024]
[0025] in, is the final predicted value, For the The adaptive weights of the modes, is the mean square error of the mode, For the The mean square error of the modes.
[0026] Specifically, the dynamic fusion step includes: constructing an expert knowledge base based on fuzzy rules;
[0027] Design a dynamic weight calculation network based on the attention mechanism:
[0028] Attention_Network
[0029] in, is the weight of expert knowledge, is the current system state vector;
[0030] Weighted fusion of neural network prediction results and expert knowledge:
[0031]
[0032] in, For the final control decision, For the neural network prediction results, Recommended value for expert knowledge.
[0033] Specifically, the multi-objective optimization step adopts an improved non-dominated sorting genetic algorithm NSGA-II, including:
[0034] Define the multi-objective optimization function:
[0035]
[0036] Maximize: Oil_quality
[0037] Minimize: Energy_consumption
[0038] Subject to:
[0039]
[0040]
[0041] in, To express the oil yield, Oil_quality Oil quality index, Energy_consumption is the energy consumption per unit of processing volume, is the vector of controlled variables, including pyrolysis temperature, residence time, and catalyst dosage; and are inequality and equality constraints respectively;
[0042] Execute the improved NSGA-II algorithm, including population initialization, non-dominated sorting, crowding calculation, selection, crossover and mutation operations;
[0043] Introducing adaptive crossover and mutation algorithms:
[0044]
[0045]
[0046] in, and are the crossover and mutation probabilities, is the fitness value of the individual with the highest fitness in the parent generation, is the fitness value of the current individual, and are the maximum and minimum fitness values in the population, respectively. and is the preset probability parameter.
[0047] The present invention has the following beneficial effects:
[0048] First, the use of a fluidized bed pyrolysis reactor greatly improves pyrolysis efficiency and solves the problem of uneven heat transfer in traditional fixed bed reactors. Second, the innovative design of two-stage catalytic cracking units in series not only improves oil quality, but also effectively extends the life of the catalyst. The first stage of catalytic cracking is mainly responsible for the initial cracking of macromolecules, while the second stage further optimizes product distribution. This design significantly increases the yield of high value-added products (such as gasoline).
[0049] In terms of control, the present invention introduces advanced multi-scale time series decomposition prediction, dynamic expert knowledge fusion and multi-objective optimization algorithms. This intelligent control system can accurately predict system changes, quickly respond to raw material fluctuations, and achieve coordinated optimization of each unit. This not only improves the adaptability and stability of the system, but also greatly improves energy efficiency.
[0050] Through the use of non-condensable gas by the generator set, this system forms a closed-loop energy utilization system, further improving the overall efficiency. The introduction of the distillation tower ensures the stability of the quality of the final product and meets the market demand for high-quality fuel.
[0051] In summary, the intelligent garbage pyrolysis oil system and control method of the present invention show significant advantages in improving oil yield and quality, increasing energy utilization efficiency, and enhancing system stability. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] Figure 1 It is the overall logic block diagram of the system of the present invention.
[0053] Figure 2 It is a logic block diagram of the garbage conveying unit of the present invention.
[0054] Figure 3 This is the logic block diagram of the fluidized bed pyrolysis reactor.
[0055] Figure 4 This is a logic block diagram of the oil and gas separation unit of the present invention.
[0056] Figure 5 This is a logic block diagram of the two-stage series catalytic cracking units of the present invention.
[0057] Figure 6 It is the logic block diagram of the distillation tower of the present invention.
[0058] Figure 7 It is the logic block diagram of the generator set of the present invention.
[0059] Figure 8 It is a logic block diagram of the control unit of the present invention.
