A Multi-condition High-Efficiency Cooperative Control System and Method for Electronic Oil Pumps in New Energy Vehicles

Abstract

This invention specifically relates to a multi-condition high-efficiency collaborative control system and method for electronic oil pumps in new energy vehicles, belonging to the field of electronic oil pump control technology for new energy vehicles. The method includes: inputting the fused state vector into a condition identification and prediction processing flow to identify the current condition mode type, and predicting the lubrication and cooling flow requirements at each moment within a preset prediction time window based on the current condition mode type, generating a target flow requirement envelope; the target flow requirement envelope is a time-series curve describing the upper and lower limits of the required flow rate at each moment within the prediction time window. This invention employs a collaborative control method involving multi-source heterogeneous signal fusion, condition identification and flow requirement feedforward prediction, multi-objective constraint optimization, multi-pump active decoupling compensation, and degradation adaptive compensation. This solves the technical problems of existing single feedback control strategies, such as delayed oil supply response under transient conditions, pressure oscillations in multi-pump parallel systems, and decreased control accuracy after long-term use.

Description

Technical Field

[0001] This invention relates to the field of electronic oil pump control technology for new energy vehicles, and in particular to a multi-condition high-efficiency collaborative control system and method for electronic oil pumps in new energy vehicles. Background Technology

[0002] The electric drive system of new energy vehicles has diverse flow and pressure requirements for lubrication and cooling. As the core hydraulic power source, the electronic oil pump needs to provide matching oil supply under different driving conditions.

[0003] Existing electronic oil pump control schemes mostly adopt a single feedback control strategy based on measured oil temperature or oil pressure, rely on sensor feedback data at the current moment for adjustment, and execute control commands independently for each oil pump, with control parameters remaining fixed after leaving the factory.

[0004] The above-mentioned single feedback control strategy has the following technical problems:

[0005] During transient operating condition switching such as rapid acceleration, high-power fast charging, and low-temperature cold start, the control strategy cannot detect the upcoming changes in flow demand in advance, resulting in a lag in the adjustment of oil pump speed relative to the actual lubrication and cooling demand. During the lag period, there may be insufficient oil film thickness or insufficient cooling flow.

[0006] When the system uses multiple oil pumps in parallel, the existing independent control strategy does not take into account the hydraulic cross-coupling effect formed between the oil pumps through the common rail oil circuit, which leads to pressure oscillation in the multi-pump system and affects the stability of oil supply.

[0007] During long-term vehicle use, oil viscosity drift and oil pump mechanical wear cause deviations between the initially calibrated control parameters and the actual system characteristics, resulting in a gradual decrease in control accuracy. Summary of the Invention

[0008] The purpose of this invention is to provide a multi-condition high-efficiency collaborative control system and method for electronic oil pumps in new energy vehicles in order to solve the above-mentioned problems.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A multi-condition efficient collaborative control method for electronic oil pumps in new energy vehicles includes:

[0011] Multi-source heterogeneous signals from multiple domain controllers are collected, and adaptive weighted fusion processing is performed on these signals to obtain a fused state vector. This fused state vector includes filtered and confidence-weighted oil pressure estimates, temperature change trend values, and driving intention feature values. The fused state vector is then input into the operating condition identification and prediction processing flow to identify the current operating condition mode type. Based on this current operating condition mode type, the lubrication and cooling flow requirements at each moment within a preset prediction time window are predicted, generating a target flow requirement envelope. This target flow requirement envelope describes the... The time series curves of the upper and lower limits of the required flow rate at each moment within the prediction time window are described; using the oil pump speed and solenoid valve opening as decision variables, and the total system energy consumption and lubrication and cooling safety redundancy as optimization objectives, the target flow rate demand envelope is used as the flow rate constraint to solve a multi-objective optimization problem to obtain the target oil pump speed and target solenoid valve opening; the target oil pump speed is converted into a speed control command for the oil pump drive circuit, and the target solenoid valve opening is converted into an opening control command for the solenoid valve drive circuit, and the control commands are sent to the corresponding actuator drive circuits for execution.

