Multi-grade waste heat grading recovery device for poultry processing plant

Abstract

The invention provides a multi-grade waste heat grading recovery device for a poultry processing factory, and solves the technical problems that existing waste heat recovery equipment is mainly designed for single-grade and low-impurity-content waste heat resources and is difficult to adapt to feather drying waste gas with high impurities and exhaust temperature fluctuation of a refrigeration house compressor in a poultry processing scene. The device can be widely applied to the field of waste heat recovery equipment. The system specifically comprises a feather drying waste gas treatment system, a refrigeration house waste heat pressure stabilizing system and a comprehensive control system, wherein the feather drying waste gas treatment system comprises a cyclone separator and an electrostatic degreaser; a gas inlet pipe of the cyclone separator is communicated with a waste gas discharge pipe, waste gas enters from the tangential direction of the cylindrical section of the separator, and solid impurities are thrown to the wall of the separator and fall to a bottom ash hopper along the wall surface; the waste gas subjected to dust removal enters an electrostatic degreaser, a high-voltage electrode and a grounding polar plate are arranged in the degreaser, grease particles in the waste gas are ionized under the action of a high-voltage electric field, and the grease particles with negative electricity are adsorbed to the grounding polar plate.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of waste heat recovery equipment, and particularly relates to a multi-grade waste heat grading recovery device for a poultry processing plant. BACKGROUND

[0002] As an important branch of the food industry, the poultry processing industry has a complex production process and high energy consumption, covering multiple high-energy consumption links such as feather drying, meat refrigeration, and cooking processing. According to the China Food Industry Energy Consumption Report, the annual comprehensive energy consumption of medium and large poultry processing plants in China is about 1200-2500 tons of standard coal, of which the energy consumption of the feather drying process accounts for 28%-35% of the total energy consumption, and the energy consumption of the refrigeration system of the cold storage accounts for 22%-28% of the total energy consumption. Under the background of the current "double carbon" target and the continuous rise in energy prices, poultry processing enterprises are facing severe pressure to save energy and reduce costs.

[0003] As one of the core technical paths for reducing industrial energy consumption, the application of waste heat recovery in the poultry processing industry has significant bottlenecks. Existing waste heat recovery equipment is mostly designed for single-grade, low-impurity-content waste heat resources, and is difficult to adapt to the special conditions of high-impurity feather drying waste gas and large temperature fluctuations of cold storage compressor exhaust gas in the poultry processing scene, resulting in a waste heat recovery efficiency generally lower than 50%. A large amount of available waste heat is directly discharged, not only causing energy waste, but also exacerbating environmental burden. Therefore, developing multi-grade waste heat grading recovery technology that adapts to the special conditions of poultry processing has become a key requirement for the industry to achieve energy-saving transformation. SUMMARY

[0004] The purpose of the present application is to solve the above technical problems, and to provide a multi-grade waste heat grading recovery device for a poultry processing plant.

[0005] To this end, the present application provides a multi-grade waste heat grading recovery device for a poultry processing plant, which comprises a feather drying waste gas treatment system, a cold storage waste heat pressure stabilizing system, and a comprehensive control system. The feather drying waste gas treatment system comprises a cyclone separator and an electrostatic oil remover. The inlet pipe of the cyclone separator is in communication with the waste gas discharge pipe. The waste gas enters the cyclone separator from the tangent direction of the cylindrical section. Under the action of centrifugal force, the solid impurities are thrown to the wall and fall to the bottom hopper along the wall. The dust-removed waste gas enters the electrostatic oil remover. The electrostatic oil remover is provided with a high-voltage electrode and a grounding electrode plate. The oil particles in the waste gas are ionized under the action of the high-voltage electric field, and the negatively charged oil particles are adsorbed to the grounding electrode plate.

[0006] Further, the feather drying waste gas treatment system further comprises a heat pipe heat exchanger. The heat pipe evaporation section of the heat pipe heat exchanger exchanges heat with the dust-removed and oil-removed waste gas. The condensation section exchanges heat with cold water. The flow channel width of the heat exchanger is 15-20mm.

[0007] Further, the cold storage waste heat stabilizing system comprises a buffer heat storage tank, the volume of the buffer heat storage tank is 1.2-1.5 times of the exhaust volume of the cold storage compressor per hour, temperature sensors and liquid level sensors are arranged in the tank, and the tank is filled with high-temperature phase change heat storage materials; a proportional regulating valve is further arranged on the buffer heat storage tank, and is installed at the intersection of the water outlet of the buffer heat storage tank and the cold water supplement pipeline, and the valve opening degree can be continuously adjusted according to the deviation between the target water temperature and the actual water temperature.

[0008] The application provides an external circulation constant temperature system applied to a slaughtering plant scalding pool, and has the following beneficial effects:

[0009] Through the three-stage treatment structure of cyclone dust removal + electrostatic oil removal + wide flow channel heat pipe heat exchange, and the precise regulation and control of cold water flow and electrode voltage by an intelligent optimization algorithm, the waste gas heat recovery efficiency is stably maintained at more than 75%. Taking a processing plant with an average daily processing capacity of 100,000 poultry as an example, about 5.2 tons of standard coal can be recovered from the daily waste heat of feather drying, and about 1898 tons of standard coal can be saved annually.