[0060] Fig. 9 This is a logic block diagram of anomaly detection and processing of the present invention. DETAILED DESCRIPTION
[0061] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects thereof are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0062] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0063] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0064] like Figure 1 As shown, the intelligent garbage pyrolysis oil system provided by the present invention mainly includes a garbage conveying unit 1, a fluidized bed pyrolysis reactor 2, an oil-gas separation unit 3, a two-stage series catalytic cracking unit 4, a distillation tower 5, a generator set 6 and a control unit 7. These units work together through an intelligent control algorithm to achieve efficient pyrolysis oil treatment of garbage.
[0065] First, let us understand in detail the structure and working principle of the garbage conveying unit 1. This unit is the starting point of the entire system and is responsible for the initial processing and transportation of garbage. Figure 2 As shown, the garbage conveying unit 1 includes a garbage receiving hopper 11 , a crusher 12 , a dryer 13 and a conveyor belt 14 .
[0066] Preferably, in one embodiment of the present invention, the garbage receiving bucket 11 adopts a large-capacity, corrosion-resistant design, with a volume of up to 50-100 cubic meters, and the inner wall adopts a special acid- and alkali-resistant coating. In order to monitor the weight of the garbage in real time, a high-precision weighing sensor is installed at the bottom of the garbage receiving bucket 11, with an accuracy of up to ±0.1%. Such a design can not only accurately grasp the feed amount, but also provide important parameters for subsequent processing.
[0067] Next, the garbage is processed by crusher 12. Crusher 12 adopts a multi-stage crushing design, including two stages of coarse crushing and fine crushing. A jaw crusher is used in the coarse crushing stage to reduce the diameter of garbage blocks from about 500mm to less than 100mm; a hammer crusher is used in the fine crushing stage to further crush the garbage to less than 20mm. It is worth mentioning that crusher 12 is equipped with an intelligent control system, which can automatically adjust the crushing force according to the type of garbage, ensuring the crushing effect while also extending the life of the equipment.
[0068] The crushed garbage enters the dryer 13 for drying. The dryer 13 adopts hot air circulation drying technology, and the heat source comes from the system waste heat recovery. There are multiple layers of drying grids inside the dryer, and the garbage is turned on it and fully exposed to the hot air. At the same time, the dryer 13 is equipped with a humidity sensor to monitor the moisture content of the garbage in real time and adjust the drying intensity and time according to the monitoring results. Normally, the drying process will reduce the moisture content of the garbage from 40-60% to below 15%, preparing for the subsequent pyrolysis process.
[0069] Finally, the dried garbage is transported to the feed port of the fluidized bed pyrolysis reactor 2 through the conveyor belt 14. The conveyor belt 14 adopts a sealed design to effectively prevent dust leakage. In addition, the conveyor belt 14 is equipped with a variable frequency motor with adjustable speed, which can flexibly adjust the feeding speed according to system requirements, usually between 0.5-2 m / s.
[0070] It is worth noting that the various devices of the garbage conveying unit 1 are connected through a sealed connection, which effectively prevents the leakage of pollutants and ensures the cleanliness of the operating environment.
[0071] Next, we will introduce the structure and working principle of the fluidized bed pyrolysis reactor 2 in detail. Figure 3 As shown, the fluidized bed pyrolysis reactor 2 adopts an advanced fluidized bed design to achieve efficient pyrolysis of garbage. It mainly includes a feed inlet 21, a gasification zone 22, a pyrolysis zone 23, a discharge port 24, a temperature control system 25 and a fluidized medium circulation system 26.
[0072] Preferably, in one embodiment of the present invention, the feed port 21 is provided with an air lock device to effectively prevent the pyrolysis gas from flowing back. The gasification zone 22 is located at the bottom of the reactor, and a proper amount of air is introduced to achieve gasification of part of the garbage and provide heat for the pyrolysis process. Normally, the temperature of the gasification zone 22 is controlled between 700-800°C, and the air volume is 20-30% of the theoretical combustion air volume.
[0073] The pyrolysis zone 23 is located in the upper middle part of the reactor and performs pyrolysis reaction under oxygen-free or oxygen-deficient conditions. The temperature of the pyrolysis zone 23 is usually controlled between 500-700°C and the residence time is 15-30 minutes. Such a combination of temperature and residence time can maximize the yield of pyrolysis oil while ensuring the quality of pyrolysis oil.