[0012] Furthermore, the multi-source heterogeneous signals include three categories: the first category is driving intention signals, including the accelerator pedal depth change rate and regenerative braking intensity signals from the vehicle controller; the second category is thermal load signals, including cell temperature data from the battery management system and power device temperature estimates from the motor controller; the third category is actuator status signals, including the current oil pump speed, the current solenoid valve opening, and oil passage pressure data obtained after high-frequency sampling by the pressure sensor; the adaptive weighted fusion processing includes: performing fast Fourier transform processing on the oil passage pressure data to obtain the pressure pulsation spectrum; performing time alignment processing on the three types of signals with the local clock of the electronic oil pump control unit as the reference time axis, and using a linear interpolation algorithm to compensate for time deviations caused by communication delays or inconsistent sampling periods to obtain a synchronization signal sequence; and using a Kalman filter algorithm to perform fusion calculation based on the synchronization signal sequence to obtain the fused state vector.

[0013] Furthermore, the weight coefficients of each signal component in the Kalman filter algorithm are dynamically adjusted according to the vehicle's driving state. Specifically, the wheel speed disturbance value is calculated based on the wheel speed sensor signals from the vehicle's electronic stability system. When the wheel speed disturbance value exceeds a preset smoothness threshold, it is determined that the vehicle is driving on a bumpy road surface. The corresponding diagonal elements in the observation noise covariance matrix corresponding to the pressure sensor fluctuation component are increased, thus weakening the Kalman gain's response to the pressure sensor fluctuation component. At the same time, the corresponding diagonal elements in the observation noise covariance matrix corresponding to the virtual pressure estimate are decreased, thus enhancing the Kalman gain's response to the virtual pressure estimate. The virtual pressure estimate is calculated based on the current oil pump speed and the pre-stored mapping relationship between speed and pressure. The smoothness threshold is a fixed threshold calibrated after collecting wheel speed disturbance data under various road conditions.

[0014] Further, the identification of the current operating condition mode type includes: inputting the multi-dimensional feature data in the fused state vector into a fuzzy neural network, and outputting the membership degree distribution of the current operating condition mode type; the input layer of the fuzzy neural network receives each feature component in the fused state vector, the hidden layer maps each feature component to a membership value in the range of each fuzzy linguistic variable through a Gaussian membership function, and extracts fuzzy rule features after weighted combination of the fuzzy membership values ​​of each feature component; the output layer outputs the membership value corresponding to each operating condition mode type after normalization by the Softmax activation function; the mean and standard deviation parameters of the Gaussian membership function are updated through the Adam optimization algorithm during the network training phase; and the current operating condition mode is determined based on the operating condition mode type with the highest membership degree in the membership degree distribution.

[0015] Furthermore, the prediction of lubrication and cooling flow demand at each moment within a preset prediction time window includes: extracting the motor torque sequence from the fused state vector within a preset sampling period in the past, inputting the motor torque sequence into a short-term flow demand predictor based on a long short-term memory network, and combining it with the flow baseline parameters corresponding to the current operating mode to output the predicted lubricating oil flow demand at each moment within the preset prediction time window; the flow baseline parameters are the lubricating oil demand benchmark flow values ​​pre-calibrated in bench tests for each operating mode, and the flow baseline parameters and the motor torque sequence are concatenated together as the input feature vector of the short-term flow demand predictor; the short-term flow demand predictor uses a mean squared error loss function, uses historical motor torque sequences as input samples, and uses measured lubricating oil flow data within the corresponding time period as supervised labels for supervised training.

[0016] Further, generating the target flow demand envelope includes: calculating the upper and lower flow limits for each moment based on the predicted flow value at each time point and the redundancy coefficient corresponding to the current operating mode; the upper flow limit for each moment point is equal to the product of the redundancy coefficient plus 1 and the predicted flow value at that moment, and the lower flow limit for each moment point is equal to the product of 1 minus the redundancy coefficient and the predicted flow value at that moment; the redundancy coefficient is a value greater than zero and less than 1, and is retrieved from a pre-stored mapping table of operating modes and redundancy coefficients according to the current operating mode; when identified as a high-power fast charging pre-operating mode, the lower flow limit is set to be no less than the corresponding flow value of the auxiliary cooling oil pump at the standby cruise speed; when identified as a high-torque low-speed motor operating mode, the predicted flow value is replaced with the fixed flow value of the oil pump in the maximum flow open-loop control mode.

