Valve coefficient optimization calculation method, device and equipment and storage medium

Abstract

The application provides a valve coefficient optimization calculation method, device and equipment and a storage medium. The method comprises the following steps: obtaining initial valve coefficients of valves in a dust removal system, an optimization strategy and a dust removal pipe network simulation model corresponding to the dust removal system, wherein the optimization strategy at least comprises a target function and a constraint condition; calculating the initial valve coefficients of the valves by using the target function and the constraint condition to obtain global optimization valve coefficients of the valves; inputting the global optimization valve coefficients into the dust removal pipe network simulation model for simulation calculation to obtain a simulation result set of the valves; constructing a machine learning model and training the machine learning model according to the simulation result set, so as to adjust valve openings of the valves by using the trained machine learning model. The initial valve coefficients are optimized, the accuracy of simulation calculation is improved, and the regulation and control effect of the dust removal system is ensured.

Description

Technical Field

[0001] This invention relates to the field of optimization calculation technology, specifically to a valve coefficient optimization calculation method, apparatus, equipment, and storage medium. Background Technology

[0002] Dust collection systems, generally consisting of dust collectors and fans, are widely used in various industries, such as the steel industry and chemical enterprises. The dust collection pipeline network, a crucial component of the system, mainly comprises regulating valves and dust collection pipes. Its complex design connects multiple dust collection points, each with a regulating valve to control the airflow. However, because different dust collection points require different amounts of dust and therefore different airflow allocations, and because these points influence each other, it's difficult to dynamically adjust the regulating valves based on actual dust conditions. Insufficient airflow allocation at each dust collection point leads to dust dispersion; excessive airflow allocation results in the suction of useful materials. Related technologies primarily address this by adjusting the valve coefficients of the regulating valves to regulate the fluid flow or pressure in the dust collection pipeline, thereby controlling the valve opening and achieving a reasonable airflow allocation at each dust collection point. Furthermore, the dust collection pipeline network is modeled and simulated, and the valves are dynamically controlled based on the simulation results.

[0003] However, with changes in production stages and equipment aging, valve coefficients can deviate from design values, causing simulation results to detach from actual operating conditions. This results in excessive errors between measured values ​​from instruments and their corresponding calculated values ​​in the simulation model, leading to low accuracy in the simulation model and poor dynamic adjustment of the dust removal system. Furthermore, when the airflow demand at the dust collection point changes, the valves cannot respond quickly to the changed airflow, failing to automatically adjust and correct themselves in a timely manner, resulting in low efficiency and high energy consumption in the dust removal system. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the present invention provides a valve coefficient optimization calculation method, apparatus, equipment and storage medium to solve at least one of the above-mentioned technical problems.

[0005] In a first aspect, the present invention provides a method for optimizing the calculation of valve coefficients, comprising:

[0006] Obtain the initial valve coefficients, optimization strategies, and simulation models of the dust removal pipeline network corresponding to the dust removal system for each valve in the dust removal system. The optimization strategy includes at least an objective function and constraints.

[0007] The initial valve coefficients of each valve are calculated using the objective function and the constraints to obtain the global optimized valve coefficients of each valve.

[0008] The globally optimized valve coefficients are input into the dust removal pipeline simulation model for simulation calculation, and the simulation result set of each valve is obtained.

[0009] A machine learning model is constructed and trained based on the simulation result set, so as to adjust the valve opening of each valve through the trained machine learning model.

[0010] In one embodiment of the present invention, before calculating the initial valve coefficients of each of the valves using the objective function and the constraints, the method further includes:

[0011] Obtain the simulated pressure and measured pressure at each dust removal point under each working condition;

[0012] The pressure error of the dust removal point is obtained by calculating the difference between the measured pressure and the corresponding simulated pressure.

[0013] The pressure error at each dust removal point is weighted and the objective function is determined.

[0014] In one embodiment of the present invention, obtaining the simulated pressure of each dust removal point under each operating condition includes:

[0015] Obtain the measured pressure and simulated pressure at each dust removal point under each working condition;

[0016] The pressure error of the dust removal point is obtained by calculating the difference between the measured pressure and the corresponding simulated pressure.

[0017] The pressure error at each dust removal point is weighted and the objective function is determined.

[0018] In one embodiment of the present invention, obtaining the simulated pressure of each dust removal point under each operating condition includes:

[0019] Obtain the historical operating data of the dust removal system;

[0020] A set of working conditions including multiple working conditions is generated based on the historical working data;

[0021] Hydraulic calculations are performed on the initial valve coefficients of each valve based on the set of operating conditions and the simulation model of the dust removal pipeline network to obtain the simulated pressure of each dust removal point under each operating condition. The valves and dust removal points correspond one-to-one.

[0022] In one embodiment of the present invention, the constraint conditions include an upper limit value of the initial valve coefficient and a lower limit value of the initial valve coefficient;

[0023] The step of calculating the initial valve coefficients of each valve using the constraints and the objective function to obtain the globally optimized valve coefficients of each valve includes:

[0024] Each valve is combined with different opening intervals of other valves to form an array matrix. Each array matrix represents a combination of valves in any opening interval, and the combinations represented by each array are not repeated. The initial valve coefficient is determined by the opening interval corresponding to each valve in the array.