[0074] The discharge port 24 is provided with a screw conveyor for continuously discharging the pyrolysis residue. The temperature control system 25 adopts a multi-point temperature measurement and partition control strategy to ensure the accurate temperature distribution of 500-800°C in the reactor. Usually 10-15 temperature measurement points are set in the reactor, and the temperature control accuracy of each temperature measurement point can reach ±5°C.
[0075] The fluidized medium circulation system 26 includes a cyclone separator and a conveying pipeline to realize the circulation of the fluidized medium. The typical fluidized medium is quartz sand with a particle size of 0.2-0.5 mm, and the circulation amount can reach 50-100 kg / m³ reactor volume.
[0076] In actual operation, the garbage enters the feed port 21 through the conveyor belt 14, undergoes preliminary pyrolysis in the gasification zone 22, and then rises to the pyrolysis zone 23 to complete deep pyrolysis. The pyrolysis products enter the oil-gas separation unit 3 from the top outlet of the reactor, and the residue is discharged from the discharge port 24. The fluidized medium is collected by the cyclone separator and re-injected into the bottom of the reactor to form a closed loop.
[0077] like Figure 4As shown, after the pyrolysis product enters the oil-gas separation unit 3, it is first cooled by the rapid condenser 31. The rapid condenser 31 adopts a spiral tube bundle structure, and the cooling medium is a low-temperature ethylene glycol solution of -5 to 5°C. The cooled mixture enters the oil-gas separator 32, which adopts multi-stage centrifugal separation technology, and the separation efficiency can reach more than 98%. In order to further improve the oil recovery rate, the system is also equipped with an oil mist collector 33, which can capture oil droplets as small as 1μm in diameter. Finally, the separated pyrolysis oil is stored in the oil storage tank 34, waiting to enter the catalytic cracking unit 4. The non-condensable gas is transported to the generator set 6.
[0078] The two-stage series catalytic cracking unit 4 is another innovative feature of this system. Figure 5 As shown, it includes a primary catalytic cracking reactor 41, a secondary catalytic cracking reactor 42, a catalyst regeneration system 43 and a product rapid cooling system 44.
[0079] The primary catalytic cracking reactor 41 adopts a fixed bed design and is mainly used for the initial cracking of large molecules. The reaction temperature is usually controlled at 450-500°C and the space velocity (WHSV) is 2-4h -1 The secondary catalytic cracking reactor 42 adopts a fluidized bed design for deep cracking and product quality optimization. Its reaction temperature is slightly higher, usually between 500-550°C, and the space velocity can reach 5-8h -1 Both reactors use modified HZSM-5 molecular sieve catalysts with a silicon-aluminum ratio of 50-100 and a pore size of 0.54-0.56nm. This catalyst has excellent selectivity and stability and can effectively increase the yield of gasoline and diesel.
[0080] The catalyst regeneration system 43 includes a regeneration tower and an air blower to achieve continuous regeneration of the catalyst. The regeneration temperature is usually controlled at 650-700°C to ensure complete combustion of carbon deposits. The activity of the regenerated catalyst can be restored to more than 95% of the original activity. The product rapid cooling system 44 adopts rapid cooling technology to reduce the temperature of the cracking product from 500°C to below 200°C within 0.1 seconds, effectively preventing secondary reactions and ensuring product quality.
[0081] The distillation tower 5 is the last key unit of the system, responsible for separating the cracking products into various finished oil products. Figure 6 As shown, the distillation tower 5 includes a preheating section 51 , a distillation section 52 , a condensation section 53 , a side extraction system 54 and a reflux system 55 .