[0010] Based on the stabilizing system of the buffer heat storage tank and the proportional regulating valve, and cooperating with the fuzzy PID control, the cold storage waste heat recovery efficiency is improved to more than 65%. The above processing plant can additionally recover about 3.3 tons of standard coal from the cold storage waste heat per day, and about 1194.5 tons of standard coal can be saved annually. DETAILED DESCRIPTION

[0011] The application will be further described below in combination with specific embodiments to help understand the content of the application. The methods used in the application are conventional methods unless otherwise specified; the raw materials and devices used are conventional commercially available products unless otherwise specified.

[0012] The application provides a multi-grade waste heat grading recovery device for a poultry processing plant, which comprises a feather drying waste gas treatment system, a cold storage waste heat stabilizing system and a comprehensive control system. The feather drying waste gas treatment system comprises a cyclone separator, an electrostatic oil remover and a heat pipe heat exchanger. The inlet pipe of the cyclone separator is connected with the waste gas discharge pipe, and the waste gas enters the separator from the tangent direction of the cylindrical section. Under the action of centrifugal force, the solid impurities such as fluff are thrown to the wall, fall along the wall to the bottom hopper, and the preliminary dust removal is realized. This stage of treatment can remove most of the fluff impurities that easily block the flow channel, providing a basis guarantee for the subsequent heat exchange link.

[0013] The dust-removed waste gas enters the electrostatic oil remover, and the high-voltage electrode and the grounding electrode plate with a voltage of 12-15kV are arranged in the oil remover. The oil particles in the waste gas are ionized under the action of the high-voltage electric field, and the negatively charged oil particles are adsorbed to the grounding electrode plate. The oil is collected to the oil storage tank through the automatic oil scraping device on the surface of the electrode plate. This stage of treatment can effectively remove the viscous oil in the waste gas, avoid the oil adhering to the heat exchange surface, and reduce the thermal resistance.

[0014] The heat pipe evaporating section of the heat pipe heat exchanger exchanges heat with the waste gas after oil and dust removal, and the condensing section exchanges heat with cold water, realizing waste heat transfer. The flow channel width of the heat exchanger is designed to be 15-20 mm, which can avoid the accumulation and blockage of residual small impurities; the working medium of the heat pipe is R134a, which can achieve a heat transfer coefficient of 800-1000 W / (m²・K) in the waste gas temperature range of 85-120℃, significantly improving the heat exchange efficiency.

[0015] The cold storage waste heat stabilizing system comprises a buffer heat storage tank, the volume of the buffer heat storage tank is 1.2-1.5 times of the exhaust volume of the cold storage compressor per hour, temperature and liquid level sensors are arranged in the tank, and the tank is filled with high-temperature phase change heat storage material. When the exhaust temperature of the compressor is too high, the excess heat is absorbed and stored by the phase change material; when the exhaust temperature is too low, the phase change material releases heat, preliminarily stabilizing the water temperature after heat exchange.

[0016] A proportional regulating valve is further arranged on the buffer heat storage tank and is installed at the intersection of the buffer heat storage tank outlet and the cold water supplement pipeline, and the valve opening degree can be continuously adjusted according to the deviation between the target water temperature 45℃ and the actual water temperature. When the actual water temperature is higher than 45℃, the proportional regulating valve increases the cold water supplement and reduces the heat storage tank outlet flow; when the actual water temperature is lower than 45℃, the proportional regulating valve reduces the cold water supplement and increases the heat storage tank outlet flow, and through accurate flow ratio, finally 45℃±2℃ hot water is stably output.

[0017] The comprehensive control system comprises a collection module, a transmission module and a processing module, the collection module comprises a waste gas inlet temperature sensor T1, a waste gas inlet pressure sensor P1, a waste gas inlet flow sensor F1, a cyclone separator outlet dust concentration sensor C1, an electrostatic oil remover outlet oil concentration sensor C2, a heat pipe heat exchanger waste gas outlet temperature sensor T2, a cold water inlet temperature sensor T3, a cold water outlet temperature sensor T4, a cold water flow sensor F2, a compressor exhaust temperature sensor T5, an exhaust pressure sensor P2, a buffer heat storage tank temperature sensor T6, a liquid level sensor L1, a hot water outlet temperature sensor T7, a hot water flow sensor F3 and a cold water supplement pipeline flow sensor F4.

[0018] The transmission module adopts the industrial Ethernet PROFINET protocol to realize data transmission between the sensors and the processing module, and the sampling frequency is set to 1Hz to ensure the real-time performance of the data; at the same time, an edge computing gateway is used to locally cache the collected data to avoid data loss caused by network interruption, and the cached data is automatically uploaded to the processing module after the network is restored.

