A power distribution method for an integrated dual-pump permanent magnet synchronous motor and the permanent magnet synchronous motor thereof

Abstract

This invention provides an integrated dual-pump permanent magnet synchronous motor power distribution method and its permanent magnet synchronous motor, relating to the field of new energy vehicle technology. During the lifting operation, real-time operating parameters and real-time lifting acceleration of the gear pump are collected, and a mapping relationship is used to determine the real-time predicted value of the load mass. Historical values ​​of the load mass are considered to determine the real-time value. Then, the real-time target values ​​of the operating parameters of the gear pump and vane pump, and the real-time power distribution ratio are determined based on the real-time vehicle speed. Finally, a PID control algorithm is used to analyze the real-time target values ​​of the operating parameters and the real-time power distribution ratio, and the motor speed is adjusted in real-time. This achieves the technical effect of balancing the load requirements of both the steering pump and the lifting pump. Furthermore, when determining the load mass, both the real-time predicted value and historical values ​​of the load mass are comprehensively considered to ensure the accuracy of the load mass calculation.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle technology, specifically to an integrated dual-pump permanent magnet synchronous motor power distribution method and its permanent magnet synchronous motor. Background Technology

[0002] With the rapid development of new energy vehicles, higher requirements have been placed on the efficiency, integration, and spatial layout of various vehicle components. Currently, the hydraulic steering system and hydraulic lifting system (such as truck bed lifting) of new energy commercial vehicles (such as trucks and dump trucks) are usually driven by two independent power sources. The steering pump is driven by an electric power steering system or a traditional engine (in hybrid models), while the lifting pump may require an additional motor or power take-off. This structural design has the disadvantages of large space occupation and high cost.

[0003] Currently, a dual-pump motor structure has been proposed in the prior art, which integrates the steering pump and the lifting pump onto the same motor, effectively solving the shortcomings of the existing independent dual-pump structure, such as large space occupation and high cost. However, since the dual-pump motor structure drives the steering pump and the lifting pump to work together through a single motor shaft, how to set a reasonable motor speed to take into account the load requirements of both the steering pump and the lifting pump remains a technical problem that urgently needs to be solved.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide an integrated dual-pump permanent magnet synchronous motor power distribution method and a permanent magnet synchronous motor thereof, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A power distribution method for an integrated dual-pump permanent magnet synchronous motor, used to adjust the speed of the integrated dual-pump permanent magnet synchronous motor to achieve power distribution between the gear pump and the vane pump, includes the following steps:

[0008] S1, obtain the lifting acceleration of the weight when the gear pump provides lifting force for lifting the same mass weight under different working parameters, so as to establish the mapping relationship between working parameters, lifting acceleration and weight mass. Working parameters include hydraulic flow rate and working pressure.

[0009] S2, during the lifting operation, collect the real-time operating parameters of the gear pump and the real-time lifting acceleration, and combine the mapping relationship to determine the real-time predicted value of the load mass, and combine the historical value of the load mass to determine the real-time value of the load mass;

[0010] S3. Based on the real-time value of the weight, determine the real-time target values ​​of the working parameters and the real-time target value of the power of the gear pump. Based on the real-time value of the weight and the real-time value of the vehicle speed, determine the real-time target values ​​of the working parameters and the real-time target value of the power of the vane pump. Combine the two real-time target values ​​of power to determine the real-time power distribution ratio of the gear pump and the vane pump.

[0011] S4 analyzes the real-time target values ​​of the operating parameters of the gear pump and vane pump based on the PID control algorithm, and adjusts the motor speed in real time in combination with the real-time power distribution ratio.

[0012] Furthermore, a mapping relationship between working parameters, lifting acceleration, and weight mass is constructed based on a deep neural network. This includes an input layer for receiving hydraulic flow, working pressure, and lifting acceleration; a hidden layer for extracting features from hydraulic flow, working pressure, and lifting acceleration; and an output layer for outputting the weight mass.

[0013] Furthermore, during the lifting operation, the logic for obtaining the real-time value of the load mass is as follows: a preset time length threshold is used as the reference point. The time interval from the start of the lifting operation to the reference point is calculated. If this time interval is less than the time length threshold, the time interval from the start of the lifting operation to the reference point is used as the reference interval. Otherwise, the reference point is used as the endpoint, and a time interval of equal length to the time length threshold is traced back as the reference interval. The median of the historical load mass values ​​at each sampling point within the reference interval is extracted, and the coefficient of variation of the historical load mass values ​​at each sampling point within the reference interval is calculated. The fusion weight is determined based on the relative error and coefficient of variation between the median of the historical load mass values ​​and the real-time predicted load mass values. The product of the median of the historical load mass values ​​and the fusion weight is used as the first weighting value. The difference between the value 1 and the fusion weight is calculated, and the product of this difference and the real-time predicted load mass values ​​is used as the second weighting value. The sum of the first weighting value and the second weighting value is used as the real-time load mass value.

[0014] Furthermore, the logic for determining the fusion weights is as follows:

[0015] After performing maximum-minimum normalization on the historical values ​​of the weight of the heavy object at each sampling point within the reference interval, the coefficient of variation is calculated. Combining the coefficient of variation with the preset high and low thresholds of the coefficient of variation, the fluctuation state of the reference interval is determined, which is a low fluctuation state, a medium fluctuation state, or a high fluctuation state, and the high threshold of the coefficient of variation is greater than the low threshold of the coefficient of variation.

[0016] By combining the relative error between the median of the historical weight and the real-time predicted weight, and the preset high and low relative error thresholds, the degree of balance between the real-time predicted weight and the median of the historical weight is determined, which is low, medium, or high balance, and the high relative error threshold is greater than the low relative error threshold.

[0017] Set the base weight of the fusion weight to 0.5, and set a base weight correction value between 0 and 0.5. When the fluctuation state is low, calculate the sum of the base weight of the fusion weight and the base weight correction value as the fusion weight correction value. When the fluctuation state is medium, use the base weight of the fusion weight as the fusion weight correction value. When the fluctuation state is high, calculate the difference between the base weight of the fusion weight and the base weight correction value as the fusion weight correction value.

[0018] A secondary correction value, less than 0.5, is preset to be the difference between the base item correction value and the base item correction value. The fusion weight correction item is then adjusted in a secondary manner based on the secondary correction value, the fluctuation state, and the degree of equilibrium, so as to determine the fusion weight.