[0025] Calculate the cumulative value of the pressure error of each dust removal point corresponding to each array, and take the initial valve coefficient corresponding to the smallest cumulative value as the global optimized valve coefficient.

[0026] In one embodiment of the present invention, the step of inputting the globally optimized valve coefficients of each valve into the dust removal pipeline simulation model for simulation calculation to obtain a simulation result set includes:

[0027] The valve opening degree of each valve within any combination of opening degree intervals;

[0028] The global optimized valve coefficient corresponding to the combination of the valve opening degree and the opening degree range of each valve is input into the dust removal pipeline simulation model for hydraulic calculation to obtain the simulation optimization results of each valve. The simulation optimization results include at least the simulation flow rate of each dust removal point and the optimized simulation pressure of each dust removal point.

[0029] The simulated flow rate of each dust removal point, the optimized simulated pressure of each dust removal point, and the valve opening of each valve are aggregated to obtain a simulation result set.

[0030] In one embodiment of the present invention, the step of constructing a machine learning model and training the machine learning model based on the simulation result set to adjust the valve opening of each valve through the trained machine learning model includes:

[0031] The simulation result set is labeled to form a sample dataset;

[0032] The machine learning model is trained based on the sample dataset;

[0033] Obtain the target flow rate for each of the dust removal points;

[0034] The target flow rate of each dust removal point is input into the trained machine learning model, so that the machine learning model outputs the target valve opening of each valve;

[0035] Adjust the current valve opening of each valve to the target valve opening.

[0036] In one embodiment of the present invention, before obtaining the initial valve coefficients of each valve in the dust removal system, the method further includes:

[0037] A valve model is constructed based on the valve type of the dust removal system to obtain the valve coefficient of each valve;

[0038] The valve coefficient is divided according to the different opening intervals of the valve opening to obtain the valve coefficient to be optimized corresponding to the opening interval;

[0039] The valve coefficients to be optimized are initialized to determine the initial valve coefficients.

[0040] In a second aspect, the present invention also provides a valve coefficient optimization calculation device, comprising:

[0041] The acquisition module is configured to acquire the initial valve coefficients, optimization strategies, and simulation models of the dust removal pipeline network corresponding to the dust removal system for each valve in the dust removal system. The optimization strategies include at least an objective function and constraints.

[0042] The optimization module is configured to calculate the initial valve coefficients of each valve using the objective function and the constraints, thereby obtaining the global optimized valve coefficients of each valve.

[0043] The simulation module is configured to input the globally optimized valve coefficients into the dust removal pipeline simulation model for simulation calculation, and obtain the simulation result set of each valve.

[0044] The training module is configured to build a machine learning model and train the machine learning model based on the simulation result set, so as to adjust the valve opening of each valve through the trained machine learning model.

[0045] In a third aspect, the present invention also provides an electronic device, comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the valve coefficient optimization calculation method as described in the above embodiments.

[0046] In a fourth aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a computer processor, causes the computer to perform the valve coefficient optimization calculation method as described in the above embodiments.

[0047] The beneficial effects of this invention are as follows: This invention proposes a valve coefficient optimization calculation method, apparatus, equipment, and storage medium. The method obtains the initial valve coefficients of each valve in a dust removal system, the optimization strategy, and the corresponding dust removal pipeline simulation model. The optimization strategy includes at least an objective function and constraints. The initial valve coefficients of each valve are calculated using the objective function and constraints to obtain the global optimized valve coefficients. These global optimized valve coefficients are then input into the dust removal pipeline simulation model for simulation calculation, resulting in a simulation result set for each valve. A machine learning model is constructed and trained based on the simulation result set to adjust the valve opening of each valve. On the one hand, this ensures the accuracy of the dust removal pipeline simulation calculation, enabling valve control through the simulation result set, resulting in good valve control effect in the dust removal system. On the other hand, the machine learning model allows for automatic and rapid valve control, improving the working efficiency of the dust removal system and thus saving energy and reducing consumption during operation.

[0048] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0049] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0050] Figure 1 This is a schematic diagram of a dust removal network in a steelmaking workshop of a steel company, illustrating an exemplary embodiment of the present invention.

[0051] Figure 2 This is a schematic diagram illustrating an exemplary embodiment of the present invention, showing the dust removal network pipe in a steelmaking workshop of a steel enterprise after the addition of instruments;

[0052] Figure 3 This is a schematic flowchart illustrating a valve coefficient optimization calculation method according to an exemplary embodiment of the present invention;

[0053] Figure 4 yes Figure 3 The flowchart of step S320 in the illustrated embodiment is shown in an exemplary embodiment;

[0054] Figure 5 This is a schematic diagram illustrating the instrument numbering of a dust removal network in a steelmaking workshop of a steel company, as shown in an exemplary embodiment of the present invention.

[0055] Figure 6 This is a schematic diagram illustrating the pressure error of a pressure gauge under multiple operating conditions, as shown in an exemplary embodiment of the present invention.

[0056] Figure 7 This is a schematic diagram illustrating optimization using the particle swarm optimization algorithm, as shown in an exemplary embodiment of the present invention.

[0057] Figure 8 This is a schematic diagram illustrating the pressure error of a pressure gauge under multiple operating conditions after optimized calculation, as shown in an exemplary embodiment of the present invention.

