Method for controlling urea injection quantity of marine diesel engine scr system and related products

Abstract

The application provides a urea injection amount control method, device, equipment and medium for a marine diesel engine SCR system. The method comprises the following steps: obtaining NOx conversion error and control parameter value of the marine diesel engine SCR system at a previous moment; determining control parameter value of the marine diesel engine SCR system at a current moment by a sliding mode adaptive controller according to NOx conversion error and control parameter value at the previous moment; and controlling urea injection amount by a PID controller according to the control parameter value at the current moment. X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment X target conversion rate, NOx conversion error and control parameter value at the previous moment The method improves urea injection amount control precision.

Description

Technical Field

[0001] This application relates to the field of marine diesel engine SCR system control technology, and in particular to a method, apparatus, equipment and storage medium for controlling the urea injection quantity of a marine diesel engine selective catalytic reduction (SCR) system. Background Technology

[0002] Marine diesel engine SCR systems can reduce NO emissions from marine diesel engines. X The NH3 produced by the thermal decomposition of urea solution undergoes a redox reaction to generate H2O and N2, thereby reducing NO emissions into the air. X Therefore, the most critical issue in a marine diesel engine SCR system is selecting the urea solution injection rate.

[0003] If too much urea is injected, some of the decomposed NH3 will escape, causing secondary pollution to the environment; if too little urea is injected, insufficient NH3 and NO will not be decomposed. X A redox reaction occurs, leading to NO X The conversion rate is low. In traditional technology, the amount of urea injected can be controlled by pre-calibrating the urea injection pulse spectrum under different operating conditions. However, this method involves a large amount of calibration work, has low control accuracy, and fails to meet the specified emission reduction requirements. Summary of the Invention

[0004] Based on this, embodiments of this application provide a method, apparatus, device, and storage medium for controlling the urea injection quantity in a marine diesel engine SCR system, which can improve the control accuracy of the urea injection quantity, thereby improving the NO content of the marine diesel engine SCR system. X The conversion efficiency meets the prescribed emission reduction requirements.

[0005] In a first aspect, embodiments of this application provide a method for controlling the urea injection quantity of a marine diesel engine SCR system, comprising:

[0006] Obtain the NO of the marine diesel engine SCR system at the previous moment. X Target conversion rate, NO X Actual conversion rate, NO X Conversion error and control parameter values; wherein, the control parameters include the proportional parameters, integral parameters and derivative parameters of the PID controller in the marine diesel engine SCR system;

[0007] Based on the previous moment NO X Target conversion rate, NO X Actual conversion rate, NO X Conversion error and control parameter values updatinga network parameter value of the sliding mode adaptive controller; wherein the sliding mode adaptive controller comprises a sliding mode equivalent control part and a sliding mode switching control part, the sliding mode equivalent control part is implemented by a first neural network, and the sliding mode switching control part is implemented by a second neural network;

[0008] the NOx at the last moment X a target conversion rate, NOx X an actual conversion rate, NOx X a conversion error, and a control parameter value of the ship diesel engine SCR system at the current moment by the updated sliding mode adaptive controller;

[0009] controlling the urea injection amount by the PID controller according to the control parameter value at the current moment.

[0010] In a second aspect, the embodiments of the present application provide a ship diesel engine SCR system urea injection amount control device, comprising:

[0011] an obtaining module, configured to obtain the NOx at the last moment of the ship diesel engine SCR system X a target conversion rate, NOx X an actual conversion rate, NOx X a conversion error, and a control parameter value of the ship diesel engine SCR system at the current moment by the updated sliding mode adaptive controller;

[0012] an updating module, configured to update the network parameter value of the sliding mode adaptive controller according to the NOx at the last moment X a target conversion rate, NOx X an actual conversion rate, NOx X a conversion error, and a control parameter value of the ship diesel engine SCR system at the current moment by the updated sliding mode adaptive controller;

[0013] a first determining module, configured to determine the NOx at the last moment X a target conversion rate, NOx X an actual conversion rate, NOx X a conversion error, and a control parameter value of the ship diesel engine SCR system at the current moment by the updated sliding mode adaptive controller;

[0014] a control module, configured to control the urea injection amount by the PID controller according to the control parameter value at the current moment.

[0015] In a third aspect, an electronic device is provided, and the electronic device comprises a memory and a processor. The memory stores a computer program. The processor implements the steps of the method for controlling the urea injection amount of the marine diesel engine SCR system according to the first aspect of the embodiments of the present application when executing the computer program.

[0016] In a fourth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program implements the steps of the method for controlling the urea injection amount of the marine diesel engine SCR system according to the first aspect of the embodiments of the present application when executed by a processor.

