Self-adaptive load balance multi-pump confluence control method and system

Abstract

The invention provides a self-adaptive load balance multi-pump confluence control method and system, and relates to the technical field of multi-pump control, and the method comprises the steps: collecting multiple pieces of confluence pump state information of a plurality of confluence pumps; acquiring operation load information; converting and averagely distributing the basic confluence flow to obtain a total confluence flow and a plurality of basic confluence flows; wear parameters of the multiple selected confluence pumps are balanced, noise and temperature of the multiple selected confluence pumps are minimized, and a confluence function is constructed; and based on the multiple basic confluence flows and the confluence functions, multi-pump confluence control optimization is conducted according to the state information of the multiple confluence pumps, an optimal confluence control parameter combination is obtained, and multi-pump confluence control is conducted on the multiple selected confluence pumps. The technical problems that in multi-pump confluence control in the prior art, accurate adjustment cannot be conducted according to the actual operation load and the state information of the pumps, load distribution of the pumps is uneven, control is inaccurate, the service life of the pumps is shortened, and the system stability is reduced are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-pump control, and in particular to a multi-pump confluence control method and system with adaptive load balancing. Background Art

[0002] In a multi-pump confluence system, how to accurately adjust the operating parameters of each pump is a major challenge. First, the traditional method cannot respond to load changes in real time, resulting in the system being unable to achieve optimal energy efficiency and stability; second, in a multi-pump system, the pump load is often uneven, some pumps may be overworked, while other pumps may be idle. This load imbalance leads to inefficient pump operation and increases the system's energy consumption and maintenance costs. At the same time, due to uneven load, some pumps may be subjected to excessive pressure and flow, resulting in excessive wear and noise problems, which not only affects the service life of the pump, but may also affect the overall stability and operating comfort of the system. Summary of the invention

[0003] The present application provides a multi-pump confluence control method and system with adaptive load balancing, aiming to solve the technical problem that in the prior art multi-pump confluence control cannot be accurately adjusted according to the actual operating load and pump status information, resulting in uneven load distribution and inaccurate control of the pump, thereby reducing the service life of the pump and the stability of the system.

[0004] The first aspect disclosed in the present application provides a multi-pump confluence control method with adaptive load balancing, the method is applied to a multi-pump confluence hydraulic device, the device includes multiple confluence pumps, the method includes: collecting multiple confluence pump status information of the multiple confluence pumps, wherein each confluence pump status information includes pump flow information and hydraulic oil temperature information; collecting the operation load information of the multi-pump confluence operation; according to the operation load information, converting and evenly distributing the basic confluence flow of the multiple confluence pumps to obtain the total confluence flow and multiple basic confluence flow of multiple selected confluence pumps; balancing the wear parameters of the multiple selected confluence pumps, minimizing the noise and temperature of the multiple selected confluence pumps, and constructing a confluence function for multi-pump confluence control; based on the multiple basic confluence flow and confluence function, according to the multiple confluence pump status information, optimizing the multi-pump confluence control, obtaining the optimal confluence control parameter combination, and performing multi-pump confluence control on the multiple selected confluence pumps.

[0005] The second aspect disclosed in the present application provides a multi-pump confluence control system with adaptive load balancing, the system is applied to a multi-pump confluence hydraulic device, the device includes multiple confluence pumps, the system is used for the above-mentioned adaptive load balancing multi-pump confluence control method, the system includes: a state information acquisition module, the state information acquisition module is used to collect multiple confluence pump state information of the multiple confluence pumps, wherein each confluence pump state information includes pump flow information and hydraulic oil temperature information; a load information acquisition module, the load information acquisition module is used to collect the operation load information of the multi-pump confluence operation; a confluence flow acquisition module, the confluence flow acquisition module is used to collect the operation load information of the multi-pump confluence operation according to the operation The invention discloses a confluence function module, which is used to construct a confluence function for multi-pump confluence control in order to balance the wear parameters of the multiple selected confluence pumps, minimize the noise and temperature of the multiple selected confluence pumps, and perform a confluence function for multi-pump confluence control; a multi-pump confluence control module, which is used to optimize multi-pump confluence control based on the multiple basic confluence flows and confluence functions and the multiple confluence pump status information, obtain the optimal confluence control parameter combination, and perform multi-pump confluence control on the multiple selected confluence pumps.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: By collecting the status information of multiple confluence pumps in real time, including pump flow and hydraulic oil temperature, as well as collecting the operation load information, the flow distribution of the pump can be automatically adjusted according to the actual load, which can ensure the load balance of the pump, reduce the overload of some pumps, and improve the working stability and efficiency of the system; by constructing the confluence function to balance the wear parameters, noise and temperature of the pump, the balanced use of the pump can be achieved in the multi-pump confluence control process, while reducing the excessive wear and noise of a single pump, extending the service life of the pump and improving the overall reliability of the system; by converting and distributing the basic confluence flow and optimizing the control in combination with the confluence function, the control parameters of the pump can be adjusted more accurately, which enables the system to dynamically adjust the operating status of the pump according to different operation loads, improving the energy efficiency and control accuracy of the system; through the iterative optimization algorithm, the control parameters are optimized based on the confluence function, and the optimal control parameter combination can be found under changing working conditions. Through cross-update and iterative adjustment, a dynamic and adaptive control optimization solution is provided to ensure that the system can maintain the best control effect under various load conditions, effectively improving the global optimization capability.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A schematic flow chart of a multi-pump confluence control method for adaptive load balancing provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of a multi-pump confluence control system with adaptive load balancing provided in an embodiment of the present application.