[0058] Figure 9 This is a schematic diagram illustrating the simulation calculation process of an exemplary embodiment of the present invention;

[0059] Figure 10 This is a schematic diagram illustrating machine learning model training as shown in an exemplary embodiment of the present invention;

[0060] Figure 11 This is a schematic diagram of the dynamic control of the dust removal system in one embodiment of the present invention;

[0061] Figure 12 This is a schematic diagram illustrating a business process according to an exemplary embodiment of the present invention;

[0062] Figure 13 This is a block diagram illustrating a valve coefficient optimization calculation device according to an exemplary embodiment of the present invention;

[0063] Figure 14 A schematic diagram of the structure of a computer system suitable for implementing an electronic device according to embodiments of the present invention is shown. Detailed Implementation

[0064] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0065] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0066] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0067] Please see Figure 1 This is a schematic diagram of a dust removal network in a steelmaking workshop of an iron and steel enterprise, illustrating an exemplary embodiment of the present invention. The dust removal system in the iron and steel industry is characterized by its large network, numerous dust collection points, and complex system structure. The dust removal network mainly consists of dust removal pipes and regulating valves. The dust removal pipes include long straight pipes, tee fittings, and elbow fittings. Figure 1 As shown, Figure 1 The dust collection system in this project comprises six dust collection points, all of which converge into a single pipe connecting the filter bags, fan, and chimney. Each dust collection point is equipped with a regulating valve, which controls the airflow at that point. Because each dust collection point is interconnected, adjusting any valve individually will affect the airflow at other points, thus influencing the overall resistance of the dust collection network. Therefore, all valves in the network must be adjusted simultaneously. To better control the airflow at each point, it is essential to understand the flow rate and pressure within the dust collection network. However, the dust collection system lacks instrumentation, operating in a "black box" state—it cannot be opened or its internal state directly observed from the outside. Internal information can only be determined through data input and output. Therefore, flow meters or pressure gauges are necessary to monitor the flow rate and pressure within the dust collection network.

[0068] Please see Figure 2 This is a schematic diagram illustrating the addition of instruments to the dust removal network pipe in a steelmaking workshop of a steel enterprise, as an exemplary embodiment of the present invention. The method of the present invention can be applied to, for example... Figure 2 The dust removal system shown. The dust removal pipe network of this dust removal system is in Figure 1 Based on the dust collection pipeline network shown, instruments are installed after the valves at each dust collection point. These instruments are pressure gauges, six in total. Figure 2 In Chinese, it is represented by "P". Figure 2 The “Q” in the figure represents the flow rate. Since the equipment in the dust removal system is connected in series, the flow rate of the equipment can be regarded as the same, so it can be replaced by the flow rate of the filter bag, fan or chimney.

[0069] Please see Figure 3 The diagram below illustrates a flow chart of a valve coefficient optimization calculation method, as shown in an exemplary embodiment of the present invention. Figure 3As shown, in an exemplary embodiment, the valve coefficient optimization calculation method includes at least steps S310 to S340, which are described in detail below:

[0070] Step S310: Obtain the initial valve coefficients, optimization strategies, and simulation models of the dust removal pipeline network corresponding to each valve in the dust removal system. The optimization strategy shall include at least the objective function and constraints.

[0071] Specifically, before step S310, a simulation model of the dust removal system's network structure is required. First, a hydraulic simulation model is constructed for each component in the dust removal network, such as valves, long straight pipes, tees, and elbows. Then, based on these component simulation models, a dust removal network simulation model is built. Specifically, a valve model is constructed by building a hydraulic simulation model for the valves. It is important to understand that the hydraulic simulation model is used to simulate dynamic changes in the actual production process and predict changes caused by certain influencing factors. This invention mainly focuses on the construction of the valve model and the dust removal network simulation model to facilitate subsequent hydraulic calculations of the dust removal network. The hydraulic calculation method can employ the nodal method to calculate the flow rate and pressure at any dust removal point in the dust removal network.

[0072] In one embodiment of the present invention, before obtaining the initial valve coefficients of each valve, a valve hydraulic calculation formula is first established based on the valve model to obtain the valve coefficients; then, the valve coefficients are divided according to different valve opening ranges, i.e., different valve opening ranges, to obtain the valve coefficients to be optimized corresponding to the opening ranges; finally, the valve coefficients to be optimized are initialized to determine the initial valve parameters. The initial valve coefficients are obtained by initializing the valve coefficients to be optimized. The valve hydraulic calculation formula is expressed as follows:

[0073] ΔP(kl)=q 2 Formula (1)

[0074] In equation (1), ΔP is the valve pressure drop, l is the valve opening, q is the valve flow rate, and k is the valve coefficient. The valve flow rate is equivalent to the flow rate at the dust removal point. The pressure at the dust removal point can be obtained from the valve pressure drop and the valve flow rate. It should be understood that because the valve coefficient is the valve resistance coefficient, and the resistance characteristics of the valve change significantly under different valve openings, the valve coefficient is not a constant, but a value that varies with the valve opening. After dividing the valve coefficient into segments according to the valve opening range, it can be represented by a piecewise function, which is expressed as follows:

[0075]

[0076] In this function, k0 corresponds to a low valve opening, k1 to a medium valve opening, and k2 to a high valve opening, representing three different opening intervals. The piecewise function divides the valve coefficients into three optimization coefficients based on these different opening intervals; each valve has three optimization coefficients. Before optimization calculations, the optimization coefficients need to be initialized to obtain initial valve coefficients. The values ​​of these initial coefficients can be arbitrary, and changes in the initial valve parameters of each valve will affect the change or redistribution of airflow at each dust collection point.

