Method and device for suppressing pressure fluctuations in an emulsion pump station

Abstract

The application discloses a pressure fluctuation suppression method and device of an emulsion pump station, and relates to the technical field of hydraulic transmission. The method comprises the following steps: real-time monitoring of pressure data at a predetermined position by deploying an emulsion pump station pressure sensor, and obtaining a pump outlet, equipment inlet and main supply pipeline pressure sequence set; inputting the sequence set into a BP neural network model, predicting the main supply pipeline pressure distribution of the next monitoring node and calculating the pressure deviation; if the pressure deviation is less than a predetermined fluctuation scale, performing valve opening degree and flow proportion optimization analysis according to the pressure distribution for the purpose of eliminating the deviation, and outputting an adjustment scheme; and performing emulsion pump station optimization control of the next monitoring node according to the adjustment scheme. The application solves the technical problems of pump station work instability and equipment damage caused by pressure fluctuation in the emulsion pump station, and achieves the technical effects of improving pump station operation stability, reducing equipment wear and controlling hysteresis by real-time monitoring and optimization adjustment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hydraulic transmission, in particular to a pressure fluctuation suppression method and device for an emulsion pump station. BACKGROUND

[0002] Emulsion pump stations are widely used in mining, metallurgy, and mechanical processing fields, mainly for providing high-pressure emulsion to meet the lubrication and cooling needs of hydraulic equipment. However, in actual operation, the pressure fluctuation problem of emulsion pump stations has always been an important factor affecting system stability and operational efficiency. Due to the complex working environment, emulsion pump stations often face multiple sources of pressure fluctuations, such as transient pressure changes caused by pump set start-up or shutdown, instability of fluid flow in pipelines, and dynamic changes in device inlet and outlet pressures. These pressure fluctuations can cause unstable operation of hydraulic equipment, and even cause equipment damage and pipeline leakage due to excessively high or low pressure. SUMMARY

[0003] The present application provides a pressure fluctuation suppression method and device for an emulsion pump station, which solves the technical problem of pump station instability and equipment damage caused by pressure fluctuations in emulsion pump stations.

[0004] The present application provides a pressure fluctuation suppression method for an emulsion pump station, which includes: during the operation of the emulsion pump station, real-time monitoring of pressure data at a predetermined position by a pressure sensor deployed in the emulsion pump station, obtaining a pump outlet pressure monitoring sequence, a device inlet pressure monitoring sequence, and a main supply pipeline pressure monitoring sequence set; inputting the pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence, and the main supply pipeline pressure monitoring sequence set into a pre-trained BP neural network model, predicting the output of the main supply pipeline predicted pressure distribution of the next monitoring node, and calculating the predicted pressure deviation; if the predicted pressure deviation is less than a predetermined fluctuation scale, optimizing valve opening and flow distribution ratio based on the main supply pipeline predicted pressure distribution to eliminate the predicted pressure deviation, and outputting an optimized adjustment scheme; and performing optimized control of the emulsion pump station at the next monitoring node according to the optimized adjustment scheme.

[0005] The application also provides a pressure fluctuation suppression device of an emulsion pump station, comprising: a real-time monitoring module: during the working process of the emulsion pump station, the pressure data of a predetermined position is monitored in real time by a pressure sensor arranged at the emulsion pump station, and a pump outlet pressure monitoring sequence, a device inlet pressure monitoring sequence and a main supply pipeline pressure monitoring sequence set are obtained; a model prediction module: the pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are input into a pre-trained BP neural network model, a main supply pipeline predicted pressure distribution of a next monitoring node is predicted and output, and a predicted pressure deviation is calculated; an optimization analysis module: if the predicted pressure deviation is less than a predetermined fluctuation scale, valve opening degree and flow distribution ratio optimization analysis is performed according to the main supply pipeline predicted pressure distribution for the purpose of eliminating the predicted pressure deviation, and an optimization adjustment scheme is output; and an optimization control module: optimization control of the emulsion pump station of the next monitoring node is performed according to the optimization adjustment scheme.

[0006] The application provides a pressure fluctuation suppression method and device of an emulsion pump station. Firstly, during the working process of the emulsion pump station, the pressure data of a predetermined position is monitored in real time by a pressure sensor arranged at the emulsion pump station, and a pump outlet pressure monitoring sequence, a device inlet pressure monitoring sequence and a main supply pipeline pressure monitoring sequence set are obtained. Then, the pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are input into a pre-trained BP neural network model, a main supply pipeline predicted pressure distribution of a next monitoring node is predicted and output, and a predicted pressure deviation is calculated. Further, if the predicted pressure deviation is less than a predetermined fluctuation scale, valve opening degree and flow distribution ratio optimization analysis is performed according to the main supply pipeline predicted pressure distribution for the purpose of eliminating the predicted pressure deviation, and an optimization adjustment scheme is output. Finally, optimization control of the emulsion pump station of the next monitoring node is performed according to the optimization adjustment scheme. The technical problems of unstable working of the pump station and damage of equipment caused by pressure fluctuation in the emulsion pump station are solved, and the technical effects of improving the running stability of the pump station, reducing equipment wear and controlling hysteresis are achieved through real-time monitoring and optimization adjustment. BRIEF DESCRIPTION OF DRAWINGS

[0007] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. In the present application, a flowchart is used to illustrate the operations performed by the device according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can be added to these processes, or one or more steps of the operation can be removed from these processes.