[0009] Explanation of reference numerals: state information acquisition module 10 , load information acquisition module 20 , confluent flow acquisition module 30 , confluent function construction module 40 , multi-pump confluent control module 50 . DETAILED DESCRIPTION

[0010] The embodiments of the present application provide a multi-pump confluence control method and system with adaptive load balancing, thereby solving the technical problem that in the prior art multi-pump confluence control cannot be accurately adjusted according to the actual operating load and pump status information, resulting in uneven load distribution and inaccurate control of the pumps, thereby reducing the service life of the pumps and the stability of the system.

[0011] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0012] Embodiment 1, as Figure 1 As shown, an embodiment of the present application provides a multi-pump confluence control method with adaptive load balancing, the method is applied to a multi-pump confluence hydraulic device, the device includes a plurality of confluence pumps, and the method includes: A plurality of merging pump status information of the plurality of merging pumps is collected, wherein each merging pump status information includes pump flow information and hydraulic oil temperature information.

[0013] The multi-pump confluence control method with adaptive load balancing provided in the embodiment of the present application is applied to a multi-pump confluence hydraulic device, wherein the device includes multiple confluence pumps, that is, multiple confluence pumps are connected to one hydraulic device to realize multi-pump confluence control. A flow sensor and a temperature sensor are installed on each confluence pump, and the flow of the hydraulic oil is monitored in real time by the flow sensor to obtain pump flow information. The pump flow information refers to the amount of hydraulic oil output by the hydraulic pump per unit time, usually in liters / minute. The pump flow information reflects the working efficiency of the pump and the operating status of the hydraulic system; the temperature of the hydraulic oil is monitored in real time by temperature sensing to obtain hydraulic oil temperature information. The hydraulic oil temperature information refers to the real-time temperature of the hydraulic oil, usually in degrees Celsius. The hydraulic oil temperature information reflects the heat load and heat dissipation of the hydraulic system. Excessive temperature may cause reduced efficiency of the hydraulic system and damage to the equipment. The pump flow information and hydraulic oil temperature information of each confluence pump are integrated to obtain multiple confluence pump status information.

[0014] Furthermore, collecting multiple confluence pump status information of the multiple confluence pumps includes: Collecting the current multiple pump flow information of the multiple confluence pumps; collecting the current multiple hydraulic oil temperature information of the multiple confluence pumps; integrating the multiple pump flow information and the multiple hydraulic oil temperature information to obtain multiple confluence pump status information.

[0015] Flow sensors are installed at the inlet and outlet of each combined pump or inside the pump body. These sensors can measure the flow rate of liquid through the pump in real time. The flow sensor is used to collect the flow data of each pump and summarize the flow information of multiple pumps. The pump flow information refers to the amount of hydraulic oil output by the hydraulic pump per unit time, usually in liters / minute. The pump flow information reflects the working efficiency of the pump and the operating status of the hydraulic system.

[0016] Temperature sensors are installed in the hydraulic oil circuit of each confluence pump. These sensors can measure the temperature of the hydraulic oil in real time. The temperature sensors are used to collect the hydraulic oil temperature data of each pump and summarize multiple hydraulic oil temperature information. The hydraulic oil temperature information refers to the real-time temperature of the hydraulic oil, usually in degrees Celsius. The hydraulic oil temperature information reflects the heat load and heat dissipation of the hydraulic system. Excessive temperature may lead to reduced efficiency of the hydraulic system and damage to the equipment.

[0017] Combine the flow information and hydraulic oil temperature information of each pump to generate a comprehensive status information, traverse multiple pumps, and obtain the status information of multiple confluence pumps to provide basic data for subsequent optimization control.

[0018] Collect the workload information of multi-pump confluence operation.

[0019] The operating load information refers to the load that the hydraulic system needs to bear in actual operation, including the weight and moving speed of the supported or lifted load. This information reflects the working requirements of the hydraulic system under specific operating conditions. Weight sensors and speed sensors are installed at key positions of the hydraulic system. The load weight is measured by the weight sensor, and the moving speed of the hydraulic actuator is measured by the speed sensor. The load weight and moving speed are read through the sensor interface to obtain the operating load information.

[0020] According to the workload information, the basic confluent flow rates of the plurality of confluent pumps are converted and evenly distributed to obtain a total confluent flow rate and a plurality of basic confluent flow rates of the plurality of selected confluent pumps.