[0017] The technical scheme provided by the embodiments of the present application obtains the NOx conversion rate of the marine diesel engine SCR system at the previous time point, the target conversion rate of the marine diesel engine SCR system at the current time point, the actual conversion rate of the marine diesel engine SCR system at the current time point, and the conversion error of the marine diesel engine SCR system at the current time point. X The target conversion rate of the marine diesel engine SCR system at the current time point, the actual conversion rate of the marine diesel engine SCR system at the current time point, and the conversion error of the marine diesel engine SCR system at the current time point are obtained. X The target conversion rate of the marine diesel engine SCR system at the current time point, the actual conversion rate of the marine diesel engine SCR system at the current time point, and the conversion error of the marine diesel engine SCR system at the current time point are obtained. X The conversion error and the control parameter value are obtained, and the control parameter comprises the proportional parameter, the integral parameter and the differential parameter of the PID controller in the marine diesel engine SCR system. X The target conversion rate of the marine diesel engine SCR system at the current time point, the actual conversion rate of the marine diesel engine SCR system at the current time point, and the conversion error of the marine diesel engine SCR system at the current time point are obtained. X The target conversion rate of the marine diesel engine SCR system at the current time point, the actual conversion rate of the marine diesel engine SCR system at the current time point, and the conversion error of the marine diesel engine SCR system at the current time point are obtained. X The network parameter value of the sliding mode adaptive controller is updated according to the conversion error and the control parameter value of the marine diesel engine SCR system at the current time point, and the sliding mode adaptive controller comprises a sliding mode equivalent control part and a sliding mode switching control part. The sliding mode equivalent control part is implemented by a first neural network, and the sliding mode switching control part is implemented by a second neural network. X The target conversion rate of the marine diesel engine SCR system at the current time point, the actual conversion rate of the marine diesel engine SCR system at the current time point, and the conversion error of the marine diesel engine SCR system at the current time point are obtained. X The target conversion rate of the marine diesel engine SCR system at the current time point, the actual conversion rate of the marine diesel engine SCR system at the current time point, and the conversion error of the marine diesel engine SCR system at the current time point are obtained. X The control parameter value of the marine diesel engine SCR system at the current time point is determined by the updated sliding mode adaptive controller. The urea injection amount is controlled by the PID controller according to the control parameter value at the current time point. That is, the sliding mode equivalent control function in the sliding mode adaptive controller is implemented by the first neural network, and the sliding mode switching control function in the sliding mode adaptive controller is implemented by the second neural network. The network parameter value of the sliding mode adaptive controller can be updated adaptively in combination with the urea injection amount, the influence of the interference factor on the control process is reduced, the control parameter value of the PID controller at the current time point is accurate, the control precision of the urea injection amount is improved, the conversion rate of the NOx is improved, and the emission reduction requirement is met. X BRIEF DESCRIPTION OF DRAWINGS BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A flowchart of the method for controlling the urea injection amount of the marine diesel engine SCR system is provided.

[0019] Figure 2A structural schematic diagram of a sliding mode adaptive controller provided by an embodiment of the present application is shown in FIG. 1.

[0020] Figure 3 A flowchart of a process for determining initial network parameter values of a sliding mode adaptive controller provided by an embodiment of the present application is shown in FIG. 4.

[0021] Figure 4 A structural schematic diagram of a urea injection amount control device for a marine diesel engine SCR system provided by an embodiment of the present application is shown in FIG. 5.

[0022] Figure 5 Another structural schematic diagram of a urea injection amount control device for a marine diesel engine SCR system provided by an embodiment of the present application is shown in FIG. 6.

[0023] Figure 6 A structural schematic diagram of an electronic device provided by an embodiment of the present application is shown in FIG. 7. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions in the embodiments of the present application are described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely intended to explain the present application and should not be used to limit the present application.

[0025] It should be noted that the execution subject of the following method embodiments can be a urea injection amount control device for a marine diesel engine SCR system, which can be implemented by software, hardware, or a combination of software and hardware to become part or all of an electronic device. Alternatively, the electronic device can be a computer, a tablet, or a portable device, etc., or can be a standalone server or a server cluster, etc., and the specific type of the electronic device is not limited in the embodiments of the present application.

[0026] Next, the marine diesel engine SCR system is introduced first.

[0027] I. SCR reaction mechanism

[0028] NH3 produced by thermal decomposition of the urea solution enters the inside of the catalyst, and a part of the NH3 enters the system and is adsorbed on the surface of the catalyst to form adsorbed ammonia (NH3(ads)), which will react with the NOx emitted by the marine diesel engine. X NH3 produced by thermal decomposition of the urea solution enters the inside of the catalyst, and a part of the NH3 enters the system and is adsorbed on the surface of the catalyst to form adsorbed ammonia (NH3(ads)), which will react with the NOx emitted by the marine diesel engine. XThe redox reaction is carried out to generate H2O and N2. The remaining ammonia gas will be left inside the catalyst in the form of gaseous ammonia (NH3), and the adsorbed ammonia and the gaseous ammonia can be converted into each other. The NH3 that is not adsorbed and the NH3 formed by desorption of the adsorbed ammonia (NH3(ads)) will largely escape from the catalyst, causing ammonia leakage. The following formula (1) represents the redox reaction of ammonia and nitrogen oxide, and formula (2) represents the adsorption and desorption reaction of ammonia.

[0029] 4NH3(ads) + 4NO x + 2O2→ 4N2+ 6H2O (1)

[0030]

[0031] Under certain conditions, the adsorbed ammonia (NH3(ads)) will react with the NO X that enters the catalyst to generate N2 and H2O, and the reaction formula is as follows:

[0032] 4NH3(ads) + 3O2→ 2N2+ 6H2O (3)

[0033] The ultimate control purpose of the SCR system of the marine diesel engine is to improve the NO X conversion rate under the premise of ensuring a small amount of ammonia escape. To control the SCR system of the marine diesel engine, a model capable of representing the dynamic changes of each component inside the SCR catalyst must be established.

[0034] II. Modeling of the SCR system

[0035] According to the Arrhenius equation, the reaction rate of each chemical reaction in the SCR system of the marine diesel engine is modeled.