[0008] Figure 1A flow chart of a pressure fluctuation suppression method of an emulsion pump station provided by an embodiment of the present application.

[0009] Figure 2 A structural diagram of a pressure fluctuation suppression device of an emulsion pump station provided by an embodiment of the present application.

[0010] Legend: real-time monitoring module 11, model prediction module 12, optimization analysis module 13, and optimization control module 14. DETAILED DESCRIPTION

[0011] The above description is only a summary of the technical solutions of the present application. In order to make the technical solutions of the present application more clear, the following detailed description can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0012] In order to make the purposes, technical solutions and advantages of the present application more clear, the following will further describe the present application with reference to the accompanying drawings, and the described embodiments should not be regarded as limiting the present application, and all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0013] In the following description, "some embodiments" are described, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict, and the term "first\second" referred to only distinguishes similar objects, and does not represent a specific order of the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, device, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0014] The embodiments of the present application provide a pressure fluctuation suppression method of an emulsion pump station, as shown in Figure 1 The method comprises:

[0015] During the working process of the emulsion pump station, the pressure data of the predetermined position is monitored in real time by the pressure sensor deployed in the emulsion pump station, and the pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are obtained.

[0016] In the embodiments of the present application, when the emulsion pump station is running, pressure sensors are installed at key positions of the pump station to collect and monitor pressure data of each preset position in real time. The data is derived from multiple important monitoring points such as the outlet of the pump, the inlet of the equipment, and the main supply pipeline. By continuously collecting these pressure data, corresponding monitoring sequences are generated, including a pump outlet pressure monitoring sequence set, an equipment inlet pressure monitoring sequence set, and a main supply pipeline pressure monitoring sequence set. The real-time monitoring of these data provides necessary basic information for subsequent pressure fluctuation analysis and optimization control.

[0017] Further, the present application provides a method for obtaining a pump outlet pressure monitoring sequence, an equipment inlet pressure monitoring sequence, and a main supply pipeline pressure monitoring sequence set, comprising:

[0018] Real-time monitoring of pressure data at predetermined positions by pressure sensors deployed at the emulsion pump station, wherein the predetermined positions include a pump outlet position, an equipment inlet position, and several pipeline positions in the main supply pipeline; collecting historical pressure data of K consecutive monitoring nodes before a next monitoring node to construct a pump outlet pressure monitoring sequence, an equipment inlet pressure monitoring sequence, and a main supply pipeline pressure monitoring sequence set, wherein the main supply pipeline pressure monitoring sequence set includes a main supply pipeline pressure distribution under K consecutive monitoring nodes.

[0019] Preferably, during the operation of the emulsion pump station, the pressure data in the emulsion pump station is monitored in real time by deploying pressure sensors at specific positions, including the pump outlet, the equipment inlet, and multiple key pipeline positions inside the main supply pipeline. The selection of these positions aims to comprehensively capture key information about pressure changes during the operation of the pump station. The pressure sensor at the pump outlet position is mainly used to monitor the output pressure of the pump station, ensuring that the supply capacity of the emulsion meets the demand. The pressure sensor at the equipment inlet position is used to monitor the pressure of the emulsion received by each device (such as hydraulic supports, hydraulic cylinders, cutting equipment, etc.), ensuring that the delivery of the emulsion meets the usage requirements of each device. The pressure sensors at multiple pipeline positions inside the main supply pipeline are distributed to record the pressure changes inside the pipeline, reflecting the distribution ratio of the emulsion between different devices, in order to dynamically grasp the overall operation of the emulsion pump station. On the basis of real-time monitoring, historical pressure data of the previous K consecutive monitoring nodes are further collected, which record the dynamic changes of the pump station in the recent period, providing timing information for subsequent analysis. By organizing the historical data and real-time data, a pump outlet pressure monitoring sequence set, an equipment inlet pressure monitoring sequence set, and a main supply pipeline pressure monitoring sequence set are constructed. The main supply pipeline pressure monitoring sequence set contains the pressure distribution of the main supply pipeline under K consecutive monitoring nodes, which reflects the pressure change trend of the emulsion in the pipeline network and whether the distribution ratio of the emulsion between devices meets the expectation. Through the above process, the key pressure information of the pump station is completely recorded and sequenced, laying a foundation for subsequent pressure prediction, distribution optimization, and implementation of overall control strategies.

[0020] The pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence, and the main supply pipeline pressure monitoring sequence set are input into a pre-trained BP neural network model to predict the main supply pipeline predicted pressure distribution of the next monitoring node and calculate the predicted pressure deviation.

[0021] In one embodiment, the pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence, and the main supply pipeline pressure monitoring sequence set are input into a pre-trained BP neural network model. The model uses its learning and prediction ability for complex pressure changes to generate the main supply pipeline predicted pressure distribution of the next monitoring node, which refers to the time point predicted in the pressure monitoring and control that will arrive during the operation of the emulsion pump station. By comparing the main supply pipeline predicted pressure distribution of the next monitoring node with the standard pressure distribution, the difference between them, i.e., the predicted pressure deviation, is calculated to evaluate the degree of pressure fluctuation of the current emulsion pump station. This process provides reliable data support for further optimizing pressure control.