[0021] According to the work load information, the total combined flow required by the load is converted and obtained. Specifically, the weight load and moving speed that need to be supported or lifted are analyzed. When the weight load is larger and the moving speed is faster, the required hydraulic flow is larger. The total combined flow is obtained based on the analysis results. The number of combined pumps required under the load is determined. Small loads do not require too many combined pumps to work, which reduces power consumption. Multiple selected combined pumps are randomly selected according to the number, and then the total combined flow is averaged to obtain multiple basic combined flows.

[0022] Furthermore, according to the workload information, the basic confluent flow rates of the plurality of confluent pumps are converted and evenly distributed to obtain a total confluent flow rate and a plurality of basic confluent flow rates of the plurality of selected confluent pumps, including: According to the workload information, a decision is made on the number of confluence pumps to be acquired, and a plurality of selected confluence pumps are randomly selected to obtain the number of confluence pumps; based on the historical operation data of multi-pump confluence, a sample workload information set and a sample total confluence flow set are collected; the sample workload information set and the sample total confluence flow set are used to train a total confluence flow converter; based on the total confluence flow converter, the workload is converted into a confluence flow to obtain a total confluence flow; the total confluence flow is allocated the number of confluence pumps to obtain a plurality of basic confluence flows.

[0023] According to the historical operation data of multi-pump confluence, a relationship model between load and flow demand is established based on the decision tree. The model is used to predict the required number of confluence pumps. The number of confluence pumps needs to be rounded according to the actual situation, and the required number of pumps is randomly selected from all available confluence pumps to carry out the operation, ensuring that the system resource utilization is optimized while meeting the operation requirements.

[0024] Extract data from historical operation records and retrieve historical operation data of multiple pumps, which includes record data of multiple pumps operation in the past period of time, including the operation load of historical operations and the corresponding total confluent flow, and extract a representative sample operation load information set, which includes operation load data, such as load weight, moving speed, etc., and a sample total confluent flow set, which includes confluent flow data corresponding to the sample operation load information.

[0025] Select a suitable model for training, such as support vector machine, neural network, etc., divide the data set into a training set and a test set. The training set is used to train the model, and the test set is used to verify the performance of the model. For example, the split ratio is 80% training set and 20% test set. Use the training set data to train the model so that it can learn the relationship between load information and total combined flow. Use the test set to evaluate the prediction performance of the model, check the accuracy, error and other performance indicators of the model, and adjust the model parameters to improve the prediction accuracy based on the feedback of the model performance. Cross-validation can be used to select the best model parameters. Finally, a model that meets the accuracy requirements is obtained as a total combined flow converter, which is used to predict the total combined flow required for job load information in real time.

[0026] The workload information is input into the total combined flow converter. The converter model uses the features and patterns learned during the training process to predict the total combined flow that matches the workload and outputs the required total combined flow, which represents the overall flow required by the system under a specific workload. The total combined flow is evenly distributed to multiple selected combined pumps. This can be achieved by simple averaging so that each pump handles an equal flow and obtains multiple basic combined flows.

[0027] Furthermore, according to the operation load information, the decision to obtain the number of confluence pumps includes: Based on the historical operation data of multi-pump confluence, a sample operation load information set and a sample number of confluence pumps are collected; using the sample operation load information set and the sample number of confluence pumps, a confluence pump activation quantity decision maker is constructed based on a decision tree; based on the confluence pump activation quantity decision maker, a decision is input on the operation load information to obtain the number of confluence pumps.

[0028] Data is extracted from historical operation records to obtain historical operation data of multi-pump confluence, which includes the operation load information of multi-pump confluence operations and the number of confluence pumps actually used. Representative operation loads and the number of pumps actually used for each set of operation load information are extracted to form a sample operation load information set and a sample number of confluence pumps.

[0029] The decision tree is a supervised learning model based on feature selection, which is used for classification and regression tasks. The model represents the decision process through a tree structure, where each node represents a feature, each branch represents a feature value, and the leaf node represents a predicted value. The workload information is extracted from the original data as a feature vector, including load weight, moving speed, etc. The number of confluence pumps is used as the target variable, and the preprocessed data set is input into the decision tree algorithm for training to generate a decision tree model. The accuracy and robustness of the model are evaluated using cross-validation technology, and finally a decision tree model that meets the accuracy requirements is obtained as the confluence pump activation quantity decider, which is used to predict new workload conditions to determine the number of confluence pumps that need to be activated.

[0030] The workload information is input into the combined flow pump activation quantity decider to predict the required combined flow pump quantity, which is used to adjust the quantity of activated pumps to adapt to the real-time workload conditions.

[0031] In order to balance the wear parameters of the multiple selected confluence pumps, minimize the noise and temperature of the multiple selected confluence pumps, and construct a confluence function for multi-pump confluence control.

[0032] In the process of multi-pump confluence control, the wear parameters of each confluence pump are calculated, and its distribution discreteness is obtained. The distribution discreteness can be represented by statistics such as standard deviation or variance. The ratio of the hydraulic oil temperature of each confluence pump to the preset temperature is calculated and the average value is obtained. The ratio of the noise of each confluence pump to the preset noise is calculated and the average value is obtained. The distribution discreteness of the wear parameters, the temperature ratio and the noise ratio are weighted to obtain the confluence function for multi-pump confluence control, so that in the process of multi-pump confluence control, the wear parameters of the confluence pumps can be balanced and the noise and temperature can be minimized.