[0019] Furthermore, the logic for determining the fluctuation state is as follows: if the coefficient of variation is not greater than the low threshold of the coefficient of variation, the reference interval is defined as a low fluctuation state; if the coefficient of variation is not less than the high threshold of the coefficient of variation, the reference interval is defined as a high fluctuation state; if the coefficient of variation is between the high threshold of the coefficient of variation and the low threshold of the coefficient of variation, the reference interval is defined as a medium fluctuation state.

[0020] The logic for determining the balance level is as follows: if the relative error is not greater than the low threshold of relative error, the balance level is set to high balance level; if the relative error is not less than the high threshold of relative error, the balance level is set to low balance level; if the relative error is between the high threshold of relative error and the low threshold of relative error, the balance level is set to medium balance level.

[0021] Furthermore, the fusion weight correction term is further corrected a second time to determine the fusion weight. The logic is as follows:

[0022] If the volatility state is low volatility, then if the equilibrium level is low equilibrium, the sum of the fusion weight correction term and the second correction value is calculated as the fusion weight. If the equilibrium level is medium equilibrium, the sum of the fusion weight correction term and 0.5 times the second correction value is calculated as the fusion weight. If the equilibrium level is high equilibrium, the fusion weight correction term under medium volatility is used as the fusion weight.

[0023] If the volatility state is not a low volatility state, then the fusion weight correction term under the high volatility state will be used as the fusion weight.

[0024] Furthermore, the logic for determining the real-time target values ​​of the working parameters is as follows: taking the mass of the object as the independent variable, and the target values ​​of the hydraulic flow and working pressure of the gear pump as the dependent variables, the mathematical expressions for the mass of the object and the target values ​​of the hydraulic flow and the working pressure of the gear pump are fitted based on polynomial fitting. The real-time value of the mass of the object is then substituted into the above two mathematical expressions to determine the real-time target values ​​of the hydraulic flow and the working pressure of the gear pump.

[0025] Using the mass of the load and the vehicle speed as independent variables, and the target values ​​of the hydraulic flow rate and working pressure of the vane pump as dependent variables, the mathematical expressions for the mass of the load, the vehicle speed, and the target values ​​of the hydraulic flow rate of the vane pump, as well as the mathematical expressions for the mass of the load, the vehicle speed, and the target values ​​of the working pressure of the vane pump, are fitted using a polynomial. The real-time values ​​of the mass of the load and the real-time values ​​of the vehicle speed are then substituted into the above two mathematical expressions to determine the real-time target values ​​of the hydraulic flow rate and the real-time target values ​​of the working pressure of the vane pump.

[0026] The real-time target power value of the gear pump is the product of its real-time target hydraulic flow rate and its real-time target working pressure value, and the real-time target power value of the vane pump is the product of its real-time target hydraulic flow rate and its real-time target working pressure value.

[0027] The calculation logic for the real-time power distribution ratio of gear pumps and vane pumps is as follows: calculate the sum of the real-time power distribution ratios of gear pumps and vane pumps, calculate the ratio of the real-time power distribution ratio of gear pumps to the sum, and use this ratio as the real-time power distribution ratio of gear pumps.

[0028] Furthermore, the logic for real-time adjustment of the motor speed is as follows:

[0029] 1) Calculate the difference between the real-time target value of the gear pump hydraulic flow and the real-time value of the gear pump hydraulic flow, and take the ratio of this difference to the real-time target value of the gear pump hydraulic flow as the real-time error of the gear pump hydraulic flow. Calculate the difference between the real-time target value of the gear pump working pressure and the real-time value of the gear pump working pressure, and take the ratio of this difference to the real-time target value of the gear pump working pressure as the real-time error of the gear pump working pressure. Sum the real-time error of the gear pump hydraulic flow and the real-time error of the gear pump working pressure as the real-time error of the gear pump.

[0030] 2) Calculate the difference between the real-time target value of the hydraulic flow rate of the vane pump and the real-time value of the hydraulic flow rate of the vane pump, and take the ratio of this difference to the real-time target value of the hydraulic flow rate of the vane pump as the real-time error of the hydraulic flow rate of the vane pump. Calculate the difference between the real-time target value of the working pressure of the vane pump and the real-time value of the working pressure of the vane pump, and take the ratio of this difference to the real-time target value of the working pressure of the vane pump as the real-time error of the working pressure of the vane pump. Sum the real-time error of the hydraulic flow rate of the vane pump and the real-time error of the working pressure of the vane pump as the real-time error of the vane pump.

[0031] 3) Calculate the product of the real-time power distribution ratio of the gear pump and the real-time error of the gear pump, and calculate the product of the real-time power distribution ratio of the vane pump and the real-time error of the vane pump. Sum the two products as the comprehensive real-time error.

[0032] 4) Using the comprehensive real-time error as the input variable, the PID control algorithm outputs a real-time speed adjustment signal to optimize the motor speed in real time.

[0033] A permanent magnet synchronous motor, using the aforementioned integrated dual-pump permanent magnet synchronous motor power distribution method, includes:

[0034] The main body of the permanent magnet synchronous motor includes the stator assembly, the rotor assembly, and the motor shaft;

[0035] A vane pump, whose pump shaft is directly connected to one end of the motor shaft via a first coupling, is used to provide steering assistance to the vehicle's power steering system.

[0036] A gear pump, whose pump shaft is directly connected to the other end of the motor shaft via a spline, is used to provide lifting force for the vehicle's lifting system.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] The integrated dual-pump permanent magnet synchronous motor power distribution method and its permanent magnet synchronous motor of the present invention first determine the mass of the load by using the operating parameters of the gear pump and the lifting acceleration, then determine the target values ​​of the operating parameters of the gear pump and the vane pump by combining the vehicle speed, and finally adjust the motor speed based on the PID control algorithm. This achieves the technical effect of taking into account the load requirements of both the steering pump and the lifting pump. Furthermore, when determining the mass of the load, the real-time predicted value and historical value of the mass of the load are comprehensively considered to ensure the accuracy of the mass calculation. This provides a theoretical basis for the subsequent accurate determination of the target values ​​of the operating parameters and further ensures that the load requirements of both the steering pump and the lifting pump are taken into account. Attached Figure Description

[0039] Figure 1 This is a schematic flowchart of the overall method of the present invention;

[0040] Figure 2 This is a diagram showing the comparison between the real-time value and the actual value of the weight. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0042] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0043] Example 1:

[0044] Please see Figure 1-2 This invention provides a power distribution method for an integrated dual-pump permanent magnet synchronous motor, used to adjust the speed of the integrated dual-pump permanent magnet synchronous motor to achieve power distribution between the gear pump and the vane pump, comprising the following steps:

[0045] S1, obtain the lifting acceleration of the weight when the gear pump provides lifting force for lifting the same mass weight under different working parameters, so as to establish the mapping relationship between working parameters, lifting acceleration and weight mass. Working parameters include hydraulic flow rate and working pressure.