[0082] The preheating section 51 uses the heat from the bottom of the tower to preheat the feed, thereby improving energy efficiency. The distillation section 52 uses high-efficiency separation fillers, the filler specific surface area can reach 250-350m² / m³, and the number of theoretical plates can reach 30-40 layers. The condensation section 53 uses graded condensation technology to obtain products with different boiling point ranges. The side extraction system 54 is used to extract intermediate distillate products such as kerosene and diesel. The reflux system 55 includes a reflux tank and a reflux pump, and the product purity is optimized by adjusting the reflux ratio (usually between 1.2-1.8).
[0083] In actual operation, the cracking products are fed from the middle of the tower and enter the distillation section 52 after preheating. The light components rise to the top of the tower, and part of them are refluxed after condensation in the condenser, and part of them are collected as gasoline products (C5-C12, boiling point 30-200°C). The heavy components fall to the bottom of the tower, and part of them are refluxed after heating in the reboiler, and part of them are collected as diesel products (C13-C20, boiling point 200-350°C). The middle fraction is collected through the side extraction system 54.
[0084] like Figure 7 As shown, the generator set 6 is mainly used to generate electricity using non-condensable gas to improve the energy utilization efficiency of the system. It includes a gas purification system, a gas turbine and a generator. The gas purification system first removes pollutants such as sulfides and dust from the non-condensable gas, and the calorific value of the purified gas is usually 15-20MJ / Nm³. The power generation efficiency of the gas turbine can reach 35-40%, and part of the generated electricity is used for the operation of the system itself, and the rest is transmitted to the power grid.
[0085] The control unit 7 is the brain of the whole system and is responsible for coordinating the operation of each unit. It uses advanced algorithms such as multi-scale time series decomposition prediction, dynamic expert knowledge fusion and multi-objective optimization to realize the intelligent operation of the system.
[0086] like Figure 8-9 As shown, below, we will introduce the specific steps of the control method in detail:
[0087] Step S1: Collect system operation data.
[0088] The control unit 7 collects key parameters such as temperature, pressure, flow, composition, etc. in real time through the sensor network distributed in each unit. The sampling frequency is usually 1-10Hz to ensure the timeliness and continuity of the data.
[0089] Step S2: Multi-scale time series decomposition and prediction.
[0090] The collected operation data is decomposed and predicted in multi-scale time series. This step uses a combination of the variational mode decomposition (VMD) algorithm and the long short-term memory network (LSTM).
[0091] First, the VMD algorithm is used to decompose the preprocessed time series data:
[0092]
[0093] in, is the original signal, For the The intrinsic mode functions, is the residual, is the decomposition scale. Usually, The value is selected between 3 and 5, which can effectively capture the multi-scale characteristics of the signal.
[0094] Then, an LSTM network is constructed for each intrinsic mode function for prediction:
[0095]
[0096] in, For the The predicted value of each mode, is the time step. The number of hidden layer nodes in the LSTM network is usually set to 50-100, and the time step is The range is chosen to be 24-48 to capture diurnal cyclical variations.
[0097] Finally, the adaptive weighted reconstruction algorithm is used to fuse the prediction results of each modality:
[0098]
[0099]
[0100] in, is the final predicted value, For the The adaptive weights of the modes, To adjust the parameters, is the mean square error of the mode, For the The mean square error of the modes. The value is usually between 0.1 and 1 and can be adjusted according to actual conditions.
[0101] Step S3: Dynamic expert knowledge fusion.
[0102] Dynamically integrate the prediction results with expert knowledge. This step first builds an expert knowledge base based on fuzzy rules, and then designs a dynamic weight calculation network based on the attention mechanism:
[0103] Attention_Network
[0104] in, is the weight of expert knowledge, is the current system state vector. The attention network usually adopts a multi-layer perceptron (MLP) structure, where the number of nodes in the input layer is the same as the dimension of the state vector, the number of nodes in the hidden layer is 1.5-2 times that of the input layer, and the output layer is a single node.
[0105] Finally, the neural network prediction results and expert knowledge are weighted and fused:
[0106]
[0107] in, For the final control decision, For the neural network prediction results, Recommended value for expert knowledge.