[0019] Because the collected raw data has noise interference, the processing module first preprocesses the raw data to ensure the accuracy of the data, and the specific processing steps are as follows:

[0020] Outlier detection and elimination: Grubbs criterion is used to detect outliers, assuming that the measurement data sequence of a sensor is x1, x2,..., xn, the mean value x and the standard deviation s of the data are calculated: n

[0021] x = (1 / n) x∑(i=1 to n) xi;

[0022] s = √[(1 / (n-1)) x∑(i=1 to n)(xi-x)²];

[0023] The Grubbs statistic G1 of each data point is calculated:

[0024] G1 = |xi-x| / s;

[0025] If G1 > Gan, the data is determined to be an outlier and is eliminated, and linear interpolation method is used to complete the data with adjacent two normal data. Wherein, Gan is the Grubbs critical value, alpha is the significance level, taking 0.05; n is the number of data samples, taking 30.

[0026] x = x i-1 + [(x i+1 -x i-1 ) / (t i+1 -t i-1 )] x (t-t i-1 );

[0027] Wherein, t i-1 , t i+1 are the collection time of adjacent normal data, t is the collection time of abnormal data.

[0028] Data filtering: Kalman filter is used to filter the data after eliminating outliers to eliminate random noise. Kalman filter is divided into prediction and update two steps.

[0029] Prediction step:

[0030] x k - = A x k-1 + B u k-1 ;

[0031] P k - = A P k-1 Aᵀ+ Q;

[0032] Wherein, x k - is the predicted state vector at time k, A is the state transition matrix, x k-1 is the optimal estimated state vector at time k-1, B is the control input matrix, u k-1 ​P is the control input at time k-1. k - Let P be the covariance matrix of the predicted state at time k. k-1 Let Q be the covariance matrix of the optimal estimated state at time k-1, and let Q be the process noise covariance matrix.

[0033] Update steps:

[0034] K k = P k - ×Hᵀ×(H×P k - ×Hᵀ+R) -1 ;

[0035] x k = x k - +K k ×(z k -H×x k - );

[0036] P k =(IK k ×H)×P k - ;

[0037] Among them, K k Let z be the Kalman gain at time k, H be the observation matrix, R be the observation noise covariance matrix, and z be the Kalman gain at time k. k Let x be the observation value at time k, i.e., the data collected by the sensor. k Let I be the optimal estimated state vector at time k, and let I be the identity matrix. Kalman filtering can improve the signal-to-noise ratio of the data by 30%-40%, ensuring the accuracy of subsequent modeling and optimization.

[0038] Based on the analysis of preprocessed data, and combined with the principles of heat transfer, fluid mechanics and systems engineering, a mathematical model of the waste heat recovery system is established, including two parts: a heat exchange model of feather drying exhaust gas and a waste heat stabilization model of cold storage.

[0039] The heat exchange model for feather drying exhaust gas is used to describe the heat exchange process of the feather drying exhaust gas processor. Its core is calculating the heat exchange capacity and heat recovery efficiency of the heat pipe heat exchanger, as detailed below:

[0040] Heat exchange calculation: According to the basic heat transfer equation, the heat exchange capacity Q1 of the heat pipe heat exchanger is equal to the heat released by the exhaust gas, and also equal to the heat absorbed by the cold water.

[0041] Q1=m_g×c pg ×(T1-T2)=m_w×c p w×(T4-T3);

[0042] wherein m_g is the exhaust gas mass flow rate, calculated from the exhaust gas volume flow rate F1 and the exhaust gas density p_g according to the ideal gas state equation p_g = (P1 x M) / (R x T1), M is the exhaust gas molar mass, taken as 29 kg / kmol; R is the gas constant, taken as 8.314 kJ / (kmol.K). It is calculated that m_g = F1 x p_g; c pg is the exhaust gas specific heat capacity at constant pressure, fitted according to the exhaust gas temperature T1, and the fitting formula is c pg = 1.005 + 0.00015 x T1; T1, T2 are the exhaust gas inlet and outlet temperatures respectively; m_w is the cold water mass flow rate, calculated from the cold water volume flow rate F2 and the cold water density p_w, m_w = F2 x p_w; c p w is the cold water specific heat capacity at constant pressure, taken as 4.186 kJ / (kg.K); T3, T4 are the cold water inlet and outlet temperatures respectively.

[0043] Heat recovery efficiency calculation: The feather drying exhaust gas heat recovery efficiency η1 is defined as the ratio of the actual heat exchange amount to the theoretical maximum exhaust heat amount of the exhaust gas:

[0044] η1 = Q1 / Q1max = [m_g x c pg x (T1 - T2)] / [m_g x c pg x (T1 - T3)] = (T1 - T2) / (T1 - T3);

[0045] wherein Q1max is the theoretical maximum exhaust heat amount of the exhaust gas, i.e. the heat release amount when the exhaust gas temperature is reduced to the cold water inlet temperature.

[0046] Flow passage blockage early warning model: Based on the dust concentration C1 at the outlet of the cyclone separator and the grease concentration C2 at the outlet of the electrostatic oil remover, a flow passage blockage early warning index S is established to predict the blockage risk of the heat pipe heat exchanger flow passage.