[0033] Furthermore, in order to balance the wear parameters of the multiple selected confluence pumps and minimize the noise and temperature of the multiple selected confluence pumps, a confluence function for multi-pump confluence control is constructed, as shown in the following formula: ; Among them, MPM is the confluence fitness, , and is the weight, is the distribution discreteness information of multiple confluence pump wear parameters of multiple selected confluence pumps analyzed by combining multiple confluence control parameters with the multiple confluence pump state information, M is the number of the multiple selected confluence pumps, is the control hydraulic oil temperature of the i-th selected confluence pump under the confluence pump state information and confluence control parameters, To preset the hydraulic oil temperature, is the noise of the merging pump of the i-th selected merging pump under the merging pump state information and merging control parameters, The default noise.

[0034] Specifically, the confluence function is as follows: ; In this confluence function, the first term The logarithmic values ​​of the wear parameters were calculated, with weights Indicates the degree of influence on confluence fitness; the second The logarithm of the temperature ratio is calculated. The temperature ratio represents the degree of deviation between the controlled hydraulic oil temperature and the preset hydraulic oil temperature. Indicates the influence of temperature on confluence fitness; the third term The logarithm of the noise ratio was calculated. The noise ratio represents the degree of deviation between the noise of the combined pump and the preset noise. Indicates the influence of noise term on confluence fitness. Through this confluence function, the wear, temperature and noise of each confluence pump are converted into logarithmic values ​​and weighted summed to calculate the overall confluence fitness, which helps to smooth and standardize the influence of different parameters on system performance, and then evaluate the confluence fitness. The fitness value calculated by the confluence function helps to determine the pros and cons of the control parameters, thereby optimizing the control strategy of the pump.

[0035] Based on the multiple basic confluence flows and confluence functions, and according to the multiple confluence pump status information, multi-pump confluence control optimization is performed to obtain an optimal confluence control parameter combination, and multi-pump confluence control is performed on the multiple selected confluence pumps.

[0036] The obtained multiple basic confluence flow rates are used as initial control parameters, and crossover operations are introduced to improve the randomness of the solution and increase the global search capability. New control parameter combinations are generated through crossover operations, and the confluence function is used to evaluate its fitness. Using crossover operations and confluence fitness functions, iterative optimization is performed on the basis of multiple basic confluence flow rates to gradually obtain better control parameter combinations, and finally the confluence control parameter combination with the largest confluence fitness is output as the optimal confluence control parameter combination. The obtained optimal confluence control parameter combination is used to perform multi-pump confluence control on multiple selected confluence pumps to achieve optimal control of multi-pump confluence.

[0037] Furthermore, based on the multiple basic confluence flows and confluence functions, and according to the multiple confluence pump status information, multi-pump confluence control optimization is performed to obtain an optimal confluence control parameter combination, including: Taking the multiple basic confluence pump flows as adjustment directions, the pump flow information of the multiple selected confluence pumps is adjusted to obtain multiple adjustment pump flow information, and the total confluence flow is distributed according to the proportion of the adjustment pump flow information to obtain multiple first confluence pump flows as a first confluence control parameter combination; based on the first confluence control parameter combination, combined with the hydraulic oil temperature information in the confluence pump status information of the multiple selected confluence pumps, aging prediction, temperature prediction and noise prediction are performed respectively, and based on the confluence function, the first confluence fitness is calculated and obtained; the multiple first confluence pump flows are randomly cross-updated to obtain multiple first cross-confluence control parameter combinations, combined with the hydraulic oil temperature information in the confluence pump status information of the multiple selected confluence pumps, aging prediction, temperature prediction and noise prediction are performed respectively, and based on the confluence function, multiple first cross-confluence fitness are calculated and obtained; continue to take the multiple basic confluence pump flows as adjustment directions, adjust the multiple adjustment pump flow information, perform iterative optimization until convergence, output the confluence control parameter combination with the largest confluence fitness, and obtain the optimal confluence control parameter combination.

[0038] The flow rates of multiple basic confluence pumps are used as adjustment directions. These flow rates are the initial flow rates of each confluence pump obtained by preliminary calculation. These flow rates are adjusted. The adjustment can be to increase or decrease the flow rate of the pump to optimize the performance. After the adjustment, multiple adjustment pump flow rate information is obtained. The multiple adjustment pump flow rate information is proportionally calculated, and the total confluence flow rate is allocated according to the proportion, so that the flow rate of each pump after adjustment is consistent with the total confluence flow rate. The final flow rate is used as the first confluence control parameter combination for subsequent control and optimization.