[0108] Step S4: multi-objective optimization.
[0109] Multi-objective optimization is performed based on the fusion results. This step uses the improved non-dominated sorting genetic algorithm NSGA-II, which includes the following sub-steps:
[0110] a) Define the multi-objective optimization function:
[0111]
[0112] Maximize: Oil_quality
[0113] Minimize: Energy_consumption
[0114] Subject to:
[0115]
[0116]
[0117] in, To express the oil yield, Oil_quality Oil quality index, Energy_consumption is the energy consumption per unit of processing volume, is the vector of controlled variables, including pyrolysis temperature, residence time, and catalyst dosage; and are inequality and equality constraints respectively;
[0118] Typical constraints include:
[0119] C Pyrolysis temperature C
[0120] min Dwell time min
[0121] kg / h Catalyst dosage kg / h
[0122] Total material balance = 0
[0123] b) Execute the improved NSGA-II algorithm, including population initialization, non-dominated sorting, crowding calculation, selection, crossover and mutation operations. The population size is usually set to 50-100, and the number of evolution generations is 100-200 generations.
[0124] c) Introducing adaptive crossover and mutation algorithms:
[0125]
[0126]
[0127] in, and are the crossover and mutation probabilities, is the fitness value of the individual with the highest fitness in the parent generation, is the fitness value of the current individual, and are the maximum and minimum fitness values in the population, respectively. and is the preset probability parameter, usually taken as: , , , .
[0128] Based on the multi-objective optimization results, the optimal control instruction set is generated:
[0129]
[0130] in, For the Then, the control instructions are converted into specific execution parameters:
[0131] Waste conveying unit: adjust the feed speed (0.5-2 m / s), crushing degree (final particle size 10-30 mm) and drying intensity (target moisture content 10-15%).
[0132] Pyrolysis reactor: precise control of temperature distribution (700-800°C in gasification zone, 500-700°C in pyrolysis zone) and residence time (15-30 minutes).
[0133] Catalytic cracking unit: adjust the catalyst dosage (WHSV2-8h -1 and reaction temperature (450-500°C for primary stage and 500-550°C for secondary stage).
[0134] Distillation tower: Optimize the top temperature (60-80°C) and reflux ratio (1.2-1.8).
[0135] Finally, the control instructions are sent to the controllers of each unit through the actuator.
[0136] Step S6: Evaluate system performance and perform feedback optimization.
[0137] Regularly collect system operation data, including pyrolysis oil yield, quality indicators, energy consumption, etc. Calculate system comprehensive performance indicators: PI Norm(Oil_yield Norm(Oil_quality Norm(Energy_consumption .
[0138] Among them, Norm() is the normalization function, is the weight of each indicator. .
[0139] if (usually set to 0.8), the current operating parameters are added to the expert knowledge base. At the same time, the latest data is used to perform incremental learning on the neural network model to continuously improve the prediction accuracy.
[0140] Preferably, in one embodiment of the present invention, an abnormality detection and processing step is also included. This step uses the isolation forest algorithm to detect system abnormalities in real time:
[0141]
[0142] in, For sample The average path length, is the normalization factor. When anomaly is detected (Anomaly_score>0.6), the emergency control strategy is triggered: reduce the feed rate to 50-70% of the normal rate;
[0143] Adjust the pyrolysis temperature to a safe range (550-650°C); increase the catalyst dosage by 20-30% to stabilize the reaction.
[0144] The intelligent garbage pyrolysis oil system and control method of the present invention realizes efficient pyrolysis oil treatment of garbage through the close cooperation of garbage conveying unit, fluidized bed pyrolysis reactor, oil and gas separation unit, two-stage series catalytic cracking unit, distillation tower and generator set. At the same time, the control method adopts advanced algorithms such as multi-scale time series decomposition prediction, dynamic expert knowledge fusion and multi-objective optimization, which greatly improves the intelligence level and operation efficiency of the system.