[0047] S = k1 x C1 + k2 x C2';

[0048] wherein k1 is the dust concentration weight coefficient, taken as 0.6; k2 is the grease concentration weight coefficient, taken as 0.4; the unit of C1 is mg / m³; C2' is the converted grease concentration, the unit is mg / m³, calculated from C2 (g / m³) x 1000. According to the experimental data, when S < 50 mg / m³, the blockage risk level is low, and the equipment can operate normally; when 50 < S < 80 mg / m³, the blockage risk level is medium, and the flow passage blowing early warning needs to be started; when S > 80 mg / m³, the blockage risk level is high, and immediate shutdown and cleaning are required to avoid equipment damage.

[0049] The cold storage waste heat stable pressure model is used for describing the temperature regulation process of the cold storage waste heat stable pressure system, and the core is to establish the relationship between the hot water outlet temperature and the control parameter, which is as follows:

[0050] The temperature T in the buffer heat storage tank is affected by the heat exchange amount Q2 of the compressor exhaust and the phase change characteristics of the heat storage material, and the dynamic change equation is as follows:

[0051] m_pcm×c p pcm×(dT6 / dt)=Q2-Q_out;

[0052] Wherein, m_pcm is the mass of the phase change heat storage material, c p pcm is the specific heat capacity of the phase change heat storage material at constant pressure, which is 2.1 kJ / (kg·K) when not phase changing, and the specific heat capacity is considered as infinite during the phase change process; dT6 represents the change amount of the temperature T6 in the heat storage tank in a small time interval; dt represents the small time interval; Q2 is the heat exchange amount transferred from the compressor exhaust to the heat storage tank, which is Q2=m_c×c p c×(T5- T6), wherein m_c is the mass flow rate of the compressor exhaust, c p c is the specific heat capacity of the refrigerant at constant pressure, which is 1.2 kJ / (kg·K), and T5 is the temperature of the compressor exhaust; Q_out is the heat released from the heat storage tank to the hot water pipeline, which is Q_out=m_h×c p w×(T6-T7), wherein m_h is the mass flow rate of the heat storage tank outlet water, and T7 is the hot water outlet temperature.

[0053] The opening degree θ of the proportional regulating valve directly determines the ratio of the cold water supplement F4 and the heat storage tank outlet water flow F_h, and further affects the hot water outlet temperature T7. According to the law of mass conservation and energy conservation, the mathematical relationship between T7 and θ is established:

[0054] T7=[F_h×ρ_w×c p w×T6+F4×ρ_w×c p w×T_c] / [(F_h+F4)×ρ_w×c p w]=[F_h×T6+F4×T_c] / (F_h+F4);

[0055] Wherein, T_c is the cold water supplement temperature, which is the ambient temperature; the relationship between F_h and θ is F_h=F_hmax×(θ / 100%), and F_hmax is the maximum outlet water flow of the heat storage tank; the relationship between F4 and θ is F4=F4max×(1-θ / 100%), and F4max is the maximum cold water supplement. F_h and F4 are substituted into the above formula, and the following can be further deduced:

[0056] T7=[F_hmax×θ×T6+F4max×(100%-θ)×T_c] / [F_hmax×θ+F4max×(100%-θ)];

[0057] Through the formula, the required valve opening θ_target can be estimated according to the target water temperature T7, target = 318K, so as to provide accurate parameter basis for subsequent real-time control.

[0058] In order to realize the double target optimization of maximum waste heat recovery efficiency and minimum system operation cost, the improved particle swarm optimization algorithm is also used to iteratively optimize the key operation parameters of the system, and the specific process includes four stages of parameter initialization, fitness function construction, particle iterative update and convergence judgment.

[0059] Optimization parameter initialization: the key parameter vector to be optimized is determined as X=[F2, θ, v_electrode, v_purge], wherein F2 is the cold water inlet flow of the heat pipe heat exchanger, the value range is 50-150 m³ / h; θ is the opening degree of the proportional regulating valve, the value range is 0-100%; v_electrode is the high-voltage electrode voltage of the electrostatic oil cleaner, the value range is 12-15 kV; v_purge is the flow passage purging air speed of the heat pipe heat exchanger, the value range is 5-15 m / s.

[0060] At the same time, the particle swarm algorithm parameters are initialized: the number of particles N=50, which can balance the optimization accuracy and calculation efficiency; the maximum iteration number G_max=100, the inertia weight ω initial value is 0.9, and it is linearly decreased to 0.4 with the iteration number, the global search ability is enhanced in the early stage, and the local convergence precision is improved in the later stage; the cognitive factor c1=2.0, the social factor c2=2.0, both factors are used to guide the particles to approach the optimal position of the group; the maximum particle speed V_max=0.2×(X_max-X_min), X_max and X_min are the maximum and minimum values of the parameters respectively, to avoid the search divergence caused by too large particle speed.