[0077] Step S320: Calculate the initial valve coefficients of each valve using the objective function and constraints to obtain the global optimized valve coefficients of each valve.

[0078] Specifically, before optimizing the initial valve coefficients, it is necessary to define the objective function and constraints. The objective function is used to allocate the weight of the pressure error of each dust removal point according to the importance and control requirements of each dust removal point. The allocation of the objective function weights may be different under different operating conditions. The constraints include the upper limit and lower limit of the initial valve coefficients.

[0079] Please see Figure 4 ,for Figure 3 The flowchart of step S320 in the illustrated embodiment is described in detail below in an exemplary embodiment:

[0080] Step S410: Obtain the simulated pressure and measured pressure of each dust removal point under each working condition.

[0081] Specifically, a pressure gauge is installed after the regulating valve at each dust collection point. The actual pressure at each dust collection point under each operating condition can be measured using the pressure gauge. The simulated pressure of each dust collection point is obtained through the following methods:

[0082] Acquire historical operating data of the dust removal system; generate a set of operating conditions including multiple operating conditions based on the historical operating data; perform hydraulic calculations on the initial valve coefficients of each valve according to the set of operating conditions and the dust removal pipeline simulation model, and obtain the simulation pressure of each dust removal point under each operating condition, with a one-to-one correspondence between valves and dust removal points.

[0083] In one embodiment of the present invention, the operating condition set represents all operating conditions that have occurred in the history of the dust removal system. To statistically analyze all operating conditions, signals such as the production status of the dust removal points, valve opening degree, and fan frequency need to be classified. For example, when a converter steelmaking process involves three dust removal cycles, the production status of the dust removal points includes multiple converter production states, including charging, oxygen blowing, slag splashing, and steel tapping. These production states can be classified into operating conditions. Furthermore, the amount of dust generated at the dust removal points varies throughout the historical production process, requiring different fan frequency settings; operating conditions can also be classified based on these different airflow frequencies. Additionally, the period during which each valve stabilizes within any opening range during the dust removal process can also serve as a basis for operating condition classification. The opening range and stabilization time can be set according to actual conditions; for example, each 5% valve opening can be considered an opening range, and the stabilization time can be set to one minute.

[0084] In one embodiment of the present invention, the historical working data of the dust removal system is used to classify all working conditions related to the production status of the dust removal point, valve opening degree and fan frequency, etc., in the historical working data. These different categories of working conditions are combined, and each combination is regarded as a new working condition, thereby obtaining a set of working conditions. The set of working conditions represents all working conditions that have occurred in the dust removal system in the past period of time.

[0085] In one embodiment of the present invention, for all operating conditions obtained by statistics, the initial valve coefficients are sequentially input into the dust removal pipeline simulation model for hydraulic calculation to obtain the simulation pressure of each dust removal point under each operating condition. Since there is a valve at each dust removal point, the simulation pressure of the corresponding dust removal point can be calculated through the initial valve coefficient of the valve.

[0086] In one embodiment of the present invention, hydraulic calculations can also be performed using a valve model, achieving the same results as the dust removal pipeline simulation model, but with different calculation times. Furthermore, by substituting the initial valve coefficients into the dust removal pipeline simulation model, the flow rate at the dust removal point can also be calculated. The same subsequent calculations can be performed as the pressure at the dust removal point to achieve the same optimization effect. However, in this embodiment, the instrument installed at the dust removal point is a pressure gauge, so flow rate calculations are not performed here.

[0087] Step S420: Calculate the difference between the measured pressure and the corresponding simulated pressure to obtain the pressure error at the dust removal point.

[0088] Specifically, the measured and simulated pressures at all dust collection points under various operating conditions are compared. The difference between the measured and simulated pressures at the same dust collection point is calculated to determine the pressure error at each dust collection point. It is important to note that the simulated pressure is calculated from the initial valve coefficients, which are obtained through random initialization. Therefore, the calculated simulated pressure is inaccurate, leading to excessive pressure errors at each dust collection point. Thus, based on the current pressure error, the initial valve coefficients need to be optimized to calculate the minimum pressure error at each dust collection point, thereby correcting the current pressure error and enabling the dust collection system to precisely control the valves.

[0089] Please see Figure 5 This is a schematic diagram illustrating the instrument numbering of a dust removal network pipe in a steelmaking workshop of a steel company, as an exemplary embodiment of the present invention. Figure 5 In China, Figure 2 The pressure gauges shown are numbered sequentially as Pressure Gauge A, Pressure Gauge B, Pressure Gauge C, Pressure Gauge D, Pressure Gauge E, and Pressure Gauge F. For example... Figure 5 As shown, each dust collection point and its regulating valve are interconnected through a dust collection pipeline network and use the same fan. This means that changes in a single valve not only affect the dust collection point corresponding to that valve but also impact other dust collection points. Therefore, when optimizing the initial valve coefficients, it is not sufficient to consider only individual valves; all valves must be considered for global optimization. All dust collection points should be treated as a whole, and the initial valve coefficients of all valves should be optimized simultaneously to obtain the optimal calculation layout.