[0036] The following formula (4) is the reduction reaction rate R scr of the SCR system of the marine diesel engine:

[0037]

[0038] The following formula (5) is the adsorption reaction rate R ads of the SCR system of the marine diesel engine:

[0039]

[0040] The following formula (6) is the desorption reaction rate R des of the SCR system of the marine diesel engine:

[0041] R des = r des k des exp[-Ea,des (RT) * θ (6)

[0042] The following formula (7) is the ammonia oxidation reaction rate R of the marine diesel engine SCR system ox :

[0043] R ox = c s k ox exp[-E a,ox / (RT)]*θ (7)

[0044] Where, r des is the desorption reaction rate constant, R is the gas constant, T is the average temperature in the catalyst, θ is the ammonia coverage, c s is the ammonia storage capacity of the catalyst, E a,des is the desorption reaction activity, E a,scr is the reduction reaction activity, S c is the surface area of active atoms, α prob is the sticking rate, k scr is the rate factor of the reduction reaction, k des is the speed factor of the desorption reaction, is the molar mass of ammonia gas, is the concentration of ammonia, E a,scr is the reduction reaction activity, is the concentration of , k ox is the ammonia oxidation reaction rate factor, E a,ox is the oxidation reaction activity.

[0045] And according to the mass conservation law of nitrogen oxides and ammonia in the marine diesel engine SCR system and the energy conservation law inside the catalyst, the chemical reaction in the catalyst of the marine diesel engine SCR system is modeled kinetically.

[0046] After modeling the marine diesel engine SCR system, the technical solution provided by the embodiment of the application is applied to the marine diesel engine SCR system model to control the urea injection amount of the marine diesel engine SCR system.

[0047] Figure 1 Figure 1 is a flowchart of a urea injection amount control method for a marine diesel engine SCR system provided by an embodiment of the application. As shown in Figure 1, the method can include: Figure 1

[0048] S101, obtaining the NO X target conversion rate, NO X actual conversion rate, NO X conversion error and control parameter value of the marine diesel engine SCR system at the last time. ​

[0049] The control parameters include a proportional parameter, an integral parameter and a differential parameter of a PID controller in the SCR system of the marine diesel engine.

[0050] The NOx conversion error is obtained by subtracting the NOx conversion rate from the NOx target conversion rate. X The NOx target conversion rate is the desired NOx conversion rate of the SCR system of the marine diesel engine. X The NOx conversion rate, in practical applications, can be set based on actual needs, and the NOx conversion rate at each moment is the same as the NOx target conversion rate. X The NOx target conversion rate is set based on actual needs, and the NOx conversion rate at each moment is the same as the NOx target conversion rate. X The NOx target conversion rate is set based on actual needs, and the NOx conversion rate at each moment is the same as the NOx target conversion rate. X The NOx actual conversion rate can be obtained based on the urea injection amount at the previous moment, i.e., through the redox reaction between ammonia and nitrogen oxides generated by the thermal decomposition of urea at the previous moment, the NOx actual conversion rate at the previous moment can be obtained. X The NOx target conversion rate is set based on actual needs, and the NOx conversion rate at each moment is the same as the NOx target conversion rate. X The NOx conversion error is obtained by subtracting the NOx conversion rate from the NOx target conversion rate. X The NOx target conversion rate is set based on actual needs, and the NOx conversion rate at each moment is the same as the NOx target conversion rate. X The NOx conversion error is obtained by subtracting the NOx conversion rate from the NOx target conversion rate. The control parameter values include the proportional parameter value, the integral parameter value and the differential parameter value of the PID controller.

[0051] S102, the NOx target conversion rate at the previous moment, the NOx actual conversion rate at the previous moment and the NOx conversion error at the previous moment are input into the sliding mode adaptive controller to update the network parameter values of the sliding mode adaptive controller. X The NOx target conversion rate at the previous moment, the NOx actual conversion rate at the previous moment and the NOx conversion error at the previous moment are input into the sliding mode adaptive controller to update the network parameter values of the sliding mode adaptive controller. X The NOx target conversion rate at the previous moment, the NOx actual conversion rate at the previous moment and the NOx conversion error at the previous moment are input into the sliding mode adaptive controller to update the network parameter values of the sliding mode adaptive controller. X The NOx target conversion rate at the previous moment, the NOx actual conversion rate at the previous moment and the NOx conversion error at the previous moment are input into the sliding mode adaptive controller to update the network parameter values of the sliding mode adaptive controller.

[0052] The sliding mode adaptive controller includes a sliding mode equivalent control part and a sliding mode switching control part, the sliding mode equivalent control part is realized by a first neural network, and the sliding mode switching control part is realized by a second neural network.

[0053] In this embodiment, the sliding mode control is combined with the neural network to solve the problems of slow learning speed and low precision of the neural network. The most critical thing to realize the sliding mode control is to determine the sliding mode surface and the control law. Generally, the sliding mode control is composed of a sliding mode switching control and a sliding mode equivalent control. The sliding mode control process can be represented by the following formulas (8) and (9):

[0054] u = u eq + u s = u eq + β sgn(s) (8)

[0055]

[0056] Sliding mode equivalent control is similar to the inverse dynamics of a marine diesel engine's SCR system. When network parameters cannot be determined in advance, the calculated equivalent control quantity differs significantly from the actual value. (See also...) Figure 2 To eliminate errors, this embodiment employs a first neural network (such as...). Figure 2 The sliding mode equivalent control function is achieved by using NN1 in the second neural network (such as NN1 in the second neural network). Figure 2 The NN2 in the model realizes the sliding mode switching control function. By adjusting the network parameter values ​​of each neural network (such as weights, biases, etc.) online, the amount of deviation of the trajectory of the marine diesel engine SCR system from the sliding surface is compensated in a timely manner, realizing full-process sliding mode control during the motion process, thereby effectively reducing chattering and optimizing the tracking performance of the marine diesel engine SCR system.