[0017] Furthermore, solving the multi-objective optimization problem includes: constructing an objective function, which is the cumulative value of weighted energy consumption and safety penalty terms at all times within the prediction time window. The objective function includes a weighted sum of three terms: the first term is the electrical power consumption of the oil pump at the corresponding speed and flow rate; the second term is the throttling heat loss power of the solenoid valve at the corresponding opening degree; and the third term is the safety penalty function. The safety penalty function is a constant greater than zero when the difference between the oil passage pressure and the minimum critical pressure required to maintain the oil film thickness is greater than zero and the oil temperature is within the normal temperature range. When the difference is less than or equal to zero or the oil temperature exceeds the preset upper limit of oil temperature, the value of the safety penalty function is an exponential function value with the weighted sum of the pressure deficiency and temperature exceedance after dimensionless processing as the exponent; the constraints include the oil pump speed being within the allowable speed range, the solenoid valve opening being within the allowable opening range, and the actual flow rate at each moment not being lower than the lower limit of flow rate; the objective function is iteratively solved under the constraints using a sequential quadratic programming algorithm, with the solution result of the previous control cycle used as the initial point to start the iteration in each control cycle.

[0018] Furthermore, it also includes feedforward compensation processing and decoupling processing; the feedforward compensation processing includes: based on the mapping relationship between the reverse torque of the oil pump and its speed and pressure, according to the current oil viscosity estimate and target pressure value, obtaining the feedforward speed increment through table lookup and linear interpolation algorithm, and superimposing the feedforward speed increment onto the target oil pump speed to obtain the compensated target oil pump speed; the decoupling processing includes: based on the hydraulic characteristic parameters of the main oil pump and the auxiliary oil pump, calculating the cross-influence coefficient of the main oil pump speed change on the actual displacement of the auxiliary oil pump using the relative gain matrix analysis method; when the target speed of the main oil pump changes, the speed compensation amount of the auxiliary oil pump is equal to the negative of the product of the cross-influence coefficient and the change in the target speed of the main oil pump; superimposing the speed compensation amount onto the target speed of the auxiliary oil pump to obtain the decoupled target speed of the auxiliary oil pump.

[0019] Furthermore, it also includes a degradation compensation step, which is triggered after the vehicle is in a low-load cruise steady-state condition and the steady-state duration exceeds a preset steady-state determination duration. This step includes: collecting the current drive current value and current speed value of the oil pump, and calculating the ratio of current to speed; comparing the ratio with the factory reference ratio to obtain a ratio offset; when the ratio offset is positive and the offset amplitude is within a first offset interval, it is determined that the viscosity has increased, and the corresponding feedforward torque compensation coefficient increment is obtained and updated in the inverse mapping relationship table used by the feedforward compensation processing; when the ratio offset is positive and the offset amplitude is within a second offset interval greater than the upper limit of the first offset interval, it is determined that the oil pump volumetric efficiency has decreased, and the corresponding speed and flow rate curve slope correction value is obtained and updated in the oil pump efficiency mapping relationship in the multi-objective optimization problem; when the ratio offset exceeds the upper limit of the second offset interval, a fault alarm is triggered; the current oil viscosity estimate is obtained from the pre-stored mapping relationship table based on the ratio offset calculated in this operation.

[0020] A multi-condition high-efficiency collaborative control system for electronic oil pumps in new energy vehicles includes:

[0021] The multi-source signal acquisition and fusion module is used to acquire multi-source heterogeneous signals from multiple domain controllers, perform adaptive weighted fusion processing on the multi-source heterogeneous signals, and obtain a fusion state vector.

[0022] The working condition identification and flow prediction module is used to input the fused state vector into the working condition identification and prediction process, identify the current working condition mode type, and predict the lubrication and cooling flow demand at each moment within a preset prediction time window based on the current working condition mode type, and generate the target flow demand envelope.

[0023] The multi-objective optimization solution module is used to solve the multi-objective optimization problem by taking the oil pump speed and solenoid valve opening as decision variables, the total system energy consumption and lubrication and cooling safety redundancy as optimization objectives, and the target flow demand envelope as flow constraint conditions, so as to obtain the target oil pump speed and target solenoid valve opening.