[0061] Fitness function construction: the fitness function f(X) is the core index to measure the advantages and disadvantages of parameter combination, and the total efficiency of waste heat recovery and the system operation cost are considered comprehensively, and the formula is as follows:

[0062] f(X)=α×[(η1×Q1+η2×Q2) / (Q1+Q2)]-β×[(P_pump+P_electrode+P_purge) / (Q1+Q2)];

[0063] Wherein, η2 is the cold storage waste heat recovery efficiency, η2=(Q2 / Q2max)×100%, Q2max is the theoretical maximum heat release of the compressor exhaust, i.e. Q2max=m_c×c pc x (T5max - T_c), T5max = 363 K;

[0064] a is the weight coefficient of waste heat recovery efficiency, and a = 0.7; b is the weight coefficient of operation cost, and b = 0.3; P_pump is the power of the cold water pump, P_pump = (p_w x g x H x F2) / (3600 x p_pump), g = 9.81 m / s2, H is the pump head, and H = 20 m, p_pump is the pump efficiency, and p_pump = 0.85; P_electrode is the electrode power of the electrostatic oil cleaner, P_electrode = (U x I) / 1000, U = v_electrode, I is the rated current of the electrode, and I = 0.5 A; P_purge is the power of the purge fan, P_purge = (p_air x v_purge3 x A x L) / (3600 x p_fan), p_air is the air density, and p_air = 1.2 kg / m3, A is the cross-sectional area of the flow channel, and A = 0.5 m2, L is the length of the flow channel, and L = 5 m, p_fan is the fan efficiency, and p_fan = 0.8.

[0065] The greater the fitness function value is, the better the comprehensive performance of the system under the current parameter combination is.

[0066] Particle iterative update: the linear decreasing inertia weight and the boundary truncation strategy are adopted to realize the update of the particle position and velocity, so as to avoid the algorithm from falling into local optimum.

[0067] Velocity update formula:

[0068] V_i^g = w x V_i^(g-1) + c1 x r1 x (pbest_i - X_i^(g-1)) + c2 x r2 x (gbest - X_i^(g-1));

[0069] wherein, V_i^g is the velocity vector of the i-th particle in the g-th iteration; X_i^(g-1) is the position vector of the i-th particle in the (g-1)-th iteration, pbest_i is the individual optimal position vector of the i-th particle, that is, the position corresponding to the maximum value of the fitness function in the iteration process of the particle; gbest is the global optimal position vector of the entire particle group, that is, the position corresponding to the maximum value of the fitness function in the iteration process of all particles; r1 and r2 are random numbers between 0 and 1. If V_i^g > V_max or V_i^g < -V_max, it is truncated to V_max or -V_max.

[0070] Position update formula:

[0071] X_i^g = X_i^(g-1) + V_i^g;

[0072] If X_i^g is out of the parameter value range, for example, F2>150 m³ / h or F2<50 m³ / h, it will be adjusted to the corresponding maximum or minimum value to ensure the physical validity of the parameters.

[0073] Convergence judgment: Calculate the global optimal fitness value f(gbest) after each iteration. Terminate the iteration when any of the following conditions is met:

[0074] The number of iterations reaches the maximum number of iterations G_max=100, and the global optimal fitness value changes by |f(gbest)^g-f (gbest)^(g-10)|<10^-6 for 10 consecutive iterations, i.e., the algorithm converges, and further iteration has no significant optimization effect.

[0075] After the iteration is terminated, the global optimal parameter vector X_opt=[F2,opt,θ_opt, v_electrode,opt,v_purge,opt] is output as the reference parameter for real-time control of the system.

[0076] Real-time control and feedback: Based on the optimal parameters output by the optimization algorithm, combined with the real-time running state of the system, a fuzzy PID control strategy is used to achieve precise regulation and control of the equipment, and a closed-loop feedback mechanism is used to continuously optimize the parameters to ensure that the system maintains optimal performance when the working conditions change.

[0077] Take the fluff drying waste gas heat recovery efficiency deviation e_η1=η1target-η1 and the hot water outlet temperature deviation e_T7=T7target- T7 as input variables, the cold water flow rate adjustment amount ΔF2 and the proportional adjustment valve opening adjustment amount Δθ as output variables, and construct a two-dimensional fuzzy PID controller, where η1target is 75% and T7target is 318 K.

[0078] Input variable fuzzification: The fuzzy subsets of e_η1 are {NB (negative big), NM (negative medium), NS (negative small), ZO (zero), PS (positive small), PM (positive medium), and PB (positive big)}, and the domain is [-10%, 10%]. The fuzzy subsets of e_T7 are {NB, NM, NS, ZO, PS, PM, and PB}, and the domain is [-5K, 5K].

[0079] Output variable fuzzification: The fuzzy subsets of ΔF2 are {NB, NM, NS, ZO, PS, PM, and PB}, and the domain is [-20 m³ / h, 20 m³ / h]. The fuzzy subsets of Δθ are {NB, NM, NS, ZO, PS, PM, and PB}, and the domain is [-10%, 10%].

[0080] Fuzzy rule base construction: Based on expert experience and 1000 sets of experimental data, core fuzzy rules are established, for example:

[0081] If the efficiency is far below the target e_η1=PB and the temperature meets the target e_T7=ZO, increase the cold water flow to improve the heat exchange efficiency ΔF2=PB, and keep the opening unchanged Δθ=ZO;

[0082] If the efficiency meets the target e_η1=ZO and the temperature is far below the target e_T7=PB, keep the flow unchanged ΔF2=ZO, and increase the valve opening to increase the proportion of hot water Δθ=PB;

[0083] If the efficiency is far above the target e_η1=NB and the temperature is far above the target e_T7=NB, reduce the cold water flow ΔF2=NB, and reduce the valve opening to increase the proportion of cold water Δθ=NB.