[0057] Optionally, the first and second neural networks can be backpropagation (BP) neural networks. Of course, the first and second neural networks can also be radial basis function neural networks or recurrent neural networks, etc.

[0058] Furthermore, when both the first and second neural networks are backpropagation (BP) neural networks, the neurons in each layer of the first neural network are fully connected. The number of neurons in the hidden layers of the second neural network is the same as the number of sliding mode functions. Each input in the second neural network is connected to only one neuron in the hidden layer, while all neurons in the hidden layers are fully connected to the output layer. The input and output of the first neural network can be determined by the sliding mode equivalent control part, and the training objective function of the first neural network is the squared error between the target output and the actual output. The input and output of the second neural network can be determined by the sliding mode switching control part, and the training objective function of the second neural network is the error function between the target output and the actual output.

[0059] When the trajectory of the marine diesel engine SCR system deviates from the sliding surface, the marine diesel engine SCR system will adopt discontinuous control to redirect the system trajectory back to the sliding surface. Real-time adjustment of two neural network parameters effectively compensates for the deviation from the sliding surface, thereby eliminating the chattering phenomenon generated by the system. When the system trajectory is within the sliding surface, the switching control input is zero, and the normal operation of the system mainly relies on equivalent control.

[0060] Therefore, the NO from the previous moment can be... X Target conversion rate, NO X Actual conversion rate, NO X The conversion error and control parameter values ​​are added to the training dataset, which is then used to train the sliding mode adaptive controller (SADC) to update its network parameter values. Specifically, during the training of the first neural network in the SADC, the NO values ​​from the previous time step are added... X Target conversion rate, NOX actual conversion rate, NO X The conversion error is taken as the input of the first neural network and the equivalent sliding mode control amount at the last moment as the expected output, and the first neural network is trained. When training the second neural network, the NO X target conversion rate, NO X actual conversion rate, NO X The conversion error is taken as the input of the second neural network and the switching sliding mode control amount at the last moment as the expected output, and the second neural network in the sliding mode adaptive controller is trained. The equivalent sliding mode control amount and the switching sliding mode control amount at the last moment can be determined based on the control parameter value of the PID controller in the marine diesel engine SCR system at the last moment.

[0061] S103, according to the NO X target conversion rate, NO X actual conversion rate and NO X conversion error, the control parameter value of the marine diesel engine SCR system at the current moment is determined through the updated sliding mode adaptive controller.

[0062] After updating the network parameter value of the sliding mode adaptive controller, the updated sliding mode adaptive controller is obtained, so that the NO X target conversion rate, NO X actual conversion rate and NO X conversion error, the control parameter value of the PID controller in the marine diesel engine SCR system at the current moment is optimized through the updated sliding mode adaptive controller, that is, the NO X target conversion rate, NO X actual conversion rate and NO X conversion error, is input into the updated sliding mode adaptive controller, and the output of the sliding mode adaptive controller is the proportional parameter value, integral parameter value and derivative parameter value of the PID controller at the current moment.

[0063] S104, according to the control parameter value at the current moment, the urea injection amount is controlled through the PID controller.

[0064] After obtaining the control parameter value of the marine diesel engine SCR system at the current moment, the urea injection amount u(k) at the current moment can be determined through the control algorithm of the PID controller, that is, through the following formula (10), and urea is injected according to the urea injection amount.

[0065] u(k) = u(k-1) + k p [e(k)-e(k-1)]+k i e(k)+k d[e(k)-2e(k-1)+e(k-2)] (10)

[0066] Where, k p k i k d These represent the proportional, integral, and derivative parameters of the PID controller, respectively; u(k-1) is the urea injection rate at time k-1; and e(k) is the NO injection rate at time k. X Conversion error, e(k-1) is NO at time k-1 X Conversion error, e(k-2) is NO at time k-2. X Conversion error.

[0067] The technical solution provided in this application embodiment obtains the NO value of the marine diesel engine SCR system at the previous moment. X Target conversion rate, NO X Actual conversion rate, NO X Conversion error and control parameter values; among which, the control parameters include the proportional, integral, and derivative parameters of the PID controller in the marine diesel engine SCR system; based on the NO value at the previous moment. X Target conversion rate, NO X Actual conversion rate, NO X Conversion error and control parameter values updating The network parameter values ​​of the sliding mode adaptive controller, which includes a sliding mode equivalent control part and a sliding mode switching control part, wherein the sliding mode equivalent control part is implemented through a first neural network and the sliding mode switching control part is implemented through a second neural network; based on the NO value at the previous time step. X Target conversion rate, NO X Actual conversion rate and NO X The conversion error is addressed by using an updated sliding mode adaptive controller to determine the control parameters of the marine diesel engine's SCR system at the current moment. Based on these control parameters, the urea injection quantity is controlled by a PID controller. In other words, the sliding mode equivalent control function in the sliding mode adaptive controller is implemented through a first neural network, and the sliding mode switching control function is implemented through a second neural network. Furthermore, the network parameter values ​​of the sliding mode adaptive controller are adaptively updated in conjunction with the urea injection quantity, reducing the impact of interference factors on the control process. This results in higher accuracy of the determined PID controller control parameters at the current moment, thereby improving the control precision of the urea injection quantity and ultimately increasing NO₂ levels. X The conversion rate meets the prescribed emission reduction requirements.