[0022] Further, the application provides inputting the pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set into a pre-trained BP neural network model, and predicting the main supply pipeline predicted pressure distribution of the next monitoring node, including:

[0023] According to the historical emulsion pump station work log in the time zone, the sample pump outlet pressure monitoring sequence, the sample equipment inlet pressure monitoring sequence and the sample main supply pipeline pressure monitoring sequence set are collected, and the main supply pipeline pressure distribution of the next monitoring node is collected to obtain a sample main supply pipeline pressure distribution; the sample pump outlet pressure monitoring sequence, the sample equipment inlet pressure monitoring sequence and the sample main supply pipeline pressure monitoring sequence set are used as input data during training, the sample main supply pipeline pressure distribution is used as supervision data, and the BP neural network is iteratively trained combined with the gradient descent algorithm to obtain a main supply pipeline pressure prediction model meeting the convergence condition; the main supply pipeline pressure prediction model is used to predict the main supply pipeline predicted pressure distribution according to the pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set.

[0024] Preferably, based on the working log of the emulsion pump station in the historical time zone (such as half a month, one month, etc.), the historical pressure monitoring data is extracted to form a sample pump outlet pressure monitoring sequence set, a sample equipment inlet pressure monitoring sequence set and a sample main supply pipeline pressure monitoring sequence set, wherein the sample pump outlet pressure monitoring sequence set records the pressure change of the pump station outlet, reflecting the output state of the emulsion, the sample equipment inlet pressure monitoring sequence set records the pressure change of the emulsion before entering each device (such as hydraulic support, hydraulic cylinder, cutting equipment, etc.), reflecting the actual pressure level of the device when receiving the emulsion, and the sample main supply pipeline pressure monitoring sequence set covers the pressure data of multiple monitoring points in the main supply pipeline at different times, showing the dynamic change of the pressure distribution in the pipeline, indicating whether the emulsion distribution is balanced, and then based on these sample sequences, the main supply pipeline pressure distribution of the next monitoring node corresponding to each sequence is collected as a reference value to form a sample main supply pipeline pressure distribution, reflecting the change of the emulsion pump station pressure in actual operation; subsequently, the collected sample pump outlet, equipment inlet and main supply pipeline pressure sequence set is taken as input data to represent the pressure change at different positions of the pump station, and the sample main supply pipeline pressure distribution under the next monitoring node is taken as supervision data to train the target output of the BP neural network model, specifically, an initial model structure for emulsion pump station pressure distribution prediction is constructed using the BP neural network, including an input layer, a hidden layer and an output layer, the weights and bias values of the BP neural network are initialized using random numbers or uniform distribution, and the input data is input into the initialized BP neural network for forward propagation, the input data includes the sample pump outlet pressure monitoring sequence, the sample equipment inlet pressure monitoring sequence and the sample main supply pipeline pressure monitoring sequence set, which is transmitted layer by layer through the input layer and the hidden layer, and the main supply pipeline pressure distribution prediction result of the next monitoring node is generated in the output layer.After that, the mean square error (MSE) is used as the loss function to calculate the loss value between the prediction result and the supervised data (sample main supply pipeline pressure distribution), and the gradient of the loss to each layer weight and bias is calculated layer by layer through the back propagation algorithm, and then the gradient descent algorithm is used to optimize the model parameters, gradually adjust the weight and bias value to minimize the value of the loss function, in each iteration, the process of forward propagation, loss calculation, back propagation and parameter update is repeated, until the loss value converges or reaches the maximum iteration times; After the training is completed, the data not used for training is used to test the performance of the model, to evaluate the accuracy of the BP neural network in predicting the main supply pipeline pressure distribution, if the prediction error on the validation set meets the expected range, the current BP neural network model is saved as the final main supply pipeline pressure prediction model, if the performance is not up to standard, adjust the hyperparameters (such as learning rate, number of hidden layer neurons, training batch size, etc.), retrain to optimize the prediction effect of the model; Then, the real-time pressure monitoring data of the emulsion pump station (pump outlet pressure monitoring sequence, device inlet pressure monitoring sequence and main supply pipeline pressure monitoring sequence set) is input into the trained main supply pipeline pressure prediction model, the model will automatically predict the main supply pipeline pressure distribution of the next monitoring node, providing accurate basis for the generation of optimization adjustment scheme, this model can be widely used in dynamic pressure control of emulsion pump station, effectively improving the operation stability and efficiency of the pump station.

[0025] Further, the application provides a calculated predicted pressure deviation, comprising:

[0026] The predicted pressure deviation is calculated by mapping the main supply pipeline pressure deviation distribution based on the standard main supply pipeline pressure distribution; the pressure deviation mean and the deviation distribution dispersion are obtained by calculating the deviation mean and the deviation distribution dispersion of the main supply pipeline pressure deviation distribution; and the predicted pressure deviation is obtained by multiplying the pressure deviation mean by the weight set according to the deviation distribution dispersion, wherein the weight and the deviation distribution dispersion are negatively correlated.