[0024] The control command generation and distribution module is used to convert the target oil pump speed into a speed control command for the oil pump drive circuit, convert the target solenoid valve opening degree into an opening control command for the solenoid valve drive circuit, and send the control commands to the corresponding actuator drive circuits for execution.

[0025] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0026] 1. This invention solves the technical problems of delayed oil supply response under transient conditions, pressure oscillation in multi-pump parallel systems, and decreased control accuracy after long-term use by the existing single feedback control strategy through a collaborative control method of multi-source heterogeneous signal fusion, operating condition identification and flow demand feedforward prediction, multi-objective constraint optimization solution, multi-pump active decoupling compensation, and degradation adaptive compensation.

[0027] 2. This invention reduces the possibility of false triggering caused by sensor reading fluctuations on bumpy roads by employing dynamic weighted Kalman filtering to fuse multi-source signals; it shortens the lag time of oil supply response under transient conditions by predicting flow demand and generating control commands in advance based on long short-term memory networks; it achieves comprehensive optimization of oil pump power consumption and solenoid valve throttling losses by introducing multi-objective optimization with exponential safety penalty functions while ensuring lubrication and cooling safety redundancy; it suppresses pressure oscillations in multi-pump parallel systems by decoupling compensation based on relative gain matrices; and it mitigates the decrease in control accuracy caused by oil aging and oil pump wear by indirectly observing changes in oil viscosity and volumetric efficiency and automatically updating control parameters by monitoring the current-to-speed ratio. Attached Figure Description

[0028] Further details, features, and advantages of this application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0029] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0030] Several embodiments of this application will now be described in more detail with reference to the accompanying drawings to enable those skilled in the art to implement this application. This application may be embodied in many different forms and for various purposes and should not be limited to the embodiments set forth herein. These embodiments are provided to make this application thorough and complete, and to fully convey the scope of this application to those skilled in the art. The embodiments described do not limit this application.

[0031] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It will be further understood that terms such as those defined in commonly used dictionaries shall be interpreted as having a meaning consistent with their meaning in the relevant field and / or the context of this specification, and shall not be interpreted in an idealized or overly formal sense unless expressly defined herein.

[0032] Example 1

[0033] Its specific implementation method is combined with the appendix Figure 1 Please provide a detailed explanation.

[0034] In this embodiment, it includes:

[0035] Step 1: Acquire multi-source heterogeneous signals and perform feature-level fusion to obtain the fused state vector.

[0036] Heterogeneous signals from multiple domain controllers are acquired via the vehicle communication bus and sensor interfaces. Adaptive weighted fusion processing is then performed on these multi-source heterogeneous signals to obtain a fused state vector. This fused state vector includes filtered and confidence-weighted oil pressure estimates, temperature change trends, and driving intent characteristics.

[0037] The aforementioned multi-source heterogeneous signals include at least the following three categories: the first category is driving intention signals, including the accelerator pedal depth change rate and brake energy recovery intensity signals from the vehicle controller; the second category is thermal load signals, including cell temperature data from the battery management system and power device temperature estimates from the motor controller; and the third category is actuator status signals, including the current oil pump speed, the current solenoid valve opening, and oil passage pressure data obtained after high-frequency sampling by the pressure sensor.

[0038] Before the fusion process, the numerical components of the three types of signals are preprocessed using the Z-score normalization method to eliminate the influence of the dimensional differences between different physical quantities on the weight comparison in the fusion calculation, so that each signal component participates in the subsequent fusion calculation on a unified dimensionless scale.

[0039] The specific process of fusion processing includes the following sub-steps:

[0040] Step 101: Perform fast Fourier transform on the oil pressure data to obtain the pressure pulsation spectrum.

[0041] Step 102: Perform time alignment processing on the above three types of signals to obtain a synchronization signal sequence. The specific method of time alignment processing is as follows: using the local clock of the electronic oil pump control unit as the reference time axis, align the timestamps carried in the signal frames reported by each domain controller through the vehicle communication bus with the reference time axis. Compensate for time deviations caused by communication delays or inconsistent sampling periods using a linear interpolation algorithm, so that each signal component corresponds to the sampled value of the same physical time under the same time index, thereby eliminating the timing misalignment caused by asynchronous sampling of multi-source signals.