[0039] Use historical data and prediction models to predict the aging status of each pump. The aging prediction model can be built based on the pump's operating time, load, and historical aging data. Use hydraulic oil temperature information and control parameter combinations to predict the temperature change of each pump under given control conditions. A regression model can be used for temperature prediction. Use noise data and control parameter combinations to predict the noise level of the pump. The noise prediction model can be trained based on noise data and operating status. Substitute the aging, temperature, and noise prediction values, as well as the temperature and noise preset values, into the confluence function. The confluence function performs weighted calculations based on multiple objectives, including minimum noise, optimal temperature, and minimum aging, to obtain the first confluence fitness of the first confluence control parameter combination to evaluate the performance of the control parameter combination.

[0040] Through random crossover operation, a new combination of confluence control parameters is generated, and new solutions are explored by exchanging the flow values ​​of the two pumps, thereby enhancing the global optimization capability and randomness. Specifically, two pumps are randomly selected from the selected confluence pumps, and the flow values ​​of the two pumps are exchanged to generate a new flow combination. The updated flow combination is recorded as the first cross-confluence control parameter combination, and the evaluation process of the aforementioned steps is repeated. Aging prediction, temperature prediction and noise prediction are performed on the first cross-confluence control parameter combination, and multiple first cross-confluence fitness values ​​are calculated based on the confluence function.

[0041] Set the number of iterations, convergence threshold and other parameters of the optimization algorithm. In each iteration, repeat the aforementioned adjustment and crossover steps to adjust the flow parameters of the pumps. For each adjusted flow combination, use the confluence function to calculate the new confluence fitness. According to the newly calculated fitness value, select the flow combination with the largest fitness as the current optimal parameter combination. If the optimization algorithm meets the convergence conditions, such as the fitness change is less than the set threshold, or the maximum number of iterations is reached, stop the optimization process; otherwise, continue to adjust the flow and proceed to the next iteration. Output the final optimal confluence control parameter combination, that is, the flow combination with the maximum confluence fitness. This combination is used in the actual multi-pump confluence control to achieve the best system performance.

[0042] Further, based on the first merging control parameter combination, combined with the hydraulic oil temperature information in the merging pump state information of the multiple selected merging pumps, aging prediction, temperature prediction and noise prediction are performed respectively, and based on the merging function, a first merging fitness is calculated, including: Based on the test data of the confluence pump, a sample confluence control parameter set and a sample hydraulic oil temperature information set are collected, and a sample wear parameter set, a sample control hydraulic oil temperature set and a sample confluence pump noise set are collected; the sample confluence control parameter set and the sample hydraulic oil temperature information set are used as input data, and the sample wear parameter set, the sample control hydraulic oil temperature set and the sample confluence pump noise set are used as supervision data to train a confluence predictor; the confluence predictor is used to perform confluence control prediction on multiple first confluence control parameters in the first confluence control parameter combination in combination with the hydraulic oil temperature information in the confluence pump status information of the multiple selected confluence pumps, so as to obtain multiple first wear parameters, multiple first control hydraulic oil temperatures and multiple first confluence pump noises; the first aging distribution discreteness information of the multiple first wear parameters is calculated, and the first confluence fitness is calculated based on the confluence function.

[0043] In a laboratory environment, the confluence pump is tested and data is collected. The obtained sample confluence control parameter set is a data set of confluence control parameters recorded under different test conditions; the sample hydraulic oil temperature information set is a data set of hydraulic oil temperature recorded during the test, and these data are used to analyze the temperature performance of the pump under different conditions; the sample wear parameter set is the pump wear-related data recorded during the test, including the degree of pump wear, wear rate, etc.; the sample control hydraulic oil temperature set is the hydraulic oil temperature data recorded under different control parameters, and these data are used to analyze the influence of control parameters on oil temperature; the sample confluence pump noise set is the pump noise data recorded during the test, and is used to analyze the relationship between noise level, control parameters and load.

[0044] Select a suitable prediction model type, such as support vector machine, random forest or neural network, select important features from the input data, construct a feature matrix, use sample confluence control parameters and sample hydraulic oil temperature information as input data, use sample wear parameters, sample control hydraulic oil temperature and sample confluence pump noise as supervision data, input the input data and supervision data into the model, adjust the model parameters through optimization algorithms, such as gradient descent, to minimize the prediction error, use cross-validation technology to evaluate the performance of the model, ensure the generalization ability of the model on different data sets, adjust the model's hyperparameters according to the evaluation results to improve the prediction performance, and obtain a prediction model that meets the accuracy requirements as a confluence predictor, which can predict the wear, hydraulic oil temperature and noise of the confluence pump based on the input data, and provide support for the optimization of multi-pump confluence control.

[0045] Each control parameter in the first confluence control parameter combination is combined with the hydraulic oil temperature of the selected confluence pump as input data and input into the confluence predictor. The predictor is used to predict the wear condition of each selected confluence pump, predict the hydraulic oil temperature of each selected confluence pump, and predict the noise level of each selected confluence pump to obtain multiple first wear parameters, multiple first control hydraulic oil temperatures, and multiple first confluence pump noises.