[0046] The model utilizes a deep neural network to construct a mapping relationship between working parameters, lifting acceleration, and the mass of the load. It includes an input layer for receiving hydraulic flow, working pressure, and lifting acceleration; a hidden layer for extracting features from these parameters; and an output layer for outputting the load mass. The input layer has three nodes for receiving hydraulic flow, working pressure, and lifting acceleration, respectively. The hidden layer can have 4-7 layers, with the number of nodes in each layer adjusted according to the specific requirements (e.g., 32, 64, or 128 nodes). The ReLU function is used to enhance the model's non-linearity. The output layer has one node for outputting the corresponding load mass. The training process for the deep neural network is as follows:

[0047] Multiple weights of different masses were set up, and lifting tests were conducted using a device of the same model as the integrated dual-pump permanent magnet synchronous motor. Specifically, for each weight, multiple sets of working parameters were set, with different hydraulic flow rates or working pressures for each set of working parameters, while ensuring that the lifting could be completed. The gear pump was used to lift the weight according to each set of working parameters, and the median lifting acceleration during each round of lifting tests was calculated as the lifting acceleration under that round of lifting tests. The working parameters, weight mass, and lifting acceleration of each round of lifting tests were summarized to construct a sample set.

[0048] It should be noted that the selection of the weight should be determined based on the actual situation. For example, if the integrated dual-pump permanent magnet synchronous motor is used in a small electric forklift, weights of 100 kg, 200 kg, 300 kg, ..., 900 kg, and 1000 kg can be selected for lifting tests. Similarly, if the integrated dual-pump permanent magnet synchronous motor is used in a stacker electric forklift, weights of 200 kg, 400 kg, 600 kg, ..., 1800 kg, and 2000 kg can be selected for lifting tests. No specific restrictions are imposed here.

[0049] It should be noted that during the lifting test, the lifting acceleration can be collected at a preset sampling interval, which can be set to once every 0.1 seconds, once every 0.5 seconds, once every 1 second, etc. There is no limitation here. The lifting acceleration time series data during the lifting test is collected in this way. In order to avoid the influence of outliers or noise, the median is taken as the final lifting acceleration.

[0050] The sample set is divided into training, testing, and validation sets in a 70:15:15 ratio. Batch size and training cycle are set, with the batch size between 32-128 and the training cycle between 30-60. Hydraulic flow rate, working pressure, and lifting acceleration from the training set are used as inputs, and the corresponding load mass is used as the output label to train the deep neural network. Mean squared error is used as the loss function. During training, backpropagation is used to update the model parameters to minimize the loss function. Specifically, optimization algorithms such as Adam and SGD can be used to update the model parameters during the training process. The model's hyperparameters, such as learning rate, batch size, and number of hidden layers, are adjusted using a validation set to optimize model performance. The initial learning rate can be set between 0.001 and 0.01. After a predetermined training period, a test set is input into the deep neural network for performance testing. If the relative error is less than a preset threshold, the training is considered complete. Otherwise, the model's hyperparameters are adjusted and retraining is performed. The preset threshold can be set between 3% and 8% to ensure that the trained model has good simulation fitting ability. The specific value can be set by the staff according to the actual situation and is not restricted here.

[0051] S2, during the lifting operation, collect the real-time operating parameters of the gear pump and the real-time lifting acceleration, and combine the mapping relationship to determine the real-time predicted value of the load mass, and combine the historical value of the load mass to determine the real-time value of the load mass;

[0052] It should be noted that the real-time operating parameters of the gear pump include the real-time hydraulic flow rate and the real-time working pressure. The real-time hydraulic flow rate can be measured by a flow meter installed on the gear pump. For a gear pump with a fixed displacement, the real-time hydraulic flow rate can also be determined by the displacement and speed of the gear pump. The real-time hydraulic flow rate is the product of the displacement and speed. The displacement can be obtained by referring to the gear pump manual. Since the gear pump is installed on the motor shaft, the speed of the gear pump is the same as the speed of the motor shaft, which can be obtained by a speed sensor or related software monitoring system. The real-time working pressure can be obtained by a pressure sensor installed on the gear pump. The real-time lifting acceleration can be obtained by an acceleration sensor installed on the lifting arm. The sampling frequency of the three is consistent to ensure timestamp alignment. The sampling frequency is generally set between 100Hz and 500Hz. If the sampling frequency is set to 100Hz, the sampling time interval is 0.01 seconds, that is, the three are sampled once every 0.01 seconds to facilitate subsequent real-time adjustment.

[0053] The method for obtaining the real-time predicted value of the weight is as follows: input the real-time working parameters and real-time lifting acceleration into the trained deep neural network to obtain the real-time predicted value of the weight. Since the real-time lifting acceleration is easily affected by noise during the acquisition process, there may be a large deviation between the real-time predicted value of the weight and the actual value of the weight. It is necessary to combine the historical value of the weight to determine the accurate real-time value of the weight.

[0054] The logic for obtaining the real-time mass value of the load during the lifting operation is as follows: A preset time length threshold is used as the reference point. The time interval from the start of the lifting operation to the reference point is calculated. If this time interval is less than the time length threshold, the time interval from the start of the lifting operation to the reference point is used as the reference interval. Conversely, if the time interval is greater than the time length threshold, the reference interval is used as the endpoint, and a time interval equal to the time length threshold is traced back to the reference interval. The median of the historical mass value of the load at each sampling point within the reference interval is extracted, and the variation of the historical mass value of the load at each sampling point within the reference interval is calculated. The coefficients are determined by the relative error and coefficient of variation of the median historical weight of the object and the real-time predicted weight. The product of the median historical weight and the fusion weight is used as the first weighting value. The difference between the value 1 and the fusion weight is calculated, and the product of this difference and the real-time predicted weight is used as the second weighting value. The sum of the first weighting value and the second weighting value is used as the real-time weight. In this way, by weighting and fusing the median historical weight and the real-time predicted weight, the accuracy of the real-time weight calculation is ensured by comprehensively considering real-time and historical data.