[0068] On the basis of the above embodiment, optionally, before S101, the method further comprises: determining an initial network parameter value of the sliding mode adaptive controller; and initializing the sliding mode adaptive controller according to the initial network parameter value.

[0069] Specifically, the initial network parameter value of the sliding mode adaptive controller can be determined by using a particle swarm algorithm or a genetic algorithm. Optionally, as shown in Figure 3 the initial network parameter value of the sliding mode adaptive controller can be determined by using the following process:

[0070] S301, initialize a population.

[0071] The number of particles in the population, the convergence condition of the algorithm (for example, the maximum number of iterations), the learning rate of a particle individual, the learning rate of the population, the speed and position of each particle can be initialized. The position of each particle in the population represents the initial weight value of the neural network to be trained, and the number of weights that play a connecting role in the neural network determines the dimension of each particle.

[0072] S302, determine the current fitness value of each particle.

[0073] Specifically, the current fitness value of each particle is determined by a predetermined fitness function. The fitness function J can be represented by the following formula (11):

[0074]

[0075] where N represents the number of samples in the training set, represents the best output value of the nth sample, and u n is the final output value of the nth sample.

[0076] S303, based on the current fitness value of each particle, determine the historical optimal value of each particle and the global optimal value of the population.

[0077] In the evolution of each generation, the current fitness value of each particle is calculated by the above fitness function. If the current fitness value is better than the historical optimal fitness value of the particle, the historical optimal value (i.e. the historical optimal position) of the particle is updated, that is, the position corresponding to the current fitness value is taken as the historical optimal value of the particle. If the current fitness value is better than the historical optimal fitness value of the population, the global optimal value (i.e. the global optimal position) of the population is updated, that is, the position corresponding to the current fitness value is taken as the global optimal value of the population.

[0078] S304, according to the fitness variance of the population, mutate the global optimal value of the population to obtain the mutated global optimal value of the population.

[0079] Specifically, in the process of determining the initial parameters of the first neural network and the second neural network, the particle swarm algorithm is prone to fall into a local optimal solution. Based on this, the global optimal value of the population can be mutated based on the fitness variance of the population.

[0080] Optionally, the process of S304 can be: determining the fitness variance of the population; determining the population mutation probability according to the fitness variance of the population, and generating a random number; wherein the random number obeys a normal distribution; when the random number is greater than the population mutation probability, determining the mutation global optimal value of the population according to the random number and the global optimal value of the population.

[0081] Specifically, the fitness variance D of the population can be determined by the following formula (12):

[0082]

[0083] Wherein, A is the number of particles in the population, f i is the current fitness value of the i-th particle, f avg is the average fitness value of the population, and f is a normalization factor for controlling the range of the fitness variance.

[0084] Then, after obtaining the fitness variance of the population, the population mutation probability p m can be determined by the following formula (13):

[0085]

[0086] Wherein, q is a random number in the interval [0, 0.4], and σ can be set based on actual needs.

[0087] Further, a random number r obeying a normal distribution is randomly generated;

[0088] The random number r is compared with the population mutation probability determined above. If the random number r is greater than the population mutation probability, the mutation global optimal value g of the population is determined according to the random number r and the global optimal value of the population; if the random number r is less than or equal to the population mutation probability, the global optimal value of the population is not mutated. Specifically, the mutation global optimal value g best of the population can be determined by the following formula (14): best

[0089]

[0090] S305, determining the weight of each particle.

[0091] Wherein, the weight of the particle can be an inertia weight, or the weight of each particle can be dynamically updated based on the current fitness value of the particle. ​

[0092] In an embodiment, the process of S305 can be determining the average fitness value and the minimum fitness value of the population, and determining the weight of each particle according to the current fitness value of each particle, the average fitness value and the minimum fitness value of the population.

[0093] Specifically, the weight ω of each particle can be determined based on the following formula (15): i

[0094]

[0095] wherein ωmax is the maximum value of the weight, ωmin is the minimum value of the weight, ωmax can be set as 0.9, ωmin can be set as 0.4, f is the current fitness value of the i-th particle, f is the average fitness value of the population, f is the minimum fitness value of the population. max min max min i avg g

[0096] S306, updating the speed and position of each particle based on the weight of each particle, the historical optimal value of each particle and the mutation global optimal value of the population.

[0097] Specifically, the speed of each particle can be updated based on the following formula (16), and the position of each particle can be updated based on the following formula (17):

[0098]

[0099]

[0100] wherein c1 and c2 are learning factors, r1 and r2 are random numbers in the interval [0, 1], y is the evolution generation, m is the population size, d is the spatial dimension of the problem to be solved, i is in the interval [1, m], j is in the interval [1, d], p is the historical optimal value of the particle in the k-th iteration, g is the mutation global optimal value of the population. best,j best,j

[0101] S307, determining whether a preset convergence condition is met.

[0102] If not, the step of determining the current fitness value of each particle in S302 is continued to be executed until the preset convergence condition is reached, so as to obtain the initial parameters of the first neural network and the second neural network. If yes, S308 is executed.