[0042] Step 103: Based on the synchronization signal sequence, a Kalman filter algorithm is used for fusion calculation to obtain a fused state vector. The input to the Kalman filter algorithm is the aforementioned synchronization signal sequence, and the output is the optimal estimate of each state component in the fused state vector. The weight coefficients of each signal component in the fusion calculation are dynamically adjusted according to the vehicle's driving state. Specifically, the wheel speed disturbance value is calculated based on the wheel speed sensor signals from the vehicle's electronic stability system. This wheel speed disturbance value reflects the degree of road bumpiness. When the wheel speed disturbance value exceeds a preset smoothness threshold, it is determined that the vehicle is driving on a bumpy road surface. At this time, the weight coefficient of the pressure sensor fluctuation component in the fusion calculation is reduced, while the weight coefficient of the virtual pressure estimate calculated based on the oil pump speed model is increased. The aforementioned virtual pressure estimate is calculated based on the current oil pump speed and the pre-stored mapping relationship between speed and pressure. This dynamic weight adjustment avoids misjudging oil pressure sensor reading fluctuations caused by road bumps as actual system depressurization or pressurization events.

[0043] Furthermore, the specific implementation of the above dynamic weight adjustment in the Kalman filter algorithm is as follows: when the bumpy road condition is determined, the corresponding diagonal elements in the observation noise covariance matrix corresponding to the pressure sensor fluctuation component are increased, so that the Kalman gain responds less to the pressure sensor fluctuation component, which is equivalent to reducing the fusion weight of the pressure sensor fluctuation component; at the same time, the corresponding diagonal elements in the observation noise covariance matrix corresponding to the virtual pressure estimate are decreased, so that the Kalman gain responds more to the virtual pressure estimate, which is equivalent to increasing the fusion weight of the virtual pressure estimate; the adjustment amounts of the above two sets of diagonal elements are pre-calibrated fixed step values, which are stored in the electronic oil pump control unit.

[0044] Furthermore, the aforementioned smoothness threshold is a fixed threshold obtained by pre-calibrating wheel speed disturbance data collected under various road conditions. It is stored in the electronic oil pump control unit and is used to distinguish the boundary of wheel speed disturbance amplitude between normal driving conditions and bumpy road conditions.

[0045] Step 2: Based on the fused state vector, perform operating condition identification and traffic demand prediction to obtain the target traffic demand envelope.

[0046] The fused state vector is input into the operating condition identification and prediction process. First, the current operating condition mode type is identified. Then, based on the identification result of the operating condition mode type, the lubrication and cooling flow requirements in the near future are predicted to obtain the target flow requirement envelope. The target flow requirement envelope is a time-series curve describing the upper and lower limits of the required flow rate at each moment within a preset future time window.

[0047] The above-mentioned working condition identification and prediction process includes the following sub-steps:

[0048] Step 201: Input the multi-dimensional feature data from the fused state vector into the fuzzy neural network and output the membership distribution of the current operating mode type. The operating mode type includes, but is not limited to, normal cruise operating mode, high-power fast charging pre-operating mode, high-torque low-speed motor operating mode, and low-temperature cold start operating mode.

[0049] The input layer of the fuzzy neural network receives the feature components from the fused state vector. The hidden layer uses a fuzzy inference mechanism to fuzzify the input features and extract working condition-related features. The output layer is a fully connected layer with the same number of neurons as the number of working condition mode types. Each neuron corresponds to one working condition mode type. After normalization by the Softmax activation function, the output layer outputs the membership value corresponding to each working condition mode type, and the sum of the membership values ​​is 1. The fuzzy neural network uses supervised training, with historical working condition labeled data as training samples and the true membership labels of each working condition mode type as supervision signals. The cross-entropy loss function is used to calculate the difference between the predicted membership distribution and the true labels. The Adam optimization algorithm is used to update the network parameters so that the network output membership distribution converges to a distribution that matches the true working condition categories.