[0084] Deburring: the barycentric method is used to convert the fuzzy output into precise control quantity, and the formula is as follows:

[0085] Δu=[Σ(i=1 to n)(μ_i×u_i)] / [Σ(i=1 to n)μ_i];

[0086] Wherein, μ_i is the membership degree of the i-th fuzzy subset, u_i is the center value of the i-th fuzzy subset, and n is the number of fuzzy subsets, which is 7.

[0087] Real-time acquisition of system operation data, compared with target value, if the following conditions occur, feedback iteration is triggered, and the improved particle swarm optimization algorithm is restarted:

[0088] |e_η1|>5%, and the duration is more than 5 minutes;

[0089] |e_T7|>2K, and the duration is more than 3 minutes;

[0090] The clogging early warning index S≥50mg / m³, and the duration is more than 10 minutes.

[0091] When feedback iteration, the current system operation data is taken as the new initial condition, the parameter vector X_opt is re-optimized, and the proportional coefficient K_p, the integral coefficient K_i and the differential coefficient K_d of the fuzzy PID controller are updated synchronously, so that the response time of the system to the working condition change is less than or equal to 10 seconds, and the anti-interference ability is significantly better than that of the traditional control method.

[0092] The above are only specific embodiments of the present application, and cannot limit the scope of the application. Therefore, the replacement of equivalent components or equivalent changes and modifications made within the scope of the patent protection of the present application shall still fall within the scope of the claims of the present application.

Claims

1. A multi-grade waste heat recovery device for poultry processing plants, characterized in that, The system includes a feather drying exhaust gas treatment system, a cold storage waste heat stabilization system, and a comprehensive control system. The feather drying exhaust gas treatment system includes a cyclone separator and an electrostatic precipitator. The inlet pipe of the cyclone separator is connected to the exhaust gas outlet pipe. The exhaust gas enters tangentially from the cylindrical section of the separator. Under the action of centrifugal force, solid impurities are thrown against the wall of the separator and fall down to the bottom ash hopper. The dust-removed exhaust gas enters the electrostatic precipitator, which is equipped with a high-voltage electrode and a grounding plate. The grease particles in the exhaust gas are ionized under the action of the high-voltage electric field, and the negatively charged grease particles are adsorbed onto the grounding plate.

2. The multi-grade waste heat recovery device for poultry processing plants according to claim 1, characterized in that, The feather drying exhaust gas treatment system also includes a heat pipe heat exchanger. The heat pipe evaporation section of the heat pipe heat exchanger exchanges heat with the exhaust gas after oil and dust removal, and the condensation section exchanges heat with cold water. The flow channel width of the heat exchanger is 15-20mm.

3. The multi-grade waste heat recovery device for poultry processing plants according to claim 2 or 3, characterized in that, The cold storage waste heat stabilization system includes a buffer heat storage tank, the volume of which is 1.2-1.5 times the hourly discharge capacity of the cold storage compressor. Temperature and liquid level sensors are installed inside the tank, which is filled with high-temperature phase change heat storage material. The buffer heat storage tank is also equipped with a proportional regulating valve, which is installed at the junction of the outlet of the buffer heat storage tank and the cold water supply pipeline. The valve opening can be continuously adjusted according to the deviation between the target water temperature and the actual water temperature.

4. The multi-grade waste heat recovery device for poultry processing plants according to claim 3, characterized in that, The integrated control system includes a data acquisition module, a transmission module, and a processing module. The data acquisition module includes a waste gas inlet temperature sensor T1, a waste gas inlet pressure sensor P1, a waste gas inlet flow sensor F1, a cyclone separator outlet dust concentration sensor C1, an electrostatic precipitator outlet grease concentration sensor C2, a heat pipe heat exchanger waste gas outlet temperature sensor T2, a cold water inlet temperature sensor T3, a cold water outlet temperature sensor T4, a cold water flow sensor F2, a compressor exhaust temperature sensor T5, an exhaust pressure sensor P2, a buffer storage tank internal temperature sensor T6, a liquid level sensor L1, a hot water outlet temperature sensor T7, a hot water flow sensor F3, and a cold water makeup pipeline flow sensor F4.