[0145] The system can not only process urban domestic waste, but also be applied to industrial waste, medical waste and other fields. For example, for industrial waste with high oil content, the oil yield and quality can be improved by adjusting the pyrolysis temperature (increased to 750-800℃) and the catalyst formula (increasing the rare earth modification ratio). For medical waste, thorough sterilization can be ensured by increasing the pyrolysis temperature (800-850℃) and extending the residence time (30-40 minutes).
[0146] In summary, the intelligent garbage pyrolysis oil system and control method provided by the present invention not only solve the problems of low efficiency and high pollution of traditional garbage treatment methods, but also realize the resource utilization of garbage, with significant economic and environmental benefits. Through the application of intelligent control algorithms, the system can adapt to the treatment needs of different types of garbage, and has broad application prospects and promotion value.
[0147] In order to verify the superiority of the intelligent garbage pyrolysis oil system and its control method of the present invention, we conducted a series of comparative experiments. The specific conditions of the embodiments and comparative examples, as well as the corresponding test results and analysis will be introduced in detail below.
[0148] Example 1: Intelligent garbage pyrolysis oil system and control method thereof using the present invention
[0149] In this embodiment, we used a complete intelligent garbage pyrolysis oil system, including a garbage conveying unit, a fluidized bed pyrolysis reactor, an oil-gas separation unit, a two-stage series catalytic cracking unit, a distillation tower and a generator set. The control method uses multi-scale time series decomposition prediction, dynamic expert knowledge fusion and multi-objective optimization algorithm.
[0150] The raw material is typical urban domestic waste, with a water content of about 40% and an organic matter content of about 60%. The system operating parameters are as follows: pyrolysis temperature: 650°C; pyrolysis residence time: 25 minutes; catalytic cracking temperature: 480°C for the first stage and 520°C for the second stage; catalyst dosage: WHSV4h - ¹.
[0151] Comparative Example 1: Traditional single-stage pyrolysis system, which only includes a simple pyrolysis reactor and an oil-gas separation unit, without catalytic cracking and distillation processes. The control method adopts traditional PID control.
[0152] Operation parameters: pyrolysis temperature: 600°C; pyrolysis residence time: 30 minutes.
[0153] Comparative Example 2: Single-stage catalytic cracking system, in which a single-stage catalytic cracking unit is added after pyrolysis, but a two-stage series design is not adopted. The control method adopts fuzzy control.
[0154] Operation parameters: pyrolysis temperature: 620℃; pyrolysis residence time: 28 minutes; catalytic cracking temperature: 500℃; catalyst dosage: WHSV3h -1 .
[0155] Comparative Example 3: A two-stage catalytic cracking system using a conventional control method. The hardware configuration of the system is the same as that of the present invention, but the control method adopts conventional PID control and expert system.
[0156] Operation parameters: pyrolysis temperature: 640℃; pyrolysis residence time: 26 minutes; catalytic cracking temperature: 470℃ for the first stage, 510℃ for the second stage; catalyst dosage: WHSV4h -1 .
[0157] We conducted a 30-day continuous operation test on the above four solutions. The main test indicators include oil yield, oil quality, energy efficiency and system stability. The test methods and standards are as follows:
[0158] 1. Oil yield: measured by weight method according to GB / T28731-2012 standard.
[0159] 2. Oil quality: Determined by gas chromatography-mass spectrometry (GC-MS) in accordance with SH / T0606-2005 standard.
[0160] 3.Energy utilization efficiency: calculated according to GB / T2589-2008 standard.
[0161] 4. System stability: standard deviation of key parameters (such as temperature, pressure, and flow) during 30 consecutive days of operation.