[0027] Optionally, after obtaining the predicted pressure distribution of the main supply pipeline, the predicted pressure distribution of the main supply pipeline is mapped with the standard main supply pipeline pressure distribution, and the difference value of each point is calculated by point-by-point comparison to determine the difference between the predicted value and the standard value, thereby obtaining the deviation of all monitoring points in the main supply pipeline from the standard pressure, and then storing these deviations point by point to form the pressure deviation distribution of the main supply pipeline, representing the pressure change in the entire pipeline; then, the deviation values of all monitoring points in the main supply pipeline pressure deviation distribution are calculated to obtain the pressure deviation mean, and the deviation distribution dispersion is calculated by the discrete calculation formula The deviation distribution dispersion is calculated by the discrete calculation formula, wherein D is the deviation distribution dispersion, N is the total number of monitoring points, the pressure deviation of the i-th monitoring point, is the mean value of the pressure deviation; then, the weight W is set according to the dispersion D of the deviation distribution, and the weight W is negatively correlated with the dispersion D of the deviation distribution, that is, , wherein, is an adjustment coefficient, used to adjust the sensitivity of the weight, and is determined according to business needs and expert decisions. The final predicted pressure deviation is calculated by multiplying the mean value of the pressure deviation by the set weight W. Through the above process, the pressure deviation of the main supply pipeline can be comprehensively evaluated from the overall and local perspectives. The mean value of the deviation reflects the degree of overall pressure deviation, and the dispersion reflects the uniformity of the pressure deviation. Combined with the weight adjustment, the predicted pressure deviation can be more accurately calculated, providing a reliable basis for subsequent pressure control and optimization scheme formulation.

[0028] If the predicted pressure deviation is less than the predetermined fluctuation range, the valve opening and flow distribution ratio optimization analysis is performed according to the predicted pressure distribution of the main supply pipeline to eliminate the predicted pressure deviation, and an optimized adjustment scheme is output.

[0029] In one embodiment, when the calculated predicted pressure deviation is less than the predetermined fluctuation range, it means that the current pressure change is within an acceptable range, but further optimization adjustment is still needed to improve the stability of the emulsion pump station. At this time, further optimization analysis is performed to eliminate the predicted pressure deviation. Specifically, according to the predicted pressure distribution of the main supply pipeline, simulation adjustment is performed in combination with the initial adjustment parameters, the fitness of each initial adjustment parameter is analyzed, and these fitness and initial adjustment parameters are used as optimization data to perform adjustment parameter optimization through a genetic algorithm to optimize the valve opening and emulsion flow distribution ratio, thereby obtaining an optimized adjustment scheme. By adjusting these parameters, fine control of the pressure can be achieved to ensure that the pressure distribution of each part of the pump station is more uniform, further reducing fluctuations. Finally, the generated optimized adjustment scheme will be used to guide the operation of the pump station to ensure that the pressure fluctuations during operation are effectively suppressed, unnecessary wear is reduced, and the stability and efficiency of the pump station are improved.

[0030] Further, the application provides that if the predicted pressure deviation is greater than or equal to a predetermined fluctuation scale, a pump rotating speed sequence, a mine dust density sequence, an emulsion temperature sequence, an environment temperature sequence, and a continuous working time length are collected; an emulsion viscosity sequence is predicted according to the mine dust density sequence, the emulsion temperature sequence, and the continuous working time length, wherein a BP neural network model is used for emulsion viscosity prediction; an auxiliary main supply pipeline predicted pressure distribution of a next monitoring node is predicted according to the pump rotating speed sequence, the environment temperature sequence, the emulsion viscosity sequence, and the continuous working time length; a comprehensive main supply pipeline predicted pressure distribution is calculated by mapping the main supply pipeline predicted pressure distribution and the auxiliary main supply pipeline predicted pressure distribution according to a predetermined weight proportion, wherein the prediction data weight is 0.8, and the auxiliary prediction data weight is 0.2; and an updated prediction pressure deviation is calculated according to the comprehensive main supply pipeline predicted pressure distribution with a standard main supply pipeline pressure distribution as a benchmark.

[0031] Optionally, when the predicted pressure deviation is greater than or equal to the predetermined fluctuation scale, it means that the current pressure change has exceeded the acceptable range, at this time, a variety of data sources related to pressure fluctuation are collected to form a pump speed sequence, a mine dust density sequence, an emulsion temperature sequence, an environmental temperature sequence and a continuous working time, wherein the pump speed sequence reflects the running state and output capacity of the pump, the mine dust density sequence indicates the concentration of particulate matter in the emulsion running environment, which may affect the pressure and flowability, the emulsion temperature sequence is used to evaluate the temperature state of the emulsion and its influence on viscosity, the environmental temperature sequence reflects the indirect influence of external environmental temperature on the operation of emulsion and pump station, and the continuous working time represents the length of time of the continuous operation of the pump station, and its cumulative effect may affect the performance of the emulsion; then, the collected mine dust density sequence, emulsion temperature sequence and continuous working time are used as inputs to predict the viscosity of the emulsion by using the viscosity prediction model trained by the BP neural network, the viscosity prediction model calculates the predicted viscosity values of the emulsion at different time points by forward propagation of the input data, and outputs the emulsion viscosity sequence composed of these predicted values, which reflects the flowability of the fluid and has a direct influence on pressure fluctuation, wherein the construction process of the viscosity prediction model is the same as that of the aforementioned main supply pipeline pressure prediction model; then, according to the pump speed sequence, the environmental temperature sequence, the emulsion viscosity sequence and the continuous working time, the auxiliary pressure prediction model is used to predict the auxiliary main supply pipeline pressure distribution of the next monitoring node, the auxiliary distribution is used to supplement and correct the accuracy of the main supply pipeline predicted pressure distribution, wherein the construction method of the auxiliary pressure prediction model is the same as that of the aforementioned main supply pipeline pressure prediction model; then, the main supply pipeline predicted pressure distribution and the auxiliary main supply pipeline predicted pressure distribution are weighted calculated according to the set weight proportion, the weight of the main supply pipeline predicted pressure distribution is 0.8, which is determined according to the business accuracy requirement and expert decision, and the weight of the auxiliary main supply pipeline predicted pressure distribution is 0.2, which is also determined according to the business accuracy requirement and expert decision, and through the weighted calculation, the comprehensive main supply pipeline predicted pressure distribution can be obtained; finally, taking the standard main supply pipeline pressure distribution as the benchmark, the comprehensive main supply pipeline pressure distribution is compared with the standard value, and the updated predicted pressure deviation is calculated by the same method as described above, which reflects the change of the main supply pipeline pressure under the comprehensive action of multiple data sources. Through the above process, the emulsion pump station is comprehensively evaluated in combination with various operating parameters and environmental conditions, the accuracy of pressure prediction is improved, and through the calculation of the comprehensive main supply pipeline pressure distribution, the actual situation of pressure fluctuation can be more accurately reflected, and a more precise optimization and adjustment scheme can be developed to ensure the stable operation of the pump station in complex environment.