[0090] Please see Figure 6 This is a schematic diagram illustrating the pressure error of a pressure gauge under multiple operating conditions, as shown in an exemplary embodiment of the present invention. Figure 5 As shown, the horizontal axis represents the operating conditions, and the vertical axis represents the pressure. The solid line represents the measured pressure at the dust removal point obtained by pressure gauge A under various operating conditions, and the dashed line represents the simulated pressure at the same dust removal point. Figure 6 It can be seen that the pressure error is relatively large under various operating conditions at this time.

[0091] Step S430: Weight the pressure error of each dust removal point and determine the objective function.

[0092] Specifically, after weighting the pressure errors at each dust removal point, the objective function can be expressed by the following formula:

[0093]

[0094] In equation (2), j represents the j-th dust removal point, and w j This represents the weight of the j-th dust removal point. This represents the measured pressure at the j-th dust collection point. denoted as the simulated pressure at the j-th dust removal point, and min is used to calculate the minimum cumulative value of the pressure error at each dust removal point under each working condition.

[0095] Specifically, based on the objective function, the cumulative value of the pressure error of each valve under all operating conditions can be expressed by the following formula:

[0096]

[0097] In equation (3), i represents the i-th working condition, j represents the j-th dust removal point, and w j,i This represents the weight of the j-th dust removal point under the i-th operating condition. This represents the measured pressure at the j-th dust removal point under the i-th operating condition. denoted as the simulated pressure at the j-th dust removal point under the i-th operating condition, and min is used to calculate the cumulative value of the minimum valve pressure error under all operating conditions.

[0098] Continue to refer to Figure 5 In one embodiment of the present invention, for Figure 5 After weighting the pressure errors at each dust removal point, the objective function is expressed by the following formula:

[0099]

[0100] Among them, in equation (4) to This represents the measured pressure from pressure gauge A to pressure gauge F. to This represents the simulated pressure at the dust removal point corresponding to pressure gauges A through F, w A to w F This represents the weight of the dust collection points corresponding to pressure gauges A through F. It's important to note that the weighting relationship between each dust collection point is expressed by the following formula:

[0101] w A +w B +w C +w D +w E +w F =1 Formula (5)

[0102] In equation (5), wA to wF represent the weights of the dust removal points corresponding to pressure gauges A to F.

[0103] Continue to refer to Figure 5 As shown, in one embodiment of the present invention, it can be based on Figure 5The pipe diameter and valve type of the dust collector network can determine the maximum pipe resistance coefficient, which can be used as a constraint on the initial valve coefficient. For example, when the maximum pipe resistance is between 0 and 25, the constraints include a minimum limit of 0 and a maximum limit of 25.

[0104] In one embodiment of the present invention, after determining the objective function and constraints, the method further includes: combining different opening intervals of each valve with different opening intervals of other valves to form an array matrix, wherein each array matrix represents a combination of valves in any opening interval, and the combination represented by each array is unique; the initial valve coefficient is determined by the opening interval corresponding to each valve in the array; calculating the cumulative value of the pressure error of each dust removal point corresponding to each array, and taking the initial valve coefficient corresponding to the smallest cumulative value as the global optimized valve coefficient.

[0105] Specifically, each valve has three different opening ranges: low, medium, and high. The initial valve coefficients corresponding to these different opening ranges are different. The combinations of opening ranges among the valves are represented as a data matrix. Each array represents a different combination of opening ranges, meaning the combination of initial valve coefficients for each valve in each array is different. Each element in the array represents the opening range of each valve, i.e., the initial valve coefficient of each valve. The pressure error at each dust collection point corresponding to each array element is calculated. The minimum pressure error at each dust collection point is summed; that is, the cumulative pressure error at each dust collection point corresponding to each array is minimized. The initial valve coefficient corresponding to the valve with the minimum pressure error is then used as the globally optimized valve coefficient. For example, assuming the dust collection system has two valves, valve A and valve B, the array matrix would be represented as follows:

[0106]

[0107] Among them, array A in the array matrix K0 B K0 For example, the array element at this time is A. K0 and B K0 Let and represent the opening range of valve A and valve B, respectively, both within the range of k0. Valve A corresponds to dust removal point A, and valve B corresponds to dust removal point B. The minimum pressure error at dust removal point A and the minimum pressure error at dust removal point B are calculated separately. The minimum pressure errors at dust removal points A and B are then added together to obtain array A. K0 B K0 The cumulative value of the pressure error at each corresponding dust removal point is the minimum cumulative value. The initial valve coefficients of valve A and valve b corresponding to the minimum cumulative value are taken as the global optimized valve coefficients. The global optimized valve coefficients are the global optimized valve coefficients corresponding to valve A and valve B in the opening range k0.

[0108] In one embodiment of the present invention, after defining the objective function and constraints, an intelligent optimization algorithm is used to optimize the initial valve coefficients of all valves in the dust removal pipeline system. The initial valve coefficients are variables in the optimization calculation; by changing the values ​​of the initial valve coefficients, the pressure error can ultimately be reduced. The initial valve coefficients of each valve participate in the optimization process to achieve global optimization. The optimization process involves continuously changing the value of each initial valve coefficient until the cumulative value of the minimum pressure error is found, and the value of the initial valve coefficient at this point is taken as the globally optimized valve coefficient. The intelligent optimization algorithm can be particle swarm optimization, genetic algorithm, annealing algorithm, etc., and this method does not restrict the intelligent optimization algorithm used. In this embodiment, by simultaneously optimizing the initial valve coefficients of each valve in the entire dust removal system, the problem of complex coupling relationships and mutual influence between dust removal points is solved.