5. The multi-grade waste heat recovery device for poultry processing plants according to claim 4, characterized in that, The processing module preprocesses the raw data collected by the sensor. The specific processing steps are as follows: The Grubbs criterion is used to detect outliers. Assume a sensor's measurement data sequence is x1, x2, ..., x... n Calculate the mean x and standard deviation s of the data: x = (1 / n) × Σ (i = 1 to n) xᵢ; s = √[(1 / (n-1)) × Σ(i=1 to n)(xᵢ-x)²]; Calculate the Grubbs statistic Gᵢ for each data point: Gᵢ=|xᵢ-x| / s; If Gᵢ>Gαn, the data is considered an outlier and is removed. The data is then supplemented using linear interpolation between two adjacent normal data points. Here, Gαn is the Grubbs critical value, α is the significance level, and n is the number of data samples. x complement = x i-1 +[(x i+1 -x i-1 ) / (t i+1 -t i-1 )]×(t - t i-1 ); Among them, t i-1 ,t i+1 t represents the collection time of adjacent normal data, and t represents the collection time of abnormal data. Kalman filtering is used to filter the data after outlier removal to eliminate random noise. Prediction steps: x k - =A×x k-1 +B×u k-1 ; P k - =A×P k-1 ×Aᵀ+Q; Where, x k - Let x be the predicted state vector at time k, A be the state transition matrix, and x be the predicted state vector at time k. k-1 Let B be the optimal estimated state vector at time k-1, and let B be the control input matrix. k-1 P is the control input at time k-1. k - Let P be the covariance matrix of the predicted state at time k. k-1 Let Q be the covariance matrix of the optimal estimated state at time k-1, and let Q be the process noise covariance matrix. Update steps: K k = P k - ×Hᵀ×(H×P k - ×Hᵀ+R) -1 ; x k = x k - +K k ×(z k -H×x k - ); P k =(I-K k ×H)×P k - ; Among them, K k Let z be the Kalman gain at time k, H be the observation matrix, R be the observation noise covariance matrix, and z be the Kalman gain at time k. k Let x be the observation value at time k, i.e., the data collected by the sensor. k Let I be the optimal estimated state vector at time k, and let I be the identity matrix.

6. The multi-grade waste heat recovery device for poultry processing plants according to claim 5, characterized in that, Based on the analysis of the pre-processed data, a mathematical model of the waste heat recovery system is established, including a heat exchange model for feather drying exhaust gas and a waste heat stabilization model for cold storage. The heat exchange model for feather drying exhaust gas is used to describe the heat exchange process of the feather drying exhaust gas processor. The specific steps for establishing the model are as follows: The heat exchange capacity of the heat pipe heat exchanger is Q1: Q1=m_g×c pg ×(T1-T2)=m_w×c p w×(T4-T3); Where m_g is the exhaust gas mass flow rate, c pg Here, T1 and T2 represent the inlet and outlet temperatures of the exhaust gas, respectively, and m_w is the mass flow rate of the chilled water. p w represents the specific heat capacity of cold water at constant pressure, and T3 and T4 are the inlet and outlet temperatures of the cold water, respectively. The heat recovery efficiency η1 of feather drying exhaust gas is defined as the ratio of the actual heat exchange to the theoretical maximum heat release of the exhaust gas: η1=Q1 / Q1max=[m_g×c pg ×(T1-T2)] / [m_g×c pg ×(T1-T3)]=(T1-T2) / (T1 -T3); Where Q1max is the theoretical maximum heat release of the exhaust gas, that is, the heat release when the exhaust gas temperature drops to the cold water inlet temperature. Based on the dust concentration C1 at the outlet of the cyclone separator and the grease concentration C2 at the outlet of the electrostatic precipitator, a flow channel blockage early warning index S is established to predict the blockage risk of the flow channel of the heat pipe heat exchanger. S = k1 × C1 + k2 × C2'; Where k1 is the dust concentration weighting coefficient, k2 is the grease concentration weighting coefficient, and C2' is the converted grease concentration; when S < 50 mg / m³, the blockage risk level is low, and the equipment can operate normally; when 50 ≤ S < 80 mg / m³, the blockage risk level is medium, and the flow channel purging warning needs to be activated; when S ≥ 80 mg / m³, the blockage risk level is high, and the machine needs to be stopped immediately for cleaning.

7. The multi-grade waste heat recovery device for poultry processing plants according to claim 6, characterized in that, The waste heat stabilization model for cold storage is used to describe the temperature control process of the waste heat stabilization system in cold storage. The specific steps for establishing the model are as follows: The temperature T inside the buffer storage tank is affected by the heat exchange capacity Q2 of the compressor exhaust and the phase change characteristics of the storage material. Its dynamic change equation is as follows: m_pcm×c p pcm×(dT6 / dt)=Q2-Q_out; Where m_pcm is the mass of the phase change thermal storage material, and c p pcm is the isobaric specific heat capacity of the phase change thermal storage material, dT6 is the change in temperature T6 inside the thermal storage tank within a small time interval, dt is the small time interval, Q2 is the heat transfer from the compressor exhaust to the thermal storage tank, and Q_out is the heat released from the thermal storage tank to the hot water pipeline. The opening degree θ of the proportional control valve determines the ratio of cold water makeup water F4 to the outlet water flow rate F_h of the thermal storage tank, which in turn affects the hot water outlet temperature T. 7； Based on the laws of conservation of mass and energy, establish the mathematical relationship between T7 and θ: T7=[F_hmax×θ×T6+F4max×(100%-θ)×T_c] / [F_hmax×θ+F4max×(100%-θ)]; Where T_c is the cold water supply temperature, F_hmax is the maximum outflow rate of the thermal storage tank, and F4max is the maximum cold water supply volume.