[0032] Further, the application provides that if the updated predicted pressure deviation is greater than or equal to the predetermined fluctuation scale, a deviation overflow ratio is calculated according to the updated predicted pressure deviation and the predetermined fluctuation scale; a predetermined optimization iteration number is matched based on the deviation overflow ratio; and valve opening degree and flow distribution ratio optimization analysis is performed according to the predicted pressure distribution of the main supply pipeline, with the purpose of eliminating the updated predicted pressure deviation and with the predetermined optimization iteration number as a constraint, to output an optimized adjustment scheme.

[0033] Optionally, if the updated predicted pressure deviation is greater than or equal to the predetermined fluctuation scale, it means that the pressure distribution of the current main supply pipeline has significant abnormalities or deviations. At this time, the updated predicted pressure deviation is calculated by ratio with the predetermined fluctuation scale to obtain a deviation overflow ratio, which represents the relative proportion of the updated pressure deviation exceeding the predetermined range, and is used to guide the number of iterations required in the optimization process. Subsequently, a reasonable optimization iteration number is matched according to the deviation overflow ratio to ensure optimization accuracy and efficiency. For example, if the deviation overflow ratio belongs to [1, 2), a smaller iteration number (such as 10 times) is selected for fast convergence, if the deviation overflow ratio belongs to [2, 3), a medium iteration number (such as 20 times) is selected, and if the deviation overflow ratio belongs to [2, +∞), a larger iteration number (such as 50 times) is selected to improve optimization accuracy. In actual operation, these iteration numbers can be dynamically adjusted to balance optimization accuracy and calculation efficiency. Then, valve opening degree and flow distribution ratio optimization analysis is performed with the purpose of eliminating the updated predicted pressure deviation and in combination with the predicted pressure distribution of the main supply pipeline, that is, initial values of valve opening degree and flow distribution ratio are randomly set, these initial values are simulated, and the simulation results are combined with genetic algorithm for iterative optimization. Valve opening degree and flow ratio are adjusted in each iteration until the predetermined optimization iteration number is reached. Finally, according to the optimization analysis results, the initial parameters with the highest fitness generated in the iteration process are output as the optimized adjustment scheme, which includes valve opening degree setting values of each key pipeline, flow distribution ratios between different devices, and adjusted predicted pressure distribution, which will be used to guide the actual operation of the pump station to eliminate or significantly reduce the updated predicted pressure deviation. By matching the optimization iteration number according to the deviation overflow ratio, the calculation complexity can be dynamically adjusted in the optimization process, ensuring the accuracy of optimization and improving the analysis efficiency. The final output optimized adjustment scheme can accurately adjust the operating parameters of the emulsion pump station, significantly suppress pressure fluctuations, reduce unnecessary wear and tear, and improve the stability and operating efficiency of the pump station.

[0034] Further, the application provides valve opening degree and flow distribution ratio optimization analysis according to the predicted pressure distribution of the main supply pipeline, which includes:

[0035] The multiple valve opening parameters and the multiple flow distribution ratios are randomly generated based on a valve opening space and a flow distribution ratio space, and initial adjustment parameters are integrated; the comprehensive main supply pipeline predicted pressure distribution is rendered to a simulation space of the emulsion pump station, and in the simulation space, simulation adjustment is performed according to the multiple initial adjustment parameters, and multiple main supply pipeline simulation pressure distributions are obtained; the multiple adjustment fitnesses are calculated according to the multiple main supply pipeline simulation pressure distributions based on a standard main supply pipeline pressure distribution as a benchmark, wherein the adjustment fitness and the deviation mean are negatively correlated; the multiple adjustment fitnesses and the multiple initial adjustment parameters are used to perform adjustment parameter optimization by using a genetic algorithm with the valve opening space and the flow distribution ratio space as constraints, until the predetermined optimization iteration number is reached, and the output optimization adjustment scheme is output.