[0103] S308, outputting the initial network parameter value of the sliding mode adaptive controller. ​​​​​​​​​​

[0104] In the embodiment, by adaptively adjusting the weight of the particle and mutating the global optimal value of the population, the algorithm is prevented from falling into a local optimal solution, the accuracy of the initial network parameter value of the sliding mode adaptive controller is improved, and the control precision of the urea injection amount is further improved.

[0105] For the convenience of those skilled in the art, the following describes the control process of the urea injection amount of the marine diesel engine SCR system with reference to the control device shown in Figure 4 As shown in Figure 4 The control device can include an initial network parameter value determination module of the sliding mode adaptive controller (i.e., A-PSO (adaptive particle swarm optimization) optimization of the initial weight value in Figure 4 , a sliding mode adaptive controller based on a BP neural network (the sliding mode adaptive controller includes a sliding mode equivalent control part and a sliding mode switching control part, the sliding mode equivalent control part is implemented through a first neural network, and the sliding mode switching control part is implemented through a second neural network), and a PID controller.

[0106] The initial network parameter value of the sliding mode adaptive controller is obtained by offline training of the sliding mode adaptive controller based on the BP neural network through the adaptive particle swarm optimization algorithm; then, the trained initial network parameter value is substituted into the sliding mode adaptive controller based on the BP neural network for online control, that is, the control parameter value (i.e., the proportional parameter value, the integral parameter value, and the differential parameter value) of the PID controller is optimized through the sliding mode adaptive controller based on the BP neural network, and is output to the PID controller, and the PID controller uses the above control parameter value to realize the injection of the urea injection amount. Further, the network parameter value of the sliding mode adaptive controller based on the BP neural network can be adaptively updated in combination with the data (i.e., the NO X target conversion rate, the NO X actual conversion rate, the NO X conversion error, and the control parameter value) in the present control process, and as time goes on, the PID controller control parameter value output by the sliding mode adaptive control module based on the BP neural network is more and more accurate, thereby improving the control precision of the urea injection amount.

[0107] Figure 5 Another structural schematic diagram of the urea injection amount control device of the marine diesel engine SCR system provided by the embodiment of the present application is shown in Figure 5 As shown in

[0108] Specifically, the acquisition module 501 is configured to acquire the NO X target conversion rate, the NO Xactual conversion rate, NO X conversion error and control parameter value; wherein, the control parameter comprises a proportional parameter, an integral parameter and a differential parameter of a PID controller in the marine diesel engine SCR system;

[0109] The updating module 502 is configured to update the network parameter value of the sliding mode adaptive controller according to the NO X target conversion rate, NO X actual conversion rate, NO X conversion error and control parameter value; wherein, the control parameter comprises a proportional parameter, an integral parameter and a differential parameter of a PID controller in the marine diesel engine SCR system;

[0110] The first determining module 503 is configured to determine the NO X target conversion rate, NO X actual conversion rate, NO X conversion error, through the updated sliding mode adaptive controller, to determine the control parameter value of the marine diesel engine SCR system at the current time.

[0111] The control module 504 is configured to control the urea injection amount through the PID controller according to the control parameter value at the current time.

[0112] Optionally, on the basis of the above embodiment, the device further comprises a second determining module and an initialization module.

[0113] Specifically, the second determining module is configured to determine the initial network parameter value of the sliding mode adaptive controller before the marine diesel engine SCR system at the previous time is acquired NO X target conversion rate, NO X actual conversion rate, NO X conversion error and control parameter value;

[0114] The initialization module is configured to initialize the sliding mode adaptive controller according to the initial network parameter value.

[0115] Based on the above embodiments, optionally, the second determining module is specifically used to initialize the population; determine the current fitness value of each particle; determine the historical best value of each particle and the global best value of the population based on the current fitness value of each particle; mutate the global best value of the population according to the fitness variance of the population to obtain the mutated global best value of the population; determine the weight of each particle; update the velocity and position of each particle based on the weight of each particle, the historical best value of each particle and the mutated global best value of the population; continue to execute the step of determining the current fitness value of each particle until the preset convergence condition is reached to obtain the initial network parameter values ​​of the sliding mode adaptive controller.

[0116] Based on the above embodiments, optionally, the second determining module is further specifically used to determine the fitness variance of the population; determine the population mutation probability based on the fitness variance of the population, and generate a random number; when the random number is greater than the population mutation probability, determine the global optimum of the population mutation based on the random number and the global optimum of the population; wherein the random number follows a normal distribution.

[0117] Based on the above embodiments, optionally, the second determining module is further specifically used to determine the average fitness value and minimum fitness value of the population; and to determine the weight of each particle based on the current fitness value of each particle, the average fitness value of the population, and the minimum fitness value.

[0118] Optionally, based on the above embodiments, the first neural network and the second neural network are BP neural networks.

[0119] In one embodiment, an electronic device, such as Figure 6 As shown, the electronic device may include a processor 601, a memory 602, an input device 603, and an output device 604; the number of processors 601 in the electronic device may be one or more. Figure 6 Taking a processor 601 as an example; the processor 601, memory 602, input device 603, and output device 604 in the electronic device can be connected via a bus or other means. Figure 6 Taking the example of a connection between China and Israel via a bus.