[0055] It should be noted that when calculating the relative error between the median historical weight of an object and the real-time predicted weight, the difference between the median historical weight and the real-time predicted weight is calculated first. Then, the ratio of this difference to the median historical weight is used as the relative error to measure the degree of difference between the median historical weight and the real-time predicted weight. The calculation logic for historical weight and real-time weight is the same. Specifically, for any historical weight, its corresponding historical time is used as the reference point, and the same method as for the real-time weight is used to calculate the historical weight. In addition, if the real-time predicted weight is the first value collected during the lifting process and there is no historical data for reference, it is directly used as the real-time weight.

[0056] It should be noted that if the time interval from the start of the lifting operation to the current time is less than the time length threshold, it means that the lifting operation has been carried out for a short time and not enough historical values ​​of the load mass can be obtained. In other words, there are few historical values ​​of the load mass available for reference in terms of the real-time prediction of the load mass. Therefore, the time interval from the start of the lifting operation to the current time is used as the reference interval, and all calculated historical values ​​of the load mass are included in the reference range. On the other hand, if the time interval from the start of the lifting operation to the current time is not less than the time length threshold, it means that the lifting operation has been carried out for a long time and enough historical values ​​of the load mass have been obtained. In other words, there are many historical values ​​of the load mass available for reference in terms of the real-time prediction of the load mass. Therefore, the current time is used as the endpoint, and a time interval of the same length as the time length threshold is traced back as the reference interval to select historical values ​​of the load mass close to the current time as the reference. Compared with using the time interval from the start of the lifting operation to the current time as the reference interval, this greatly reduces the amount of data processing. The specific time length threshold is set by the staff according to the actual situation, such as 1 second, 2 seconds, 3 seconds, etc., and there is no restriction here.

[0057] The logic for determining the fusion weight is as follows: after performing maximum-minimum normalization on the historical values ​​of the weight at each sampling point within the reference interval, the coefficient of variation is calculated. The coefficient of variation is the ratio of the mean to the standard deviation. The larger the value, the greater the volatility of the historical values ​​of the weight within the reference interval, which means the lower the reliability of the historical values ​​of the weight within the reference interval. Conversely, the smaller the value, the smaller the volatility of the historical values ​​of the weight within the reference interval, which means the higher the reliability of the historical values ​​of the weight within the reference interval.

[0058] By combining the coefficient of variation, a preset high threshold for the coefficient of variation, and a low threshold for the coefficient of variation, the fluctuation state of the reference interval is determined, which is a low fluctuation state, a medium fluctuation state, or a high fluctuation state. The high threshold for the coefficient of variation is greater than the low threshold for the coefficient of variation. The logic for determining the fluctuation state is as follows: if the coefficient of variation is not greater than the low threshold for the coefficient of variation, it means that the historical value of the heavy object mass in the reference interval has low volatility, so the reference interval is defined as a low fluctuation state. If the coefficient of variation is not less than the high threshold for the coefficient of variation, it means that the historical value of the heavy object mass in the reference interval has high volatility, so the reference interval is defined as a high fluctuation state. If the coefficient of variation is between the high threshold for the coefficient of variation and the low threshold for the coefficient of variation, it means that the historical value of the heavy object mass in the reference interval has moderate volatility, so the reference interval is defined as a medium fluctuation state.

[0059] As one implementation method, the low threshold value of the coefficient of variation is between 5% and 10%, and the high threshold value of the coefficient of variation is between 25% and 30%. In this way, the reference interval is determined as a low fluctuation state, a medium fluctuation state, or a high fluctuation state. The specific values ​​are set by the staff according to the actual situation and are not limited here.

[0060] By combining the relative error, a preset high threshold for relative error, and a low threshold for relative error, the degree of balance between the real-time predicted value and the median historical value of the heavy object's mass is determined, categorized as low, medium, or high balance. The high threshold for relative error must be greater than the low threshold. If the relative error is not greater than the low threshold, it indicates a small deviation between the median historical value and the real-time predicted value, meaning they are relatively close. In this case, the balance is considered high, and the degree of balance is set to high. If the relative error is not greater than the low threshold, it indicates a small deviation between the median historical value and the real-time predicted value, meaning they are relatively balanced. If the relative error is less than the high threshold, it indicates a large deviation between the historical median value of the heavy object's mass and the real-time predicted value. This means there is a significant discrepancy between the two values, indicating a serious conflict. In this case, the balance level is set to low. If the relative error is between the high and low thresholds, it indicates a moderate deviation between the historical median value of the heavy object's mass and the real-time predicted value. This means the balance level is moderate, and the balance level is set to medium.

[0061] As one implementation method, the low threshold for relative error is between 5% and 10%, and the high threshold for relative error is between 15% and 20%, thereby determining the degree of balance as low, medium, or high. The specific values ​​are set by the staff according to the actual situation and are not limited here.

[0062] The fusion weight base term is set to 0.5 to initialize the contribution ratio of the median of the historical weight value and the real-time predicted weight value to the calculation of the real-time weight value to be the same. A base term correction value between 0 and 0.5 is set. When the fluctuation state is low, it indicates that the historical weight value of the object is relatively stable within the reference interval. This indicates that the higher the confidence level of the median of the historical weight value being close to the true weight value, the higher the confidence level. The sum of the fusion weight base term and the base term correction value is calculated as the fusion weight correction term to increase the contribution ratio of the median of the historical weight value to the calculation of the real-time weight value.

[0063] When the fluctuation state is medium fluctuation state, it indicates that the stability of the historical values ​​of the weight of the object within the reference interval is moderate. This also indicates that the confidence level of the median of the historical values ​​of the weight of the object being close to the true value of the weight of the object is moderate. In other words, the confidence level is moderate. Therefore, the basic term of the fusion weight is used as the fusion weight correction term and is not adjusted.

[0064] When the fluctuation state is high, it indicates that the stability of the historical values ​​of the weight within the reference interval is poor. This means that the lower the confidence level of the median of the historical weight values ​​being close to the true weight values, the lower the confidence level. Therefore, the difference between the basic term and the correction value of the basic term of the fusion weight is calculated as the fusion weight correction term to reduce the contribution of the median of the historical weight values ​​to the calculation of the real-time weight values.

[0065] As an implementation method, the value of the base term correction is preferably between 0.2 and 0.3 to avoid over-adjustment or under-adjustment. For example, the specific value of the base term correction can be 0.25, so that when the fluctuation state is low fluctuation state, medium fluctuation state, and high fluctuation state, the fusion weight correction term is 0.75, 0.5, and 0.25 respectively.