[0036] Optionally, multiple sets of initial parameters are randomly generated in a preset valve opening space and flow distribution ratio space, including multiple valve opening parameters and multiple flow distribution ratio parameters, and the valve opening parameters and the flow distribution ratio parameters are integrated into a plurality of initial adjustment parameter sets, each of which represents a possible adjustment scheme; then, the comprehensive main supply pipeline predicted pressure distribution is loaded into a simulation space of the emulsion pump station, which is a virtual environment for simulating the interaction of liquid flow, pressure change and various control parameters in the pump station, and is constructed by combining fluid dynamics simulation tools and three-dimensional modeling software, the structure data and actual operation data of pumps, pipelines, valves and other components, and can accurately simulate the actual situation of emulsion flow in the pipeline. By inputting each set of initial adjustment parameters into the simulation space for simulation, the influence of valve opening and flow ratio adjustment on pressure distribution can be understood, and the simulation space can record the results of each simulation, thereby outputting multiple main supply pipeline simulation pressure distributions representing pressure changes under different adjustment schemes; then, the standard main supply pipeline pressure distribution is taken as a benchmark to compare each set of simulation pressure distribution, the deviation mean is calculated in the same way as described above, and the multiple adjustment fitnesses are mapped according to the calculated deviation mean, and the adjustment fitness F is negatively correlated with the deviation mean, that is, The greater the adjustment fitness is, the closer the adjustment scheme is to the standard pressure distribution; then, the initial adjustment parameters are optimized by using a genetic algorithm to find the optimal valve opening and flow distribution ratio. Specifically, the initial parameter set with better performance is selected as the parent according to the adjustment fitness, the selected parent parameter set is combined to generate a new child parameter set, then a part of the child parameter set is randomly mutated to increase the diversity of the parameter space, the new parameter set is substituted into the simulation space to be simulated again, the new adjustment fitness is calculated, and the above process is repeated until a predetermined optimization iteration number is reached; finally, the adjustment parameter set with the highest final fitness is selected as the optimized adjustment scheme of the emulsion pump station, which includes the valve opening setting value of each key pipeline, the flow distribution ratio between different devices, and the predicted pressure distribution after adjustment. Through random generation of parameters, simulation adjustment and genetic algorithm optimization, the optimal combination of valve opening and flow distribution ratio can be efficiently explored, and the final optimized adjustment scheme can effectively reduce pressure deviation and equipment wear, ensure stable operation of the emulsion pump station, and improve pump station operation efficiency and reliability.

[0037] According to the optimized adjustment scheme, the optimized control of the emulsion pump station of the next monitoring node is performed.

[0038] In one embodiment, according to the optimized adjustment scheme, the emulsion pump station is adjusted in real time, and the optimized control of the next monitoring node is performed. Specifically, according to the valve opening and flow distribution ratio set by the optimization scheme, the operation parameters of each part of the pump station are automatically adjusted to ensure that the pressure of the main supply pipeline reaches the expected target and reduces any possible pressure fluctuation. After this adjustment, the optimized control of the next monitoring node is performed, i.e., the above process is performed again. In this way, through continuous real-time adjustment, the pump station can be continuously and stably operated to avoid unnecessary wear and ensure that the delivery and distribution of emulsion meet the actual requirements.

[0039] In the foregoing, with reference to Figure 1 A pressure fluctuation suppression method for an emulsion pump station according to an embodiment of the present application is described in detail. Next, with reference to Figure 2 A pressure fluctuation suppression device for an emulsion pump station according to an embodiment of the present application is described.

[0040] The pressure fluctuation suppression device for an emulsion pump station according to the embodiment of the present application solves the technical problem of unstable operation of the pump station and damage to the equipment caused by pressure fluctuation in the emulsion pump station, and achieves the technical effects of improving the operation stability of the pump station, reducing equipment wear, and controlling hysteresis through real-time monitoring and optimized adjustment. The pressure fluctuation suppression device for an emulsion pump station includes a real-time monitoring module 11, a model prediction module 12, an optimization analysis module 13, and an optimized control module 14.

[0041] The real-time monitoring module 11: during the operation of the emulsion pump station, the pressure data at the predetermined positions are monitored in real time by the pressure sensors deployed at the emulsion pump station, and the pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are obtained; the model prediction module 12: the pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are input into the pre-trained BP neural network model, the main supply pipeline predicted pressure distribution of the next monitoring node is predicted and output, and the predicted pressure deviation is calculated; the optimization analysis module 13: if the predicted pressure deviation is less than the predetermined fluctuation scale, the valve opening and the flow distribution ratio are optimized and analyzed according to the main supply pipeline predicted pressure distribution for the purpose of eliminating the predicted pressure deviation, and the optimization adjustment scheme is output; the optimization control module 14: according to the optimization adjustment scheme, the optimization control of the emulsion pump station of the next monitoring node is performed.

[0042] Further, the real-time monitoring module 11 further comprises:

[0043] The pressure data at the predetermined positions are monitored in real time by the pressure sensors deployed at the emulsion pump station, wherein the predetermined positions include the pump outlet position, the equipment inlet position and a plurality of pipeline positions in the main supply pipeline; the historical pressure data of the K consecutive monitoring nodes before the next monitoring node are collected, and the pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are constructed, wherein the main supply pipeline pressure monitoring sequence set includes the main supply pipeline pressure distribution under the K consecutive monitoring nodes.

[0044] Further, the model prediction module 12 further comprises:

[0045] According to the emulsion pump station operation log in the historical time zone, the sample pump outlet pressure monitoring sequence, the sample equipment inlet pressure monitoring sequence and the sample main supply pipeline pressure monitoring sequence set are collected, and the main supply pipeline pressure distribution under the next monitoring node is collected, and the sample main supply pipeline pressure distribution is obtained; the sample pump outlet pressure monitoring sequence, the sample equipment inlet pressure monitoring sequence and the sample main supply pipeline pressure monitoring sequence set are used as the input data during the training, the sample main supply pipeline pressure distribution is used as the supervision data, the BP neural network is iteratively trained combined with the gradient descent algorithm, and the main supply pipeline pressure prediction model meeting the convergence condition is obtained; the main supply pipeline pressure prediction model is used to predict the main supply pipeline predicted pressure distribution according to the pump outlet pressure monitoring sequence, the equipment inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set.