[0120] The memory 602 can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the urea injection amount control method of the marine diesel engine SCR system in the embodiments of the present application (for example, the obtaining module 501, the updating module 502, the first determining module 503 and the control module 504 in the urea injection amount control device of the marine diesel engine SCR system). The processor 601 executes various functions and data processing of the electronic device by running the software programs, instructions and modules stored in the memory 602, that is, implements the urea injection amount control method of the marine diesel engine SCR system as described above.

[0121] The memory 602 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and application programs required by at least one function; the data storage area can store data created according to the use of the electronic device, etc. In addition, the memory 602 can include a high-speed random access memory, and can also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state memory device. In some examples, the memory 602 can further include a memory remotely arranged with respect to the processor 601, which can be connected to the device / terminal / server through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0122] The input device 603 can be used to receive input digital or character information, and generate key signal input related to user settings and function control of the electronic device. The output device 604 can include a display device such as a display screen.

[0123] In one embodiment, a computer readable storage medium is also provided, which stores a computer program. The computer program is executed by a processor to perform a urea injection amount control method of a marine diesel engine SCR system, which includes:

[0124] obtaining NOx conversion error and control parameter value; wherein the control parameter includes proportional parameter, integral parameter and differential parameter of a PID controller in the marine diesel engine SCR system; X target conversion rate, NOx conversion error and control parameter value; wherein the control parameter includes proportional parameter, integral parameter and differential parameter of a PID controller in the marine diesel engine SCR system; X actual conversion rate, NOx conversion error and control parameter value; wherein the control parameter includes proportional parameter, integral parameter and differential parameter of a PID controller in the marine diesel engine SCR system; X target conversion rate, NOx conversion error and control parameter value; wherein the control parameter includes proportional parameter, integral parameter and differential parameter of a PID controller in the marine diesel engine SCR system;

[0125] obtaining NOx conversion error and control parameter value; wherein the control parameter includes proportional parameter, integral parameter and differential parameter of a PID controller in the marine diesel engine SCR system; X target conversion rate, NOx conversion error and control parameter value; wherein the control parameter includes proportional parameter, integral parameter and differential parameter of a PID controller in the marine diesel engine SCR system; X actual conversion rate, NOx conversion error and control parameter value; wherein the control parameter includes proportional parameter, integral parameter and differential parameter of a PID controller in the marine diesel engine SCR system; Xupdating the network parameter value of the sliding mode adaptive controller according to the conversion error and the control parameter value; wherein the sliding mode adaptive controller comprises a sliding mode equivalent control part and a sliding mode switching control part, the sliding mode equivalent control part is implemented by a first neural network, and the sliding mode switching control part is implemented by a second neural network; the NO X the target conversion rate, the NO X the actual conversion rate, and the NO X the conversion error, determining the control parameter value of the SCR system of the marine diesel engine at the current time through the updated sliding mode adaptive controller;

[0126] controlling the urea injection amount through the PID controller according to the control parameter value at the current time.

[0127] Of course, the computer readable storage medium provided by the embodiment of the application is not limited to the method operations as described above when the computer program is executed, and can also perform related operations in the urea injection amount control method of the SCR system of the marine diesel engine provided by any embodiment of the application.

[0128] Through the above description of the embodiments, those skilled in the art can clearly understand that the application can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the application can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk or an optical disk, etc., including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the application.

[0129] It is worth noting that in the above embodiment of the search device, each unit and module included is only divided according to functional logic, but is not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy distinction, and do not limit the protection scope of the application.

[0130] It is to be noted that the above-mentioned embodiments illustrate rather than limit the application, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the application. The word "comprising" does not exclude the presence of elements or steps other than those listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. It is further noted that characteristics relating to the different embodiments can be combined, and not just those within respective sections of the description.

Claims

1. A method of controlling the amount of urea injection for a marine diesel engine SCR system, characterized by, The method comprises the following steps: acquiring NOx conversion rate of the ship diesel engine SCR system at a previous time X target conversion rate, NOx X actual conversion rate, NOx X conversion error and control parameter value; wherein the control parameter comprises a proportional parameter, an integral parameter and a differential parameter of a proportional-integral-differential (PID) controller in the ship diesel engine SCR system NO X target conversion rate, NO X actual conversion rate, NO X conversion error and control parameter value update network parameter value of a sliding mode adaptive controller; wherein the sliding mode adaptive controller comprises a sliding mode equivalent control part and a sliding mode switching control part, the sliding mode equivalent control part is realized by a first neural network, and the sliding mode switching control part is realized by a second neural network; According to the NO of the last time X Target conversion rate, NO X Actual conversion rate and NO X Conversion error, through the updated sliding mode adaptive controller, determine the control parameter value of the ship diesel engine SCR system at the current time controlling the urea injection amount by the PID controller according to the control parameter value at the current time; acquiring NOx conversion rate of the SCR system of the ship diesel engine at a previous time point X target conversion rate, NOx conversion rate of the SCR system of the ship diesel engine at the previous time point X actual conversion rate, NOx conversion rate of the SCR system of the ship diesel engine at the previous time point X conversion error and the control parameter value, further comprising: determining an initial network parameter value of the sliding mode adaptive controller; initializing the sliding mode adaptive controller according to the initial network parameter value; the step of determining the initial network parameter value of the sliding mode adaptive controller comprises: initializing a population; determining a current fitness value of each particle; determining a historical optimal value of each particle and a global optimal value of the population based on the current fitness value of each particle; mutating the global optimal value of the population according to a fitness variance of the population to obtain a mutated global optimal value of the population; determining a weight of each particle, wherein the weight of each particle is an inertia weight or a weight of each particle dynamically updated based on the current fitness value of each particle; updating a speed and a position of each particle based on the weight of each particle, the historical optimal value of each particle and the mutated global optimal value of the population; continuing to perform the step of determining the current fitness value of each particle until a preset convergence condition is reached to obtain the initial network parameter value of the sliding mode adaptive controller; the step of updating the network parameter value of the sliding mode adaptive controller according to the NOx target conversion rate, the NOx actual conversion rate, the NOx conversion error and the control parameter value at the previous time comprises: In training the first neural network in the sliding mode adaptive controller, the NO X target conversion rate, NO X actual conversion rate, NO X The first neural network in the sliding mode adaptive controller is trained with the conversion error as input and the sliding mode equivalent control amount at the last time as expected output. In training the second neural network in the sliding mode adaptive controller, the NO X target conversion rate, NO X actual conversion rate, NO X conversion error as the input of the second neural network and the sliding mode switching control amount of the last time as the expected output, the second neural network in the sliding mode adaptive controller is trained; wherein the sliding mode equivalent control amount at the previous time and the sliding mode switching control amount at the previous time are determined based on the control parameter value of the PID controller at the previous time in the SCR system of the marine diesel engine.