[0046] For multiple first wear parameters, the mean of the wear parameters is first calculated, and then the variance of the wear parameters is calculated based on the mean. This variance describes the distribution discreteness of the wear parameters. The discreteness reflects the variation range and consistency of the wear parameters. The smaller the variance, the smaller and more consistent the variation range of the wear parameters. Substitute the first aging distribution discreteness information into the confluence function, and combine it with other parameter items to calculate the first confluence fitness.

[0047] In summary, the multi-pump confluence control method with adaptive load balancing provided in the embodiment of the present application has the following technical effects: By collecting the status information of multiple confluence pumps in real time, including pump flow and hydraulic oil temperature, as well as collecting the operation load information, the flow distribution of the pump can be automatically adjusted according to the actual load, which can ensure the load balance of the pump, reduce the overload of some pumps, and improve the working stability and efficiency of the system; by constructing the confluence function to balance the wear parameters, noise and temperature of the pump, the balanced use of the pump can be achieved in the multi-pump confluence control process, while reducing the excessive wear and noise of a single pump, extending the service life of the pump and improving the overall reliability of the system; by converting and distributing the basic confluence flow and optimizing the control in combination with the confluence function, the control parameters of the pump can be adjusted more accurately, which enables the system to dynamically adjust the operating status of the pump according to different operation loads, improving the energy efficiency and control accuracy of the system; through the iterative optimization algorithm, the control parameters are optimized based on the confluence function, and the optimal control parameter combination can be found under changing working conditions. Through cross-update and iterative adjustment, a dynamic and adaptive control optimization solution is provided to ensure that the system can maintain the best control effect under various load conditions, effectively improving the global optimization capability.

[0048] Embodiment 2 is based on the same inventive concept as the multi-pump confluence control method for adaptive load balancing in the above embodiment. Figure 2 As shown, the embodiment of the present application provides a multi-pump confluence control system with adaptive load balancing, the system is applied to a multi-pump confluence hydraulic device, the device includes a plurality of confluence pumps, and the system includes: A state information acquisition module 10, the state information acquisition module 10 is used to collect state information of multiple confluence pumps of the multiple confluence pumps, wherein each confluence pump state information includes pump flow information and hydraulic oil temperature information; a load information acquisition module 20, the load information acquisition module 20 is used to collect the operation load information of the multi-pump confluence operation; a confluence flow acquisition module 30, the confluence flow acquisition module 30 is used to convert and evenly distribute the basic confluence flow of the multiple confluence pumps according to the operation load information, and obtain the total confluence flow and multiple selected confluence pumps. A plurality of basic confluent flows of the pumps; a confluent function construction module 40, wherein the confluent function construction module 40 is used to balance the wear parameters of the plurality of selected confluent pumps, minimize the noise and temperature of the plurality of selected confluent pumps, and construct a confluent function for multi-pump confluent control; a multi-pump confluent control module 50, wherein the multi-pump confluent control module 50 is used to optimize the multi-pump confluent control based on the plurality of basic confluent flows and confluent functions and according to the plurality of confluent pump status information, obtain the optimal confluent control parameter combination, and perform multi-pump confluent control on the plurality of selected confluent pumps.

[0049] Furthermore, the system further includes a confluence pump status information acquisition module to perform the following operation steps: Collecting the current multiple pump flow information of the multiple confluence pumps; collecting the current multiple hydraulic oil temperature information of the multiple confluence pumps; integrating the multiple pump flow information and the multiple hydraulic oil temperature information to obtain multiple confluence pump status information.

[0050] Furthermore, the system further includes a basic confluent flow acquisition module to perform the following operation steps: According to the workload information, a decision is made on the number of confluence pumps to be acquired, and a plurality of selected confluence pumps are randomly selected to obtain the number of confluence pumps; based on the historical operation data of multi-pump confluence, a sample workload information set and a sample total confluence flow set are collected; the sample workload information set and the sample total confluence flow set are used to train a total confluence flow converter; based on the total confluence flow converter, the workload is converted into a confluence flow to obtain a total confluence flow; the total confluence flow is allocated the number of confluence pumps to obtain a plurality of basic confluence flows.

[0051] Furthermore, the system further includes a confluence pump quantity acquisition module to perform the following operation steps: Based on the historical operation data of multi-pump confluence, a sample operation load information set and a sample number of confluence pumps are collected; using the sample operation load information set and the sample number of confluence pumps, a confluence pump activation quantity decision maker is constructed based on a decision tree; based on the confluence pump activation quantity decision maker, a decision is input on the operation load information to obtain the number of confluence pumps.

[0052] Furthermore, in order to balance the wear parameters of the multiple selected confluence pumps and minimize the noise and temperature of the multiple selected confluence pumps, a confluence function for multi-pump confluence control is constructed, as shown in the following formula: ; Among them, MPM is the confluence fitness, , and is the weight, is the distribution discreteness information of multiple confluence pump wear parameters of multiple selected confluence pumps analyzed by combining multiple confluence control parameters with the multiple confluence pump state information, M is the number of the multiple selected confluence pumps, is the control hydraulic oil temperature of the i-th selected confluence pump under the confluence pump state information and confluence control parameters, To preset the hydraulic oil temperature, is the noise of the merging pump of the i-th selected merging pump under the merging pump state information and merging control parameters, The default noise.