[0046] Further, the model prediction module 12 further comprises:

[0047] Based on the standard main supply pipeline pressure distribution, a mapping deviation calculation is performed on the predicted main supply pipeline pressure distribution to determine the main supply pipeline pressure deviation distribution; a deviation mean calculation and a deviation distribution discrete analysis are performed on the main supply pipeline pressure deviation distribution to obtain the pressure deviation mean and the deviation distribution discreteness; a weight is set according to the deviation distribution discreteness, and the predicted pressure deviation is obtained by multiplying the pressure deviation mean value, wherein the weight is negatively correlated with the deviation distribution discreteness.

[0048] Furthermore, the optimization analysis module 13 also includes:

[0049] If the predicted pressure deviation is greater than or equal to the predetermined fluctuation scale, the pump speed sequence, mine dust density sequence, emulsion temperature sequence, ambient temperature sequence and continuous working time are collected; the emulsion viscosity sequence is predicted based on the mine dust density sequence, emulsion temperature sequence and continuous working time, wherein the emulsion viscosity is predicted using the BP neural network model; the auxiliary main supply pipeline predicted pressure distribution of the next monitoring node is predicted based on the pump speed sequence, ambient temperature sequence, emulsion viscosity sequence and continuous working time; the main supply pipeline predicted pressure distribution and the auxiliary main supply pipeline predicted pressure distribution are mapped and calculated according to the predetermined weight ratio to obtain the comprehensive main supply pipeline predicted pressure distribution, wherein the predicted data weight is 0.8 and the auxiliary predicted data weight is 0.2; based on the standard main supply pipeline pressure distribution, the updated predicted pressure deviation is calculated based on the comprehensive main supply pipeline predicted pressure distribution.

[0050] Furthermore, the optimization analysis module 13 also includes:

[0051] If the updated predicted pressure deviation is greater than or equal to the predetermined fluctuation scale, the deviation overflow ratio is calculated based on the updated predicted pressure deviation and the predetermined fluctuation scale; a predetermined number of optimization iterations is obtained based on the deviation overflow ratio matching; with the purpose of eliminating the updated predicted pressure deviation and with the predetermined number of optimization iterations as a constraint, an optimization analysis of the valve opening and flow distribution ratio is performed according to the predicted pressure distribution of the main supply pipeline, and an optimized adjustment plan is output.

[0052] Furthermore, the optimization analysis module 13 also includes:

[0053] The valve opening degree parameters and the flow distribution ratios are randomly generated based on a valve opening degree space and a flow distribution ratio space, and initial adjustment parameters are integrated; the comprehensive main supply pipeline predicted pressure distribution is rendered to a simulation space of the emulsion pump station, and in the simulation space, simulation adjustment is performed according to the initial adjustment parameters, and main supply pipeline simulation pressure distributions are obtained; the adjustment fitness is calculated according to the main supply pipeline simulation pressure distributions based on a standard main supply pipeline pressure distribution, wherein the adjustment fitness and the deviation mean are negatively correlated; the adjustment parameter optimization is performed by using a genetic algorithm according to the adjustment fitness and the initial adjustment parameters, with the valve opening degree space and the flow distribution ratio space as constraints, until the predetermined optimization iteration number is reached, and the output optimization adjustment scheme is output.

[0054] The pressure fluctuation suppression device for the emulsion pump station provided in the embodiments of the present application can perform the pressure fluctuation suppression method for the emulsion pump station provided in any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0055] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server, and each unit and module included is only divided according to the 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 mutual differentiation, and do not limit the protection scope of the present application.

[0056] The specific embodiments described above do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method of suppressing pressure fluctuations in an emulsion pump station, characterized by, The method comprises: During the operation of the emulsion pump station, the pressure data at predetermined positions are monitored in real time by the pressure sensors arranged at the emulsion pump station to obtain a pump outlet pressure monitoring sequence, a device inlet pressure monitoring sequence and a main supply pipeline pressure monitoring sequence set; The pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are input into a pre-trained BP neural network model to predict the main supply pipeline predicted pressure distribution of the next monitoring node and calculate the predicted pressure deviation, including: According to the emulsion pump station operation log in the historical time zone, sample pump outlet pressure monitoring sequences, sample device inlet pressure monitoring sequences and sample main supply pipeline pressure monitoring sequence sets are collected, and the main supply pipeline pressure distribution under the next monitoring node is collected to obtain a sample main supply pipeline pressure distribution; The sample pump outlet pressure monitoring sequences, the sample device inlet pressure monitoring sequences and the sample main supply pipeline pressure monitoring sequence sets are used as input data during training, and the sample main supply pipeline pressure distribution is used as supervision data. The BP neural network is iteratively trained combined with the gradient descent algorithm to obtain a main supply pipeline pressure prediction model meeting the convergence condition; The main supply pipeline pressure prediction model is used to predict the main supply pipeline predicted pressure distribution according to the pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set; The main supply pipeline predicted pressure distribution is mapped and deviation calculated based on the standard main supply pipeline pressure distribution to determine the main supply pipeline pressure deviation distribution; The main supply pipeline pressure deviation distribution is subjected to deviation mean calculation and deviation distribution dispersion analysis to obtain the pressure deviation mean and the deviation distribution dispersion; The weight is set according to the deviation distribution dispersion, and the pressure deviation mean is multiplied by the weight to obtain the predicted pressure deviation, wherein the weight and the deviation distribution dispersion are negatively correlated; If the predicted pressure deviation is less than a predetermined fluctuation scale, valve opening degree and flow distribution ratio optimization analysis is performed according to the main supply pipeline predicted pressure distribution to eliminate the predicted pressure deviation, and an optimized adjustment scheme is output; The emulsion pump station of the next monitoring node is subjected to optimized control according to the optimized adjustment scheme.