[0109] Please see Figure 7 This is a schematic diagram illustrating optimization using a particle swarm optimization algorithm, as an exemplary embodiment of the present invention. Figure 7 As shown, if the intelligent optimization algorithm used is particle swarm optimization, then... Figure 5 Taking six valves as an example, valves A to F correspond to pressure gauges A to F. Each valve has three initial valve coefficients, corresponding to different opening ranges: low, medium, and high. Therefore, global optimization of these 18 initial valve coefficients is required. Each particle consists of 18 initial valve coefficients, and each generation consists of m particles, where m can take any value depending on the required computation speed and time. First, the first generation of particles is initialized, resulting in m particles. The pressure error corresponding to the initial valve coefficient in each particle is calculated. Then, half of the particles with the largest pressure error are eliminated, and the initial valve coefficient values ​​in the other half with smaller pressure errors are changed to form a new generation. The calculation steps of the first generation of particles are iterated until the pressure error meets the requirements or the number of iterations reaches the upper limit. The initial valve coefficient at this point is used as the global optimization valve coefficient. It is important to note that the initial valve coefficients must meet constraints during optimization calculations, i.e., they can only be changed between a minimum limit of 0 and a maximum limit of 25. After optimization, the pressure error can be significantly reduced, the accuracy of simulation calculation is improved, enabling the dust removal system to precisely adjust valves and achieve good dynamic adjustment effect.

[0110] Please see Figure 8 This is a schematic diagram illustrating the pressure error of a pressure gauge under multiple operating conditions after optimized calculation, as shown in an exemplary embodiment of the present invention. Figure 8 yes Figure 5The pressure error between the measured pressure and the simulated pressure of pressure gauge A after optimization calculation is clearly shown to be significantly reduced compared to before optimization calculation.

[0111] Step S330: Input the global optimized valve coefficients of each valve into the dust removal pipeline simulation model for simulation calculation, and obtain the simulation result set.

[0112] Specifically, the simulation result set is determined as follows: traverse the valve opening of each valve within any combination of opening intervals; input the valve opening of each valve and the globally optimized valve coefficient corresponding to the combination of opening intervals into the dust removal pipeline simulation model for hydraulic calculation to obtain the simulation optimization results of each valve. The simulation optimization results include at least the simulated flow rate and the optimized simulated pressure of each dust removal point; aggregate the simulated flow rate of each dust removal point, the optimized simulated pressure of each dust removal point, and the valve opening of each valve to obtain the simulation result set.

[0113] Please see Figure 9 This is a schematic diagram illustrating a simulation calculation process as an exemplary embodiment of the invention. Figure 9 As shown, after obtaining the simulation results, the optimized simulation calculation process involves knowing the valve opening at each dust collection point, then determining the globally optimized valve coefficient corresponding to that valve opening, and thus calculating the simulated flow rate and optimized simulated pressure at the dust collection point. However, in practical applications, it is necessary to first allocate the air volume at each dust collection point based on the actual production situation of the dust collection system. The target flow rate and target pressure at each dust collection point are known, i.e., the actual flow rate and actual pressure at each dust collection point are known. At this point, it is necessary to calculate the globally optimized valve coefficient to find the target valve opening for each valve, and use the target valve opening to guide the actual operation of the on-site regulating valves to facilitate valve control. It is important to understand that the globally optimized valve coefficient is a piecewise function. Due to the characteristics of piecewise functions, the calculation of each valve will consume a significant amount of computation time and resources during the trial calculation of valve opening. Excessive computation time may lead to low efficiency of the dust collection system, untimely control, and inability to meet the on-site production conditions, resulting in the failure of the dust collection system and its inability to protect the working environment and reduce air pollution.

[0114] In one embodiment of the present invention, to avoid the need for trial calculations for piecewise functions, a machine learning method is used instead of trial calculations, significantly improving work efficiency. Since the simulation result set will be used to train the machine learning model, if the simulation calculation is based solely on the valve openings of each valve under all operating conditions that have occurred in the history of the dust removal system, the resulting simulation result set will not be comprehensive. By traversing the range of possible valve openings within any given opening interval, the possible valve openings are input into the dust removal network simulation model for hydraulic calculations, resulting in a more comprehensive simulation result set and better performance for training the machine learning model. For example, a 2% opening interval is considered a low opening interval for valves. If a valve is within a 2% opening interval, then each valve has 50 different valve openings, and six valves could have 50 to the power of 6 possible combinations of valve openings. By inputting the valve openings of each valve in these combinations into the dust removal network simulation model and performing calculations using a hydraulic program, the simulation optimization results corresponding to the valve openings of each valve can be obtained. The hydraulic program is a mature existing technology and will not be elaborated upon here.

[0115] Step S340: Construct a machine learning model and train the machine learning model based on the simulation result set, so as to adjust the valve opening of each valve through the trained machine learning model.