[0162] The test results are shown in the following table:
[0163]
[0164] It can be seen from the test results that Example 1 of the present invention shows obvious advantages in various indicators:
[0165] First, in terms of oil yield, Example 1 reached 42.5%, which is much higher than other comparative examples. This is mainly due to the design of two-stage series catalytic cracking units, which enables more complete conversion of pyrolysis oil. In particular, the significant increase in gasoline yield (25.3%) reflects the advantages of this system in producing high value-added products.
[0166] Secondly, the oil quality index has also been significantly improved. The octane number (RON) of gasoline has reached 92.5, and the cetane number of diesel has reached 52.3, both of which are better than other solutions. This shows that the system of the present invention not only increases product output, but also significantly improves product quality.
[0167] Furthermore, the energy efficiency reached 78.6%, which is 3.3 percentage points higher than the closest comparative example 3. This reflects the advantages of this system in energy recovery and recycling, and provides strong support for achieving low-carbon environmental protection goals.
[0168] Finally, the system stability index also performed well. The temperature fluctuation was controlled within the range of ±2.5℃, and the yield fluctuation was only ±1.8%. This is due to the advanced control algorithm adopted by the present invention, especially the multi-scale time series decomposition prediction and dynamic expert knowledge fusion technology, which enables the system to better cope with raw material fluctuations and external interference.
[0169] It is worth noting that although Comparative Example 3 uses similar hardware configuration, due to the lack of advanced control algorithm support, its various indicators are slightly inferior to those of Example 1. This further proves the importance and effectiveness of the control method of the present invention.
[0170] In summary, the intelligent garbage pyrolysis oil system and its control method of the present invention have shown obvious advantages in improving the yield and quality of oil products, increasing energy utilization efficiency, and enhancing system stability. This not only provides an efficient and environmentally friendly solution for urban domestic waste treatment, but also opens up a new way for the resource utilization of other types of waste. With the expansion of raw material types and the expansion of processing scale, the advantages of the present invention will become more prominent and have broad application prospects.
[0171] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, replacement, and improvement made within the principle of the present invention should be included in the protection scope of the present invention.
Claims
1. An intelligent garbage pyrolysis oil system, It is characterized in that include: A garbage conveying unit, used for pre-processing and conveying garbage; A fluidized bed pyrolysis reactor is used to receive the pre-treated garbage and perform pyrolysis to produce pyrolysis oil and non-condensable gas; The oil-gas separation unit is used to receive the pyrolysis products and separate them into pyrolysis oil and non-condensable gas; the two-stage catalytic cracking units are used to receive the separated pyrolysis oil and perform deep processing to generate cracking products; the distillation tower is used to receive the cracking products and separate them into gasoline and diesel finished oil; The generator set is used to receive non-condensable gas and generate electricity; the control unit is used to collect system operation data, execute intelligent control algorithms, and coordinate the operation of each unit; The fluidized bed pyrolysis reactor comprises: a feed port provided with an air lock device, connected to a conveyor belt; a gasification zone located at the bottom of the reactor, used to introduce an appropriate amount of air to achieve gasification of part of the garbage; a pyrolysis zone located in the middle and upper part of the reactor, used to carry out pyrolysis reaction under oxygen-free or oxygen-deficient conditions; a discharge port provided with a screw conveyor, used to continuously discharge pyrolysis residues; a temperature control system with multi-point temperature measurement and zone control, used to ensure accurate temperature distribution of 500-800°C; a fluidized medium circulation system including a cyclone separator and a conveying pipeline; wherein the pyrolysis product enters the oil-gas separation unit from the outlet at the top of the reactor, and the fluidized medium is collected by the cyclone separator and then re-injected into the bottom of the reactor to form a closed-loop circulation; The two-stage series catalytic cracking unit comprises: a primary catalytic cracking reactor of fixed bed design for preliminary cracking of macromolecules; a secondary catalytic cracking reactor of fluidized bed design for deep cracking and product quality optimization; a catalyst