2. The method of claim 1, wherein, The pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are obtained, including: The pressure data at predetermined positions are monitored in real time by the pressure sensors arranged at the emulsion pump station, wherein the predetermined positions include the pump outlet position, the device inlet position and a plurality of pipeline positions in the main supply pipeline; The historical pressure data of K consecutive monitoring nodes before the next monitoring node are collected to construct the pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set, wherein the main supply pipeline pressure monitoring sequence set includes the main supply pipeline pressure distribution under the K consecutive monitoring nodes.

3. The method of claim 1, wherein the pressure fluctuation of the emulsion pump station is suppressed by, If the predicted pressure deviation is greater than or equal to the predetermined fluctuation scale, the pump speed sequence, the mine dust density sequence, the emulsion temperature sequence, the environmental temperature sequence and the continuous working time are collected; According to the mine dust density sequence, the emulsion temperature sequence and the continuous working time prediction, an emulsion viscosity sequence is obtained, wherein the emulsion viscosity is predicted by using a BP neural network model; According to the pump rotating speed sequence, the environment temperature sequence, the emulsion viscosity sequence and the continuous working time prediction, an auxiliary main supply pipeline predicted pressure distribution of the next monitoring node is obtained; According to the main supply pipeline predicted pressure distribution and the auxiliary main supply pipeline predicted pressure distribution, a comprehensive main supply pipeline predicted pressure distribution is obtained by mapping calculation according to a predetermined weight ratio, wherein the prediction data weight is 0.8 and the auxiliary prediction data weight is 0.2; According to the comprehensive main supply pipeline predicted pressure distribution, an updated prediction pressure deviation is obtained by taking the standard main supply pipeline pressure distribution as a reference.

4. The method of claim 3, wherein the pressure fluctuation of the emulsion pump station is suppressed by, If the updated prediction pressure deviation is greater than or equal to a predetermined fluctuation scale, a deviation overflow ratio is obtained according to the updated prediction pressure deviation and the predetermined fluctuation scale; The predetermined optimization iteration number is obtained by matching based on the deviation overflow ratio; According to the main supply pipeline predicted pressure distribution, valve opening degree and flow distribution ratio optimization analysis is performed for the purpose of eliminating the updated prediction pressure deviation, and an optimization adjustment scheme is outputted, wherein the predetermined optimization iteration number is taken as a constraint.

5. The method of claim 4, wherein the pressure fluctuation of the emulsion pump station is suppressed by, According to the main supply pipeline predicted pressure distribution, valve opening degree and flow distribution ratio optimization analysis includes: A plurality of valve opening degree parameters and a plurality of flow distribution ratios are randomly generated based on a valve opening degree space and a flow distribution ratio space, and a plurality of initial adjustment parameters are obtained by integration; The comprehensive main supply pipeline predicted pressure distribution is rendered to a simulation space of the emulsion pump station, and a plurality of main supply pipeline simulation pressure distributions are obtained by simulating adjustment according to the plurality of initial adjustment parameters in the simulation space; According to the plurality of main supply pipeline simulation pressure distributions, a plurality of adjustment fitnesses are obtained by taking the standard main supply pipeline pressure distribution as a reference, wherein the adjustment fitness and the deviation mean are negatively correlated; According to the plurality of adjustment fitnesses and the plurality of initial adjustment parameters, adjustment parameter optimization is performed by using a genetic algorithm, taking the valve opening degree space and the flow distribution ratio space as constraints, until the predetermined optimization iteration number is reached, and the output optimization adjustment scheme is outputted.

6. A pressure fluctuation suppressing device for an emulsion pump station, characterized by comprising: The device is used to implement the pressure fluctuation suppression method of the emulsion pump station according to any one of claims 1-5, and includes: A real-time monitoring module: in the working process of the emulsion pump station, the pressure data of the predetermined position is monitored in real time by the pressure sensor arranged in the emulsion pump station, and a pump outlet pressure monitoring sequence, a device inlet pressure monitoring sequence and a main supply pipeline pressure monitoring sequence set are obtained; A model prediction module: the pump outlet pressure monitoring sequence, the device inlet pressure monitoring sequence and the main supply pipeline pressure monitoring sequence set are inputted into a pre-trained BP neural network model, a main supply pipeline predicted pressure distribution of the next monitoring node is predicted and outputted, and a prediction pressure deviation is calculated; An optimization analysis module: if the prediction pressure deviation is less than a predetermined fluctuation scale, valve opening degree and flow distribution ratio optimization analysis is performed according to the main supply pipeline predicted pressure distribution for the purpose of eliminating the prediction pressure deviation, and an optimization adjustment scheme is outputted. An optimization control module: according to the optimization adjustment scheme, performing optimization control of the emulsion pump station of the next monitoring node.

Images