[0066] A secondary correction value, less than 0.5, is preset to be the difference between the base item correction value and the base item correction value. This secondary correction is then applied to the fusion weight correction item based on the secondary correction value, fluctuation status, and balance degree to determine the fusion weight. The logic is as follows:

[0067] If the fluctuation state is low, it indicates that the historical values ​​of the weight within the reference interval are relatively stable, which in turn indicates a high degree of confidence that the median of the historical weight is close to the actual weight, i.e., high reliability. On this basis, if the equilibrium degree is low, it indicates that the median of the historical weight differs significantly from the real-time predicted weight, which in turn indicates a low degree of confidence that the real-time predicted weight is close to the actual weight, i.e., low reliability. In this case, the sum of the fusion weight correction term and the secondary correction value is calculated as the fusion weight, thereby significantly increasing the weight of the median of the historical weight in the calculation of the real-time weight, ensuring the accuracy of the real-time weight.

[0068] If the balance level is moderate, it means that the difference between the median historical weight of the object and the real-time predicted weight is moderate. On the basis of a high confidence level that the median historical weight is close to the actual weight, the moderate difference between the median historical weight and the real-time predicted weight indicates that the confidence level that the real-time predicted weight is close to the actual weight is moderate. In other words, the confidence level is moderate. Therefore, the sum of the fusion weight correction term and 0.5 times the second correction value is calculated as the fusion weight. This slightly increases the weight of the median historical weight in the calculation of the real-time weight, ensuring the accuracy of the real-time weight while also taking into account the consideration of the real-time predicted weight.

[0069] If the balance is high, it means that the difference between the median historical weight of the object and the real-time predicted weight is small. On the basis of a high confidence level that the median historical weight of the object is close to the actual weight, the small difference between the median historical weight of the object and the real-time predicted weight indicates that the confidence level that the real-time predicted weight is close to the actual weight is also high, that is, the reliability is also high. The fusion weight correction term under the medium fluctuation state is used as the fusion weight to give the median historical weight of the object and the real-time predicted weight a balanced contribution ratio in the calculation of the real-time weight. While ensuring the accuracy of the real-time weight, the consideration of the real-time predicted weight is also taken into account.

[0070] If the fluctuation state is not a low fluctuation state, it indicates that the confidence level of the median of the historical value of the heavy object mass is moderate or low, which means that the reference value of the median of the historical value of the heavy object mass is not strong. The fusion weight correction term under the high fluctuation state is used as the fusion weight to give the real-time predicted value of the heavy object mass a higher contribution ratio in the calculation process of the real-time value of the heavy object mass, focusing on the real-time predicted value of the heavy object mass and reducing the impact of adverse noise factors in the reference interval.

[0071] It should be noted that setting the difference between the secondary correction value and the basic correction value to be less than 0.5 avoids the problem of setting the contribution ratio of the real-time predicted weight of the heavy object to 0 or a negative value when the fusion weight is not less than 1. This ensures that the real-time predicted weight of the heavy object is taken into consideration. The secondary correction value is preferably half of the difference between 0.5 and the basic correction value to ensure that the real-time predicted weight of the heavy object has a certain contribution ratio in the calculation of the real-time weight of the heavy object. For example, when the basic correction value is 0.25, the secondary correction value is set to 0.125 to ensure that the real-time predicted weight of the heavy object is taken into consideration.

[0072] As one implementation method, a lifting test was conducted using a 500 kg weight. The real-time weight values ​​calculated based on this technical solution are shown in Table 1 below. Table 1 clearly shows that as the lifting test progresses, the real-time weight value calculated based on this technical solution gradually approaches the true weight value, stabilizing after a lifting time exceeding two seconds. Once stable, the relative error is less than 1%, achieving effective and accurate monitoring of the weight's mass. Further details can be found below. Figure 2 As shown below Figure 2 The blue line graph represents the real-time weight of the object, while the red line represents the actual weight of the object. Figure 2 It can also be clearly seen that as the lifting test proceeds, the real-time value of the weight gradually approaches the true value of the weight.

[0073] Table 1. Comparison of Real-time and Actual Mass Values ​​of Heavy Objects

[0074]

[0075] S3. Based on the real-time value of the weight, determine the real-time target values ​​of the working parameters and the real-time target value of the power of the gear pump. Based on the real-time value of the weight and the real-time value of the vehicle speed, determine the real-time target values ​​of the working parameters and the real-time target value of the power of the vane pump. Combine the two real-time target values ​​of power to determine the real-time power distribution ratio of the gear pump and the vane pump.

[0076] The logic for determining the real-time target values ​​of the working parameters is as follows: Multiple load test points with different weights are set. Based on the lifting test, suitable working parameters of the gear pump are determined when lifting the load at each load test point. These parameters include suitable hydraulic flow rate and suitable working pressure. The suitable hydraulic flow rate and suitable working pressure are used as the target values ​​of hydraulic flow rate and working pressure under the corresponding load weight. The load weight is used as the independent variable, and the target values ​​of hydraulic flow rate and working pressure of the gear pump are used as the dependent variables. Based on polynomial fitting, mathematical expressions for the load weight and gear pump hydraulic flow rate target values, as well as mathematical expressions for the load weight and gear pump working pressure target values, are fitted. The real-time value of the load weight is substituted into the above two mathematical expressions to determine the real-time target values ​​of hydraulic flow rate and working pressure of the gear pump. Polynomial fitting is an existing technology and will not be elaborated here. Furthermore, during the polynomial fitting process, the gradient descent method is used to iteratively optimize the polynomial coefficients to ensure that the fitting accuracy meets the requirements, such as a relative error of less than 5%. It is recommended that the total degree of the polynomial be between 3 and 6 to avoid overfitting while ensuring accurate fitting. The gradient descent method is used for iterative optimization.

[0077] It should be noted that the lifting test was conducted using a device of the same model as the integrated dual-pump permanent magnet synchronous motor. Specifically, for each heavy object, multiple sets of working parameters were set, each with different hydraulic flow or working pressure, while ensuring that the lifting could be completed. The gear pump was then used to conduct lifting tests on the heavy object according to each set of working parameters. This was done to obtain the energy efficiency of the gear pump in lifting the same heavy object under different working parameters. The energy efficiency of the gear pump was used as the evaluation index, which is the ratio of the output power to the input power of the gear pump. For the same heavy object, the working parameters corresponding to the lifting test with the highest energy efficiency of the gear pump were extracted as the appropriate working parameters for that heavy object mass. Alternatively, the target values ​​of hydraulic flow and working pressure for the corresponding heavy object mass could be determined based on expert analysis and existing common knowledge. This is existing technology and will not be elaborated here.