[0116] Specifically, the simulation result set is labeled to form a sample dataset; a machine learning model is trained based on the sample dataset; the target flow rate of each dust removal point is obtained; the target flow rate of each dust removal point is input into the trained machine learning model to output the target valve opening of each valve; and the current valve opening of each valve is adjusted to the target valve opening. In this embodiment, because the trained machine learning model has fast calculation speed and high accuracy, it saves computing time and computing resources.

[0117] Please see Figure 10 This is a schematic diagram illustrating machine learning model training as an exemplary embodiment of the present invention. Figure 10 As shown, when the trained machine learning model is used in actual operation, it is only necessary to input the target flow rate of each dust removal point into the machine learning model. The machine learning model will automatically output the target valve opening degree of each valve corresponding to each dust removal point to guide valve operation, realize valve control, and make the dynamic control effect of the dust removal system good and timely. The machine learning algorithms used by the machine learning model include, but are not limited to, Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and BackPropagation Neural Network (BP), etc., and this invention does not limit them.

[0118] Please see Figure 11 This is a schematic diagram illustrating the dynamic control of a dust removal system in one embodiment of the present invention. Figure 11 As shown, in one embodiment of the present invention, because the machine learning model has high computational efficiency and can meet the needs of dynamic control of the dust removal system, when the air volume demand of each dust removal point changes, the machine learning model can quickly respond to the target flow rate corresponding to the air volume of each dust removal point, output the target valve opening required by the valve at each dust removal point, and feed back the target valve opening of each valve to each regulating valve in the dust removal system. The valve operation is controlled by the actuator in the regulating valve, and the current valve opening of the valve is corrected to the target valve opening, so as to realize the dynamic control of the valve by the dust removal system.

[0119] Please see Figure 12 This is a schematic diagram illustrating a business process as shown in an exemplary embodiment of the present invention. Figure 12 As shown, the business process of this method and the specific explanation of each process stage are introduced. First, the valve model and dust removal pipeline network simulation model are constructed. The dust removal pipeline network simulation model provides a clear understanding of the pipeline network, which includes components such as long straight pipes, tees, elbows, and regulating valves. Second, the initial valve coefficients of each valve are obtained. A calculation formula for the valve coefficients is established using the valve model, and then the valve coefficients are initialized to obtain the initial valve coefficients. Third, based on the historical operating data of the dust removal system, the operating conditions are classified and statistically analyzed to obtain a set of operating conditions. The classification and statistics are based on factors such as the production status of dust removal points and valve openings during historical operations. The set of operating conditions includes all operating conditions that occurred in the dust removal system during historical operations. Fourth, based on the set of operating conditions, the initial valve coefficients are input into the dust removal pipeline network simulation model for hydraulic calculations. That is, the initial valve coefficients of each valve are input into the dust removal pipeline network simulation model, and hydraulic calculations are performed sequentially on all statistically obtained operating conditions to obtain the simulated pressure of each dust removal point. Fifth, determine the objective function for optimization calculations based on the installation status of pressure gauges in the dust removal system at the production site, and determine the constraints based on the pipeline type of the dust removal system. Sixth, optimize the initial valve coefficients to find the globally optimized valve coefficients that minimize the cumulative pressure error. Seventh, input the valve opening degree of each valve and the globally optimized valve coefficients under the corresponding opening degree combinations into the dust removal pipeline network simulation model for simulation calculations, obtaining a simulation result set. Eighth, train a machine learning model using the simulation result set, taking the simulated flow rate and optimized simulated pressure at each dust removal point as input, and the corresponding valve opening degree as output. Ninth, after training the machine learning model, input the target flow rate of each dust removal point into the trained machine learning model, and the machine learning model will output the target valve opening degree for each valve.

[0120] Please see Figure 13The diagram illustrates a block diagram of a valve coefficient optimization calculation device according to an exemplary embodiment of the present invention. This exemplary valve coefficient optimization calculation device includes: an acquisition module 1310, an optimization module 1320, a simulation module 1330, and a training module 1340.

[0121] The acquisition module 1310 is configured to acquire the initial valve coefficients, optimization strategies, and simulation models of the dust removal pipeline network corresponding to each valve in the dust removal system. The optimization strategies include at least the objective function and constraints.

[0122] The optimization module 1320 is configured to calculate the initial valve coefficients of each valve using the objective function and constraints, and obtain the global optimized valve coefficients of each valve.

[0123] The simulation module 1330 is configured to input the globally optimized valve coefficients into the dust removal pipeline simulation model for simulation calculation, and obtain the simulation result set of each valve;

[0124] Training module 1340 is configured to build a machine learning model and train the machine learning model based on the simulation result set, so as to adjust the valve opening of each valve through the trained machine learning model.

[0125] It should be noted that the valve coefficient optimization calculation device and the valve coefficient optimization calculation method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the valve coefficient optimization calculation device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0126] Please see Figure 14 The present invention also provides an electronic device 1400, including a memory 1410, a processor 1420, and a computer program stored in the memory and executable on the processor. When the processor 1420 executes the computer program, it implements the steps of the valve coefficient optimization calculation method as described in any of the above embodiments.

[0127] In this embodiment, the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. The processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0128] The present invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a computer's processor, causes the computer to perform the valve coefficient optimization calculation method as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not assembled into the electronic device.