regeneration system including a regeneration tower and an air blower; and a product rapid cooling system for preventing secondary reactions; wherein the primary and secondary catalytic cracking reactors are connected in series, the catalyst regeneration system is connected to the two reactors to form a catalyst circulation, and the product rapid cooling system is connected to the outlet of the secondary catalytic cracking reactor and to the feed inlet of the distillation tower; The distillation tower comprises: a preheating section using the heat from the bottom of the tower; a distillation section using a high-efficiency separation filler; a condensation section using graded condensation; a side extraction system for extracting the intermediate distillate product; and a reflux system including a reflux tank and a reflux pump; wherein the preheating section is connected to the outlet of the two-stage catalytic cracking unit connected in series, the top product of the condensation section is gasoline, the bottom product is diesel, and the side extraction system is used to extract the intermediate distillate product; The control method of the intelligent garbage pyrolysis oil system comprises the following steps: collecting system operation data; performing multi-scale time series decomposition and prediction on the operation data; dynamically fusing the prediction results with expert knowledge; performing multi-objective optimization based on the fusion results; generating and executing control instructions; evaluating system performance and performing feedback optimization; The multi-scale time series decomposition and prediction step comprises: The preprocessed time series data is decomposed using the variational mode decomposition algorithm: ; in, is the original signal, For the The intrinsic mode functions, is the residual, To decompose the scale; Construct a long short-term memory network for each intrinsic mode function for prediction: ; in, For the The predicted value of each mode, is the time step; Adopt adaptive weighted reconstruction algorithm to fuse the prediction results of each modality: ; in, is the final predicted value, For the The adaptive weights of the modes, is the mean square error of the mode; The dynamic fusion step includes: constructing an expert knowledge base based on fuzzy rules; Design a dynamic weight calculation network based on the attention mechanism: ; in, is the weight of expert knowledge, is the current system state vector; Weighted fusion of neural network prediction results and expert knowledge: ; in, For the final control decision, For the neural network prediction results, Recommended values for expert knowledge; The multi-objective optimization step adopts an improved non-dominated sorting genetic algorithm NSGA-II, including: Define the multi-objective optimization function: ; in, To control the variable vector, including pyrolysis temperature, residence time, catalyst dosage, etc.; and are inequality and equality constraints respectively; Execute the improved NSGA-II algorithm, including population initialization, non-dominated sorting, crowding calculation, selection, crossover and mutation operations; Introducing adaptive crossover and mutation algorithms: ; in, and are the crossover and mutation probabilities, is the fitness value of the better individual in the parent generation, is the fitness value of the current individual, and are the maximum and minimum fitness values in the population, respectively. and is the preset probability parameter.
2. According to the intelligent garbage pyrolysis oil system of claim 1, It is characterized in that The garbage conveying unit includes: a garbage receiving bucket equipped with a weighing sensor; a multi-stage crusher, including two stages of coarse crushing and fine crushing, which automatically adjusts the crushing force according to the type of garbage; a dryer using hot air circulation drying, equipped with a humidity sensor; a sealed conveyor belt, equipped with a speed-adjustable variable frequency motor; wherein the garbage receiving bucket, crusher, dryer and conveyor belt are connected in sequence, and the leakage of pollutants is prevented by a sealed connection, and the conveyor belt is connected to the feed port of the fluidized bed pyrolysis reactor.
3. The intelligent garbage pyrolysis oil system according to claim 1, It is characterized in that The oil-gas separation unit comprises: a rapid condenser adopting a spiral tube bundle structure and using a low-temperature ethylene glycol solution as a cooling medium; an oil-gas separator adopting a multi-stage centrifugal separation technology; an oil mist collector for capturing tiny oil droplets; and an oil storage tank provided with a level gauge and a temperature sensor; wherein the rapid condenser is connected to the top outlet of the fluidized bed pyrolysis reactor, the oil-gas separator, the oil mist collector and the oil storage tank are connected in sequence, the oil storage tank is connected to a two-stage series catalytic cracking unit, and the non-condensable gas outlet is connected to a generator set.