[0078] Multiple load test points with different weights and vehicle speed test points are set and combined to form multiple load steering test combinations. Each load steering test combination corresponds to one load test point and one vehicle speed test point. Based on the load steering test, the appropriate operating parameters of the vane pump under each load steering test combination are determined, including the appropriate hydraulic flow rate and the appropriate working pressure. The appropriate hydraulic flow rate and the appropriate working pressure are used as the target values ​​of hydraulic flow rate and working pressure under the corresponding load steering test combination. The weight and vehicle speed are used as independent variables, and the target values ​​of hydraulic flow rate and working pressure of the vane pump are used as dependent variables. Based on polynomial fitting, mathematical expressions for the target values ​​of hydraulic flow rate and hydraulic pressure of the vane pump, as well as mathematical expressions for the target values ​​of working pressure of the vane pump, are obtained. The real-time values ​​of weight and vehicle speed are substituted into the above two mathematical expressions to determine the real-time target values ​​of hydraulic flow rate and working pressure of the vane pump.

[0079] It should be noted that the same device as the integrated dual-pump permanent magnet synchronous motor is used for the load steering test. Specifically, when conducting the load steering test for each load steering test combination, the steering angle is set to the maximum value under normal use to ensure that the final suitable working parameters can meet the steering requirements of most people. With stable steering as the constraint, multiple sets of working parameters are set, and the hydraulic flow or working pressure of each set of working parameters is different. The vane pump is used to conduct load steering tests on the heavy object according to each set of working parameters. The energy efficiency of the vane pump is used as the evaluation index, which is the ratio of the output power to the input power of the vane pump. For the same load steering test combination, the working parameters corresponding to the load steering test with the highest energy efficiency of the vane pump are extracted as the suitable working parameters under that load steering test combination. In addition, the target values ​​of hydraulic flow and working pressure under the corresponding load steering test combination can also be determined based on expert analysis and existing common knowledge. This is existing technology and will not be elaborated here.

[0080] It should be noted that when determining whether stable steering is met, it can be done by analyzing the vehicle's slip angle and steering response time during steering. For example, a slip angle of no more than 5 degrees and a steering response time of no more than 1 second are considered necessary conditions for stable steering.

[0081] Among them, the real-time target power value of the gear pump is the product of its real-time target hydraulic flow rate and its real-time target working pressure value, and the real-time target power value of the vane pump is the product of its real-time target hydraulic flow rate and its real-time target working pressure value. This is common knowledge and will not be elaborated here.

[0082] The calculation logic for the real-time power distribution ratio of gear pumps and vane pumps is as follows: calculate the sum of the real-time power distribution ratios of gear pumps and vane pumps, calculate the ratio of the real-time power distribution ratio of gear pumps to the sum, and use this ratio as the real-time power distribution ratio of gear pumps; and calculate the ratio of the real-time power distribution ratio of vane pumps to the sum, and use this ratio as the real-time power distribution ratio of vane pumps.

[0083] S4, based on the PID control algorithm, analyzes the real-time target values ​​of the working parameters of the gear pump and the vane pump, and combines the real-time power distribution ratio to adjust the motor speed in real time.

[0084] The logic for real-time adjustment of the motor speed is as follows:

[0085] 1) Calculate the difference between the real-time target value of the gear pump hydraulic flow and the real-time value of the gear pump hydraulic flow, and take the ratio of this difference to the real-time target value of the gear pump hydraulic flow as the real-time error of the gear pump hydraulic flow. Calculate the difference between the real-time target value of the gear pump working pressure and the real-time value of the gear pump working pressure, and take the ratio of this difference to the real-time target value of the gear pump working pressure as the real-time error of the gear pump working pressure. Sum the real-time error of the gear pump hydraulic flow and the real-time error of the gear pump working pressure as the real-time error of the gear pump.

[0086] It should be noted that when calculating the real-time error of the gear pump hydraulic flow, the difference between the real-time target value and the real-time hydraulic flow of the gear pump is first calculated. This difference is used to measure the degree of deviation of the gear pump hydraulic flow from its target value. Then, it is divided by the real-time target value of the gear pump hydraulic flow to perform dimensionless processing on the above difference. This facilitates the fusion with the dimensionless real-time error of the gear pump working pressure, which is also described later. This avoids the problem of overemphasizing higher-order-of-magnitude data and ignoring lower-order-of-magnitude data when merging data of different orders of magnitude. The calculation logic for the real-time error of the gear pump working pressure is the same. Finally, the dimensionless real-time error of the gear pump hydraulic flow and the real-time error of the gear pump working pressure are summed to obtain the real-time error of the gear pump, which characterizes the comprehensive deviation of the gear pump working parameters. The larger the value, the further the current state of the gear pump is from the target state, and the more necessary it is to adjust the motor speed to make the working parameters approach the target value.

[0087] 2) Calculate the difference between the real-time target value of the hydraulic flow rate of the vane pump and the real-time value of the hydraulic flow rate of the vane pump, and take the ratio of this difference to the real-time target value of the hydraulic flow rate of the vane pump as the real-time error of the hydraulic flow rate of the vane pump. Calculate the difference between the real-time target value of the working pressure of the vane pump and the real-time value of the working pressure of the vane pump, and take the ratio of this difference to the real-time target value of the working pressure of the vane pump as the real-time error of the working pressure of the vane pump. The sum of the real-time error of the hydraulic flow rate of the vane pump and the real-time error of the working pressure of the vane pump is taken as the real-time error of the vane pump. The calculation logic is the same as that of the gear pump, and will not be elaborated here.

[0088] 3) Calculate the product of the real-time power distribution ratio of the gear pump and the real-time error of the gear pump, and calculate the product of the real-time power distribution ratio of the vane pump and the real-time error of the vane pump. Sum the two products as the comprehensive real-time error.

[0089] It should be noted that in steps 1 and 2) above, a dimensionless calculation method was used to calculate the real-time error of the gear pump and the vane pump. This makes the calculated real-time error a relative value. The pump with a higher real-time power target value has a larger hydraulic flow and working pressure than the pump with a lower real-time power target value. Even if the real-time error is the same, the pump with a higher real-time power target value has a larger actual error value than the pump with a lower real-time power target value. Therefore, it needs to be given a higher contribution ratio. Here, the real-time power distribution ratio of each of the two pumps is used to weight the corresponding real-time error so that the comprehensive real-time error is more instructive.