[0050] Furthermore, the fuzzification process of the hidden layer of the aforementioned fuzzy neural network is as follows: for each feature component received by the input layer, a corresponding membership function is set, mapping the value of each feature component to the membership value of each feature component in the range of each fuzzy linguistic variable; the membership function is a Gaussian membership function, and the mean and standard deviation parameters of the Gaussian membership function are updated together with other network parameters through the Adam optimization algorithm during the network training phase; the hidden layer then performs a weighted combination of the fuzzy membership values ​​of each feature component, extracts the fuzzy rule features related to the working condition category, and then passes them to the output layer to calculate the membership distribution of the working condition category.

[0051] Step 202: Determine the current working condition mode based on the working condition mode type with the highest membership degree in the membership degree distribution.

[0052] Step 203: Extract the motor torque sequence within the past preset sampling time period from the fused state vector, input the motor torque sequence into the short-term flow demand predictor based on Long Short-Term Memory (LSTM) network, combine it with the flow baseline parameters corresponding to the current operating mode, and output the predicted value of lubricating oil demand flow at each moment within the future preset prediction time window.

[0053] The input layer of the short-term traffic demand predictor based on a Long Short-Term Memory (LSTM) network receives the motor torque sequence from a preset sampling period in the past and the traffic baseline parameters corresponding to the current operating mode. The hidden layer of the LSM network extracts temporal features from the input sequence. The output layer is a linear fully connected layer that maps the hidden state of the LSM network at the final moment to the traffic prediction values ​​at each moment within a preset prediction time window. The number of neurons in the output layer is the same as the number of moments in the prediction time window, and each neuron corresponds to the traffic prediction value at one moment. The short-term traffic demand predictor based on the LSM network adopts a supervised training method, using historical motor torque sequences as input samples and measured lubricating oil traffic flow data within the corresponding time period as supervision labels. The mean squared error loss function is used to measure the deviation between the predicted value and the measured value, and the Adam optimization algorithm is used to update the network parameters so that the predicted output converges to the result that matches the measured traffic flow. In this process, the motor torque sequence input to the short-term flow demand predictor based on the long short-term memory network is preprocessed using a mean normalization method based on the range during both the training and inference phases to eliminate the dimensional differences between the torque value range and the flow prediction value, thus ensuring the convergence of the network training. The normalized flow prediction value output by the network is restored to the flow prediction value in actual physical units after inverse normalization transformation, and is used as the input for step 204.

[0054] Furthermore, the flow baseline parameter corresponding to the current operating mode is the lubricating oil demand benchmark flow value pre-calibrated in bench tests for each operating mode, and is stored in the electronic oil pump control unit in the form of a lookup table; in the input layer of the short-term flow demand predictor based on the long short-term memory network, the flow baseline parameter and the motor torque sequence are concatenated together as the input feature vector, so that the short-term flow demand predictor based on the long short-term memory network outputs the flow prediction result that matches the lubrication demand benchmark of the operating mode under different operating modes.

[0055] Step 204: Based on the predicted traffic volume at each time point and the redundancy coefficient corresponding to the current operating mode, calculate the upper and lower limits of traffic volume at each time point to generate the target traffic demand envelope. Specifically, let the first... The predicted flow rate at time is The redundancy coefficient corresponding to the current operating mode is Then the first Maximum flow rate at any time and traffic lower limit They are respectively:

[0056] ;

[0057] ;

[0058] in, For the first The predicted flow rate at any given time, in liters per minute. This is the time index within the prediction time window, and its value range is each discrete time within the prediction time window; The redundancy coefficient is greater than zero and less than 1. The value of the redundancy coefficient is retrieved from a pre-stored mapping table of operating conditions and redundancy coefficients based on the current operating mode, to ensure... Not less than zero; and Units and The same, together forming the target flow demand envelope, in the first... The upper and lower limits of time.

[0059] It should be noted that the identification conditions for the above-mentioned high-power fast charging pre-condition are as follows: the fused state vector contains a valid charging gun connection signal, the battery state of charge (SOC) is lower than a preset low charge threshold, and the battery temperature is higher than a preset high temperature threshold. When identified as a high-power fast charging pre-condition, the lower limit of the target flow demand envelope generated in step 204 is set to be no less than the corresponding flow value of the auxiliary cooling oil pump at the standby cruise speed, in order to prepare cooling capacity in advance.