8. The multi-grade waste heat recovery device for poultry processing plants according to claim 7, characterized in that, The processing module also employs an improved particle swarm optimization algorithm to iteratively optimize the system's operating parameters. The specific optimization steps are as follows: The key parameter vector to be optimized is X=[F2,θ, v_electrode,v_purge], where F2 is the cold water inlet flow rate of the heat pipe heat exchanger, θ is the opening degree of the proportional control valve, v_electrode is the high-voltage electrode voltage of the electrostatic precipitator, and v_purge is the purge velocity of the heat pipe heat exchanger flow channel. The number of particles N=50, the maximum number of iterations G_max=100, the initial value of the inertia weight ω is 0.9, and it decreases linearly to 0.4 with the number of iterations, the cognitive factor c1=2.0, the social factor c2=2.0, the maximum particle velocity V_max=0.2×(X_max-X_min), where X_max and X_min are the maximum and minimum values ​​of the parameters, respectively. Fitness function f(X): f(X)=α×[(η1×Q1+η2×Q2) / (Q1+Q2)]-β×[(P_pump+P_electrode+P_purge) / (Q1+Q2)]; Where η2 is the waste heat recovery efficiency of the cold storage, α is the weighting coefficient of waste heat recovery efficiency, β is the weighting coefficient of operating cost, P_pump is the power of the chilled water pump, P_electrode is the power of the electrostatic precipitator electrode, and P_purge is the power of the purge fan. A linearly decreasing inertia weight and boundary truncation strategy are used to update particle position and velocity. The velocity update formula is as follows: V_i^g=ω×V_i^(g-1)+c1×r1×(pbest_i-X_i^(g-1))+c2×r2×(gbest-X_i^(g-1)); Where V_i^g is the velocity vector of the i-th particle in the g-th iteration, X_i^(g-1) is the position vector of the i-th particle in the g-1-th iteration, pbest_i is the individual optimal position vector of the i-th particle, gbest is the global optimal position vector of the entire particle swarm, and r1 and r2 are random numbers between 0 and 1. If V_i^g > V_max or V_i^g < -V_max, truncate it to V_max or -V_max, then the position update formula is: X_i^g = X_i^(g-1) + V_i^g; If X_i^g exceeds the parameter value range, adjust it to the corresponding maximum or minimum value; After each iteration, the global optimal fitness value f(gbest) is calculated. The iteration terminates when any of the following conditions are met: When the number of iterations reaches the maximum number of iterations G_max=100, the change in the global optimal fitness value after 10 consecutive iterations is |f(gbest)^gf(gbest)^(g-10)|<10^-6, which means the algorithm converges and further iterations will not have a significant optimization effect. After the iteration terminates, the globally optimal parameter vector X_opt=[F2,opt,θ_opt, v_electrode,opt,v_purge,opt] is output as the reference parameter for real-time control of the system.

9. The multi-grade waste heat recovery device for poultry processing plants according to claim 8, characterized in that, Based on the optimal parameters output by the optimization algorithm and combined with the real-time operating status of the system, a fuzzy PID control strategy is adopted to achieve precise control of the equipment, and the parameters are continuously optimized through a closed-loop feedback mechanism. The specific control steps are as follows: A two-dimensional fuzzy PID controller is constructed with the deviation of heat recovery efficiency of feather drying exhaust gas e_η1=η1target-η1 and the deviation of hot water outlet temperature e_T7=T7target-T7 as input variables, and the adjustment amount of cold water flow rate ΔF2 and the adjustment amount of proportional control valve opening Δθ as output variables. The fuzzy subset of e_η1 is {NB, NM, NS, ZO, PS, PM, PB}, and the universe of discourse is [-10%, 10%]; the fuzzy subset of e_T7 is {NB, NM, NS, ZO, PS, PM, PB}, and the universe of discourse is [-5K, 5K]. The fuzzy subset of ΔF2 is {NB,NM,NS,ZO,PS,PM,PB}, and the universe of discourse is [-20m³ / h, 20m³ / h]; the fuzzy subset of Δθ is {NB,NM,NS,ZO,PS,PM,PB}, and the universe of discourse is [-10%, 10%]; Based on experimental data, core fuzzy rules are established; The centroid method is used to convert the output of fuzzy rules into precise control quantities, as shown in the following formula: Δu=[Σ(i=1 to n)(μ_i×u_i)] / [Σ(i=1 to n)μ_i]; Where μ_i is the membership degree of the i-th fuzzy subset, u_i is the center value of the i-th fuzzy subset, and n is the number of fuzzy subsets; The system collects operational data in real time and compares it with the target value. If any of the following conditions occur, a feedback iteration is triggered, and the improved particle swarm optimization algorithm is restarted: |e_η1|>5%, and the duration exceeds 5 minutes; |e_T7|>2K, and the duration exceeds 3 minutes; The congestion warning indicator S ≥ 50 mg / m³ and lasts for more than 10 minutes; During feedback iteration, the current system running data is used as the new initial condition to re-optimize the parameter vector X_opt, and the proportional coefficient K_p, integral coefficient Ki, and derivative coefficient K_d of the fuzzy PID controller are updated synchronously.