[0053] Furthermore, the system further includes an optimal confluence control parameter combination acquisition module to perform the following operation steps: Taking the multiple basic confluence pump flows as adjustment directions, the pump flow information of the multiple selected confluence pumps is adjusted to obtain multiple adjustment pump flow information, and the total confluence flow is distributed according to the proportion of the adjustment pump flow information to obtain multiple first confluence pump flows as a first confluence control parameter combination; based on the first confluence control parameter combination, combined with the hydraulic oil temperature information in the confluence pump status information of the multiple selected confluence pumps, aging prediction, temperature prediction and noise prediction are performed respectively, and based on the confluence function, the first confluence fitness is calculated and obtained; the multiple first confluence pump flows are randomly cross-updated to obtain multiple first cross-confluence control parameter combinations, combined with the hydraulic oil temperature information in the confluence pump status information of the multiple selected confluence pumps, aging prediction, temperature prediction and noise prediction are performed respectively, and based on the confluence function, multiple first cross-confluence fitness are calculated and obtained; continue to take the multiple basic confluence pump flows as adjustment directions, adjust the multiple adjustment pump flow information, perform iterative optimization until convergence, output the confluence control parameter combination with the largest confluence fitness, and obtain the optimal confluence control parameter combination.

[0054] Furthermore, the system further includes a first confluence fitness acquisition module to perform the following operation steps: Based on the test data of the confluence pump, a sample confluence control parameter set and a sample hydraulic oil temperature information set are collected, and a sample wear parameter set, a sample control hydraulic oil temperature set and a sample confluence pump noise set are collected; the sample confluence control parameter set and the sample hydraulic oil temperature information set are used as input data, and the sample wear parameter set, the sample control hydraulic oil temperature set and the sample confluence pump noise set are used as supervision data to train a confluence predictor; the confluence predictor is used to perform confluence control prediction on multiple first confluence control parameters in the first confluence control parameter combination in combination with the hydraulic oil temperature information in the confluence pump status information of the multiple selected confluence pumps, so as to obtain multiple first wear parameters, multiple first control hydraulic oil temperatures and multiple first confluence pump noises; the first aging distribution discreteness information of the multiple first wear parameters is calculated, and the first confluence fitness is calculated based on the confluence function.

[0055] Through the above-mentioned detailed description of the multi-pump confluence control method with adaptive load balancing in this specification, those skilled in the art can clearly understand the multi-pump confluence control system with adaptive load balancing in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0056] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-pump confluence control method with adaptive load balancing, characterized in that: The method is applied to a multi-pump confluence hydraulic device, the device includes a plurality of confluence pumps, and the method includes: Collecting multiple confluence pump status information of the multiple confluence pumps, wherein each confluence pump status information includes pump flow information and hydraulic oil temperature information; Collecting the workload information of multi-pump confluence operation; According to the workload information, converting and evenly distributing the basic confluent flow rates of the plurality of confluent pumps to obtain a total confluent flow rate and a plurality of basic confluent flow rates of the plurality of selected confluent pumps; To balance the wear parameters of the plurality of selected confluence pumps, minimize the noise and temperature of the plurality of selected confluence pumps, and construct a confluence function for performing multi-pump confluence control; Based on the multiple basic confluence flows and confluence functions, and according to the multiple confluence pump status information, multi-pump confluence control optimization is performed to obtain an optimal confluence control parameter combination, and multi-pump confluence control is performed on the multiple selected confluence pumps.

2. The multi-pump confluence control method with adaptive load balancing according to claim 1, characterized in that: Collecting multiple confluence pump status information of the multiple confluence pumps includes: Collecting current flow information of multiple pumps of the multiple confluence pumps; Collecting current multiple hydraulic oil temperature information of the multiple confluence pumps; The plurality of pump flow information and the plurality of hydraulic oil temperature information are integrated to obtain the plurality of confluence pump status information.

3. The multi-pump confluence control method with adaptive load balancing according to claim 1, characterized in that: According to the workload information, the basic confluent flow rates of the plurality of confluent pumps are converted and evenly distributed to obtain a total confluent flow rate and a plurality of basic confluent flow rates of the plurality of selected confluent pumps, including: According to the workload information, a decision is made to obtain the number of merging pumps, and a plurality of selected merging pumps having the number of merging pumps are randomly selected; Based on the historical operation data of multi-pump confluence, collect sample operation load information set and sample total confluence flow set; Using the sample workload information set and the sample total combined flow set to train a total combined flow converter; Performing confluent flow conversion on the workload based on the total confluent flow converter to obtain a total confluent flow; The total confluent flow is allocated according to the number of confluent pumps to obtain a plurality of basic confluent flows.