[0129] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0130] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0131] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for optimizing valve coefficient calculation, characterized in that, include: The process involves obtaining the initial valve coefficients, optimization strategies, and a simulation model of the dust removal pipeline network corresponding to each valve in the dust removal system. The optimization strategy includes at least an objective function and constraints, with the constraints including an upper limit and a lower limit for the initial valve coefficients. Before obtaining the initial valve coefficients for each valve in the dust removal system, the process further includes: constructing a valve model based on the valve type of the dust removal system to obtain the valve coefficients for each valve; dividing the valve coefficients according to different valve opening intervals to obtain the valve coefficients to be optimized corresponding to each opening interval; and initializing the valve coefficients to be optimized to determine the initial valve coefficients. The initial valve coefficients of each valve are calculated using the objective function and the constraints to obtain the globally optimized valve coefficients of each valve. This includes: combining different opening intervals of each valve with different opening intervals of other valves to form an array matrix. Each array matrix represents a combination of valves in any opening interval, and the combinations represented by each array are unique. The initial valve coefficients are determined by the opening intervals corresponding to each valve in the array. The cumulative value of the pressure error at each dust removal point corresponding to each array is calculated, and the initial valve coefficient corresponding to the smallest cumulative value is taken as the globally optimized valve coefficient. The globally optimized valve coefficients are input into the dust removal pipeline simulation model for simulation calculation, and the simulation result set of each valve is obtained. A machine learning model is constructed and trained based on the simulation result set, so as to adjust the valve opening of each valve through the trained machine learning model.

2. The valve coefficient optimization calculation method as described in claim 1, characterized in that, Before calculating the initial valve coefficients of each valve using the objective function and the constraints, the method further includes: Obtain the simulated pressure and measured pressure at each dust removal point under each working condition; The pressure error of the dust removal point is obtained by calculating the difference between the measured pressure and the corresponding simulated pressure. The pressure error at each dust removal point is weighted and the objective function is determined.

3. The valve coefficient optimization calculation method as described in claim 2, characterized in that, The process of obtaining the simulated pressure of each dust removal point under each operating condition includes: Obtain the historical operating data of the dust removal system; A set of working conditions including multiple working conditions is generated based on the historical working data; Hydraulic calculations are performed on the initial valve coefficients of each valve based on the set of operating conditions and the simulation model of the dust removal pipeline network to obtain the simulated pressure of each dust removal point under each operating condition. The valves and dust removal points correspond one-to-one.

4. The valve coefficient optimization calculation method as described in claim 1, characterized in that, The globally optimized valve coefficients of each valve are input into the dust removal pipeline simulation model for simulation calculation, resulting in a simulation result set, including: The valve opening degree of each valve within any combination of opening degree intervals; The global optimized valve coefficient corresponding to the combination of the valve opening degree and the opening degree range of each valve is input into the dust removal pipeline simulation model for hydraulic calculation to obtain the simulation optimization results of each valve. The simulation optimization results include at least the simulation flow rate of each dust removal point and the optimized simulation pressure of each dust removal point. The simulated flow rate of each dust removal point, the optimized simulated pressure of each dust removal point, and the valve opening of each valve are aggregated to obtain a simulation result set.

5. The valve coefficient optimization calculation method as described in claim 4, characterized in that, The step of constructing a machine learning model and training the machine learning model based on the simulation result set, so as to adjust the valve opening of each valve through the trained machine learning model, includes: The simulation result set is labeled to form a sample dataset; The machine learning model is trained based on the sample dataset; Obtain the target flow rate for each of the dust removal points; The target flow rate of each dust removal point is input into the trained machine learning model, so that the machine learning model outputs the target valve opening of each valve; Adjust the current valve opening of each valve to the target valve opening.

6. A valve coefficient optimization calculation device, characterized in that, include: The acquisition module is used to acquire the initial valve coefficients, optimization strategies, and simulation models of the dust removal pipeline network corresponding to each valve in the dust removal system. The optimization strategy includes at least an objective function and constraints, and the constraints include an upper limit and a lower limit of the initial valve coefficients. Before acquiring the initial valve coefficients of each valve in the dust removal system, the module further includes: constructing a valve model based on the valve type of the dust removal system to obtain the valve coefficients of each valve; dividing the valve coefficients according to different valve opening intervals to obtain the valve coefficients to be optimized corresponding to the opening intervals; and initializing the valve coefficients to be optimized to determine the initial valve coefficients. An optimization module is used to calculate the initial valve coefficients of each valve using the objective function and the constraints, to obtain the global optimized valve coefficients of each valve. This includes: combining different opening intervals of each valve with different opening intervals of other valves to form an array matrix, where each array matrix represents a combination of valves in any opening interval, and the combinations represented by each array are unique; the initial valve coefficients are determined by the opening intervals corresponding to each valve in the array; calculating the cumulative pressure error of each dust removal point corresponding to each array, and using the initial valve coefficient corresponding to the smallest cumulative value as the global optimized valve coefficient. The simulation module is used to input the globally optimized valve coefficients into the dust removal pipeline simulation model for simulation calculation, and obtain the simulation result set of each valve; The training module is used to build a machine learning model and train the machine learning model based on the simulation result set, so as to adjust the valve opening of each valve through the trained machine learning model.

7. An electronic device, characterized in that, The electronic device includes; One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the electronic device to implement the valve coefficient optimization calculation method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, It stores a computer program that enables the computer to execute the valve coefficient optimization calculation method as described in any one of claims 1 to 5, or one or more processors.

Images