[0090] 4) Using the comprehensive real-time error as the input variable, the PID control algorithm outputs a real-time speed adjustment signal to optimize the motor speed in real time. The specific mathematical expression is as follows:

[0091]

[0092] In the formula, This is a real-time speed adjustment signal, used to adjust and change the motor speed. To account for real-time errors, , , These are the proportional gain coefficient, integral gain coefficient, and differential gain coefficient, respectively. The lifting operation is about to begin. To advance the work at this moment, The comprehensive error at time t is calculated using the same logic as the comprehensive real-time error, where t is the time variable from the start of the lifting operation to the current moment of the lifting operation. The calculation logic for the real-time differential term of the comprehensive error is as follows: calculate the difference between the comprehensive real-time error and the previous comprehensive error, and calculate the ratio of this difference to the sampling time interval, which is used as the real-time differential term of the comprehensive error. The sampling time interval is specifically determined based on the sampling frequency of the hydraulic flow. For example, if the sampling frequency is set to 100Hz, the sampling time interval is 0.01 seconds, that is, sampling is performed every 0.01 seconds. Calculate the difference between the comprehensive real-time error under the current sampling round and the comprehensive error under the previous sampling round, and then divide the difference by 0.01 seconds to obtain the real-time differential term of the comprehensive error. The motor speed is adjusted in real time through the speed adjustment signal output by the above PID control algorithm to iteratively reduce the comprehensive error, ensuring that both the gear pump and the vane pump approach their real-time target values ​​of working parameters, thereby ensuring the smooth progress of the lifting operation and providing sufficient steering assistance for the steering operation that is being performed or will be performed, so as to facilitate steering operation at any time.

[0093] As an implementation method, the proportional gain coefficient directly affects the system's response speed, and its specific value is generally set between 0.5 and 1. The integral gain coefficient is used to eliminate steady-state error, and its specific value is generally set between 0.01 and 1, which is smaller than the proportional gain coefficient. The derivative gain coefficient is used to predict the future behavior of the system, thereby reducing overshoot and oscillation, and its specific value is generally set between 0 and 1, which is smaller than the proportional gain coefficient. In addition, during the process of adjusting the motor speed based on the PID control algorithm, the proportional gain coefficient, integral gain coefficient, and derivative gain coefficient are adjusted in real time based on the Ziegler-Nichols method in the existing technology to ensure the smooth progress of motor speed adjustment optimization.

[0094] Example 2:

[0095] This invention provides a permanent magnet synchronous motor, using the integrated dual-pump permanent magnet synchronous motor power distribution method in Embodiment 1 above, including:

[0096] The main body of the permanent magnet synchronous motor includes the stator assembly, the rotor assembly, and the motor shaft;

[0097] A vane pump, whose pump shaft is directly connected to one end of the motor shaft via a first coupling, is used to provide steering assistance to the vehicle's power steering system.

[0098] A gear pump, whose pump shaft is directly connected to the other end of the motor shaft via a spline, is used to provide lifting force for the vehicle's lifting system.

[0099] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0100] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0101] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0102] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A power distribution method for an integrated dual-pump permanent magnet synchronous motor, used to adjust the speed of the integrated dual-pump permanent magnet synchronous motor to achieve power distribution between the gear pump and the vane pump, characterized in that, Includes the following steps: S1, obtain the lifting acceleration of the weight when the gear pump provides lifting force for lifting the same weight under different operating parameters, so as to establish the mapping relationship between operating parameters, lifting acceleration and weight mass. Operating parameters include hydraulic flow and operating pressure. S2, during the lifting operation, collect the real-time operating parameters of the gear pump and the real-time lifting acceleration, and combine the mapping relationship to determine the real-time predicted value of the load mass, and combine the historical value of the load mass to determine the real-time value of the load mass; S3. Based on the real-time value of the weight, determine the real-time target values ​​of the working parameters and the real-time target value of the power of the gear pump. Based on the real-time value of the weight and the real-time value of the vehicle speed, determine the real-time target values ​​of the working parameters and the real-time target value of the power of the vane pump. Combine the two real-time target values ​​of power to determine the real-time power distribution ratio of the gear pump and the vane pump. S4, based on the PID control algorithm, analyzes the real-time target values ​​of the working parameters of the gear pump and the vane pump, and combines the real-time power distribution ratio to adjust the motor speed in real time. The logic for obtaining the real-time value of the load mass during the lifting operation is as follows: A preset time length threshold is used as the reference point. The time interval from the start of the lifting operation to the reference point is calculated. If this time interval is less than the time length threshold, the time interval from the start of the lifting operation to the reference point is used as the reference interval. Conversely, if the time interval is greater than the time length threshold, the reference point is used as the endpoint, and a time interval equal to the time length threshold is traced back to the reference interval. The median of the historical load mass values ​​at each sampling point within the reference interval is extracted, and the coefficient of variation of the historical load mass values ​​at each sampling point within the reference interval is calculated. Based on the relative error and coefficient of variation between the median of the historical load mass values ​​and the real-time predicted load mass values, a fusion weight is determined. The product of the median of the historical load mass values ​​and the fusion weight is used as the first weighting value. The difference between the value 1 and the fusion weight is calculated, and the product of this difference and the real-time predicted load mass values ​​is used as the second weighting value. The sum of the first and second weighting values ​​is used as the real-time load mass value.

2. The power distribution method for an integrated dual-pump permanent magnet synchronous motor according to claim 1, characterized in that, The system constructs a mapping relationship between working parameters, lifting acceleration, and the mass of the heavy object based on a deep neural network. It includes an input layer for receiving hydraulic flow, working pressure, and lifting acceleration, a hidden layer for extracting features from hydraulic flow, working pressure, and lifting acceleration, and an output layer for outputting the mass of the heavy object.

3. The power distribution method for an integrated dual-pump permanent magnet synchronous motor according to claim 1, characterized in that, The logic for determining the fusion weights is as follows: After performing maximum-minimum normalization on the historical values ​​of the weight of the heavy object at each sampling point within the reference interval, the coefficient of variation is calculated. Combining the coefficient of variation with the preset high and low thresholds of the coefficient of variation, the fluctuation state of the reference interval is determined, which is a low fluctuation state, a medium fluctuation state, or a high fluctuation state, and the high threshold of the coefficient of variation is greater than the low threshold of the coefficient of variation. By combining the relative error between the median of the historical weight and the real-time predicted weight, and the preset high and low relative error thresholds, the degree of balance between the real-time predicted weight and the median of the historical weight is determined, which is low, medium, or high balance, and the high relative error threshold is greater than the low relative error threshold. Set the base weight of the fusion weight to 0.5, and set a base weight correction value between 0 and 0.