4. The multi-pump confluence control method with adaptive load balancing according to claim 3 is characterized in that: According to the work load information, a decision is made to obtain the number of confluence pumps, including: Based on the historical operation data of multi-pump confluence, collect the sample operation load information set and the number of sample confluence pumps; Using the sample workload information set and the number of sample confluence pumps, based on a decision tree, a confluence pump activation number decision maker is constructed; Based on the combined pump activation quantity decider, an input decision is made on the work load information to obtain the number of combined pumps.

5. The multi-pump confluence control method with adaptive load balancing according to claim 1, characterized in that: In order to balance the wear parameters of the multiple selected confluence pumps and minimize the noise and temperature of the multiple selected confluence pumps, a confluence function for multi-pump confluence control is constructed, as shown in the following formula: ； Among them, MPM is the confluence fitness, , and is the weight, is the distribution discreteness information of multiple confluence pump wear parameters of multiple selected confluence pumps analyzed by combining multiple confluence control parameters with the multiple confluence pump state information, M is the number of the multiple selected confluence pumps, is the control hydraulic oil temperature of the i-th selected confluence pump under the confluence pump state information and confluence control parameters, To preset the hydraulic oil temperature, is the noise of the merging pump of the i-th selected merging pump under the merging pump state information and merging control parameters, The default noise.

6. The multi-pump confluence control method with adaptive load balancing according to claim 1, characterized in that: Based on the multiple basic confluence flows and confluence functions, and according to the multiple confluence pump status information, multi-pump confluence control optimization is performed to obtain an optimal confluence control parameter combination, including: Taking the multiple basic confluence pump flows as adjustment directions, adjusting the pump flow information of the multiple selected confluence pumps to obtain multiple adjusted pump flow information, distributing the total confluence flow according to the proportion of the adjusted pump flow information, and obtaining multiple first confluence pump flows as a first confluence control parameter combination; Based on the first merging control parameter combination, combined with the hydraulic oil temperature information in the merging pump status information of the multiple selected merging pumps, aging prediction, temperature prediction and noise prediction are performed respectively, and based on the merging function, a first merging fitness is calculated; The flow rates of the multiple first confluence pumps are randomly cross-updated to obtain multiple first cross-confluence control parameter combinations, and aging prediction, temperature prediction and noise prediction are performed respectively in combination with the hydraulic oil temperature information in the confluence pump state information of the multiple selected confluence pumps, and multiple first cross-confluence fitness degrees are calculated based on the confluence function; Continue to use the multiple basic confluence pump flow rates as adjustment directions, adjust the multiple adjustment pump flow information, perform iterative optimization until convergence, output the confluence control parameter combination with the maximum confluence fitness, and obtain the optimal confluence control parameter combination.

7. The multi-pump confluence control method with adaptive load balancing according to claim 6, characterized in that: Based on the first merging control parameter combination, combined with the hydraulic oil temperature information in the merging pump state information of the multiple selected merging pumps, aging prediction, temperature prediction and noise prediction are performed respectively, and based on the merging function, a first merging fitness is calculated, including: Based on the test data of the confluence pump, a sample confluence control parameter set, a sample hydraulic oil temperature information set, a sample wear parameter set, a sample control hydraulic oil temperature set and a sample confluence pump noise set are collected; Using the sample confluence control parameter set and the sample hydraulic oil temperature information set as input data, and using the sample wear parameter set, the sample control hydraulic oil temperature set and the sample confluence pump noise set as supervision data to train a confluence predictor; The confluence predictor is used to respectively combine the plurality of first confluence control parameters in the first confluence control parameter combination with the hydraulic oil temperature information in the confluence pump state information of the plurality of selected confluence pumps to perform confluence control prediction, and obtain a plurality of first wear parameters, a plurality of first control hydraulic oil temperatures and a plurality of first confluence pump noises; The discreteness information of the first aging distribution of the plurality of first wear parameters is obtained by calculation, and the first confluence fitness is obtained by calculation based on the confluence function.

8. Adaptive load balancing multi-pump confluence control system, characterized in that: The system is applied to a multi-pump confluence hydraulic device, the device includes a plurality of confluence pumps, and is used to implement the multi-pump confluence control method for adaptive load balancing according to any one of claims 1 to 7. The system includes: A status information acquisition module, wherein the status information acquisition module is used to acquire status information of multiple confluence pumps of the multiple confluence pumps, wherein each confluence pump status information includes pump flow information and hydraulic oil temperature information; A load information collection module, which is used to collect operation load information of multi-pump confluence operation; A confluent flow acquisition module, the confluent flow acquisition module is used to convert and evenly distribute the basic confluent flows of the multiple confluent pumps according to the workload information, to obtain a total confluent flow and multiple basic confluent flows of multiple selected confluent pumps; A confluence function building module, the confluence function building module is used to balance the wear parameters of the multiple selected confluence pumps, minimize the noise and temperature of the multiple selected confluence pumps, and build a confluence function for multi-pump confluence control; A multi-pump confluence control module, the multi-pump confluence control module is used to optimize the multi-pump confluence control based on the multiple basic confluence flows and confluence functions and according to the multiple confluence pump status information, obtain the optimal confluence control parameter combination, and perform multi-pump confluence control on the multiple selected confluence pumps.

Images