2. The method of claim 1, wherein, the step of mutating the global optimal value of the population according to the fitness variance of the population to obtain the mutated global optimal value of the population comprises: determining the fitness variance of the population; determining a population mutation probability and generating a random number according to the fitness variance of the population, wherein the random number follows a normal distribution; when the random number is greater than the population mutation probability, determining the mutated global optimal value of the population according to the random number and the global optimal value of the population.

3. The method of claim 1, wherein, the step of determining the weight of each particle comprises: determining an average fitness value and a minimum fitness value of the population; determining the weight of each particle according to the current fitness value of each particle, the average fitness value and the minimum fitness value of the population.

4. The method according to any one of claims 1 to 3, characterized in that, The types of the first neural network and the second neural network are back propagation (BP) neural networks.

5. A marine diesel engine SCR system urea injection amount control device characterized by comprising: The method comprises the following steps: The acquisition module is configured to acquire NOx conversion rate of the SCR system of the marine diesel engine at a previous time point X target conversion rate, NOx X actual conversion rate, NOx X conversion error and a control parameter value; wherein the control parameter comprises a proportional parameter, an integral parameter and a differential parameter of a PID controller in the SCR system of the marine diesel engine. an updating module configured to update the network parameter values of the sliding mode adaptive controller according to the NO X a target conversion rate, NO X an actual conversion rate, NO X a conversion error and control parameter value updating sliding mode adaptive controller; wherein the sliding mode adaptive controller comprises a sliding mode equivalent control part and a sliding mode switching control part, the sliding mode equivalent control part is implemented by a first neural network, and the sliding mode switching control part is implemented by a second neural network. a first determining module configured to determine a target conversion rate of NOx at a current time according to a NOx conversion rate at a previous time and a target conversion rate of NOx at the previous time X a target conversion rate of NOx, a NOx conversion rate at the previous time X a NOx conversion rate at the previous time and a NOx conversion rate at the current time X a conversion error, and determine a control parameter value of the SCR system of the marine diesel engine at the current time through the updated sliding mode adaptive controller controlling the urea injection amount by the PID controller according to the control parameter value at the current time; A second determining module is configured to determine NOx conversion rate of the SCR system of the ship diesel engine at a previous time point according to the first determining module. X Target conversion rate, NOx X Actual conversion rate, NOx X Before determining the conversion error and the control parameter value, initial network parameter values of the sliding mode adaptive controller are determined. initializing the sliding mode adaptive controller according to the initial network parameter value; The second determining module is specifically configured to initialize a population; determine current fitness values of particles; determine historical optimal values of the particles and a global optimal value of the population based on the current fitness values of the particles; mutate the global optimal value of the population according to a fitness variance of the population, to obtain a mutated global optimal value of the population; determine weights of the particles, wherein the weights of the particles are inertia weights or weights of the particles dynamically updated based on the current fitness values of the particles; update speeds and positions of the particles based on the weights of the particles, the historical optimal values of the particles and the mutated global optimal value of the population; continue to perform the step of determining the current fitness values of the particles until a preset convergence condition is reached. The updating module is specifically configured to: when training the first neural network in the sliding mode adaptive controller, taking the NOx target conversion rate, the NOx actual conversion rate and the NOx conversion error at the last moment as inputs of the first neural network and taking the sliding mode equivalent control amount at the last moment as expected output, training the first neural network in the sliding mode adaptive controller; when training the second neural network in the sliding mode adaptive controller, taking the NOx target conversion rate, the NOx actual conversion rate and the NOx conversion error at the last moment as inputs of the second neural network and taking the sliding mode switching control amount at the last moment as expected output, training the second neural network in the sliding mode adaptive controller; wherein the sliding mode equivalent control amount at the last moment and the sliding mode switching control amount at the last moment are determined based on a control parameter value of a PID controller in the SCR system of the marine diesel engine at the last moment.

6. An electronic device, comprising: comprising: a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the method according to any one of claims 1 to 4 when executing the computer program.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, the computer program, when executed by the processor, implements the steps of the method according to any one of claims 1 to 4.

Images