5. When the fluctuation state is low, calculate the sum of the base weight of the fusion weight and the base weight correction value as the fusion weight correction value. When the fluctuation state is medium, use the base weight of the fusion weight as the fusion weight correction value. When the fluctuation state is high, calculate the difference between the base weight of the fusion weight and the base weight correction value as the fusion weight correction value. A secondary correction value, less than 0.5, is preset to be the difference between the base item correction value and the base item correction value. The fusion weight correction item is then adjusted in a secondary manner based on the secondary correction value, the fluctuation state, and the degree of equilibrium, so as to determine the fusion weight.

4. The integrated dual-pump permanent magnet synchronous motor power distribution method according to claim 3, characterized in that, The logic for determining the fluctuation state is as follows: if the coefficient of variation is not greater than the low threshold of the coefficient of variation, the reference interval is defined as a low fluctuation state; if the coefficient of variation is not less than the high threshold of the coefficient of variation, the reference interval is defined as a high fluctuation state; if the coefficient of variation is between the high threshold of the coefficient of variation and the low threshold of the coefficient of variation, the reference interval is defined as a medium fluctuation state. The logic for determining the balance level is as follows: if the relative error is not greater than the low threshold of relative error, the balance level is set to high balance level; if the relative error is not less than the high threshold of relative error, the balance level is set to low balance level; if the relative error is between the high threshold of relative error and the low threshold of relative error, the balance level is set to medium balance level.

5. The power distribution method for an integrated dual-pump permanent magnet synchronous motor according to claim 3, characterized in that, The logic for making a second adjustment to the fusion weight correction term to determine the fusion weight is as follows: If the volatility state is low volatility, then if the equilibrium level is low equilibrium, the sum of the fusion weight correction term and the second correction value is calculated as the fusion weight. If the equilibrium level is medium equilibrium, the sum of the fusion weight correction term and 0.5 times the second correction value is calculated as the fusion weight. If the equilibrium level is high equilibrium, the fusion weight correction term under medium volatility is used as the fusion weight. If the volatility state is not a low volatility state, then the fusion weight correction term under the high volatility state will be used as the fusion weight.

6. The power distribution method for an integrated dual-pump permanent magnet synchronous motor according to claim 1, characterized in that, The logic for determining the real-time target values ​​of the working parameters is as follows: taking the mass of the object as the independent variable, and the target values ​​of the hydraulic flow and working pressure of the gear pump as the dependent variables, the mathematical expressions for the mass of the object and the target values ​​of the hydraulic flow and the working pressure of the gear pump are fitted based on polynomial fitting. The real-time value of the mass of the object is then substituted into the above two mathematical expressions to determine the real-time target values ​​of the hydraulic flow and the working pressure of the gear pump. Using the mass of the load and the vehicle speed as independent variables, and the target values ​​of the hydraulic flow rate and working pressure of the vane pump as dependent variables, the mathematical expressions for the mass of the load, the vehicle speed, and the target values ​​of the hydraulic flow rate of the vane pump, as well as the mathematical expressions for the mass of the load, the vehicle speed, and the target values ​​of the working pressure of the vane pump, are fitted using a polynomial. The real-time values ​​of the mass of the load and the real-time values ​​of the vehicle speed are then substituted into the above two mathematical expressions to determine the real-time target values ​​of the hydraulic flow rate and the real-time target values ​​of the working pressure of the vane pump. The real-time target power value of the gear pump is the product of its real-time target hydraulic flow rate and its real-time target working pressure value, and the real-time target power value of the vane pump is the product of its real-time target hydraulic flow rate and its real-time target working pressure value. The calculation logic for the real-time power distribution ratio of gear pumps and vane pumps is as follows: calculate the sum of the real-time power distribution ratios of gear pumps and vane pumps, calculate the ratio of the real-time power distribution ratio of gear pumps to the sum, and use this ratio as the real-time power distribution ratio of gear pumps.

7. The power distribution method for an integrated dual-pump permanent magnet synchronous motor according to claim 1, characterized in that, The logic for real-time adjustment of the motor speed is as follows: 1) Calculate the difference between the real-time target value of the gear pump hydraulic flow and the real-time value of the gear pump hydraulic flow, and take the ratio of this difference to the real-time target value of the gear pump hydraulic flow as the real-time error of the gear pump hydraulic flow. Calculate the difference between the real-time target value of the gear pump working pressure and the real-time value of the gear pump working pressure, and take the ratio of this difference to the real-time target value of the gear pump working pressure as the real-time error of the gear pump working pressure. Sum the real-time error of the gear pump hydraulic flow and the real-time error of the gear pump working pressure as the real-time error of the gear pump. 2) Calculate the difference between the real-time target value of the hydraulic flow rate of the vane pump and the real-time value of the hydraulic flow rate of the vane pump, and take the ratio of this difference to the real-time target value of the hydraulic flow rate of the vane pump as the real-time error of the hydraulic flow rate of the vane pump. Calculate the difference between the real-time target value of the working pressure of the vane pump and the real-time value of the working pressure of the vane pump, and take the ratio of this difference to the real-time target value of the working pressure of the vane pump as the real-time error of the working pressure of the vane pump. Sum the real-time error of the hydraulic flow rate of the vane pump and the real-time error of the working pressure of the vane pump as the real-time error of the vane pump. 3) Calculate the product of the real-time power distribution ratio of the gear pump and the real-time error of the gear pump, and calculate the product of the real-time power distribution ratio of the vane pump and the real-time error of the vane pump. Sum the two products as the comprehensive real-time error. 4) Using the comprehensive real-time error as the input variable, the PID control algorithm outputs a real-time speed adjustment signal to optimize the motor speed in real time.

8. A permanent magnet synchronous motor, using the integrated dual-pump permanent magnet synchronous motor power distribution method according to any one of claims 1-7, characterized in that, include: The main body of the permanent magnet synchronous motor includes the stator assembly, the rotor assembly, and the motor shaft; A vane pump, whose pump shaft is directly connected to one end of the motor shaft via a first coupling, is used to provide steering assistance to the vehicle's power steering system. A gear pump, whose pump shaft is directly connected to the other end of the motor shaft via a spline, is used to provide lifting force for the vehicle's lifting system.

Images