[0060] It should be noted that the identification conditions for the above-mentioned high torque low speed motor operating condition are as follows: the motor torque value included in the fused state vector is greater than a preset proportional threshold of the maximum torque, the vehicle speed is lower than a preset low speed threshold, and the duration of the above state exceeds a preset duration threshold. When the motor is identified as operating in a high torque low speed condition, the flow prediction value output in step 203 is replaced with the fixed flow value of the oil pump in the maximum flow open-loop control mode, which is not subject to the constraints of the conventional adjustment limit.

[0061] It should be noted that the above-mentioned preset sampling time period can be the past 10 seconds, the above-mentioned preset prediction time window can be the future 2 seconds, and the above-mentioned specific time parameters can be adjusted according to the response speed and computing power of the actual control system.

[0062] Step 3: Based on the target flow demand envelope, solve for the optimal hydraulic state parameters under multi-objective constraints to obtain the target oil pump speed and target solenoid valve opening.

[0063] Using the oil pump speed and solenoid valve opening as decision variables, the total system energy consumption and lubrication and cooling safety redundancy as optimization objectives, and the target flow demand envelope as flow constraint, a multi-objective optimization problem is solved to obtain the target oil pump speed and target solenoid valve opening.

[0064] The above solution process includes the following sub-steps:

[0065] Step 301: Based on the efficiency mapping relationship between the oil pump's rotational speed and electrical power consumption, and the throttling loss relationship of the solenoid valve at different opening degrees, construct the objective function. The expression of the objective function is:

[0066] ;

[0067] in, This refers to the oil pump speed, measured in revolutions per minute. This refers to the oil flow rate, measured in liters per minute. The solenoid valve opening degree has a range of values. The dimensionless normalized value, where 0 represents all off and 1 represents all on; For the oil pump at the speed and traffic The power consumption is expressed in watts. The value is obtained by querying a pre-stored efficiency mapping table; For the solenoid valve at the opening degree The power loss due to throttling caused by excess oil flowing back through the overflow valve is expressed in watts. The value is obtained by querying the pre-stored throttling loss mapping table; This is the current oil temperature, in degrees Celsius. This is the difference between the oil passage pressure and the minimum critical pressure required to maintain the oil film thickness, expressed in Pascals. The value is the safety penalty function value, and the value is a dimensionless penalty value. , These are dimensionless weighting coefficients; The unit is watt, so that the objective function After unifying all dimensions to the power dimension, a weighted sum is performed.

[0068] Furthermore, the above objective function middle, and All within the prediction time window Calculate the objective function for each of the corresponding decision variable values. The summation of the above values ​​at each moment within the prediction time window is used as the overall optimization objective, i.e., the objective function. The complete form is for all moments within the prediction time window. The weighted energy consumption and safety penalty terms are accumulated to reflect the comprehensive consideration of the optimization objective on the system state throughout the entire prediction time window.

[0069] Step 302: Set constraints. Constraints include: oil pump speed. Within the permissible speed range Inside, among which and These are the minimum and maximum permissible speeds of the oil pump, both in revolutions per minute (rpm), determined by the oil pump hardware specifications and pre-stored in the electronic oil pump control unit; solenoid valve opening degree. Within the allowable opening range Inside, among which and These represent the minimum and maximum allowable opening degrees of the solenoid valve, respectively; and the time intervals corresponding to the target flow demand envelope. The oil pump rotates at a certain speed. and opening degree The actual traffic that can be provided is no less than the minimum traffic limit at that moment. .

[0070] Step 303: Solve the above objective function under constraints using a constrained optimization algorithm to obtain the objective function. The target oil pump speed and target solenoid valve opening corresponding to the minimum value.

[0071] Furthermore, the above-mentioned constraint optimization solution algorithm employs the Sequential Quadratic Programming (SQP) algorithm, which transforms the objective function... A second-order Taylor expansion approximation is performed on the current decision variables to linearize the constraints, constructing a quadratic programming subproblem. The objective function is then obtained by iteratively solving this subproblem. The search direction descends, gradually converging to the optimal solution that satisfies all constraints. Within each control cycle, the solution result of the previous control cycle is used as the initial point to start the iteration, so as to reduce the number of iteration steps required for convergence and ensure that the solution is completed within the control cycle time.

[0072] It should be noted that the above security penalty function It exhibits nonlinear characteristics. When Greater than zero and Safety penalty function when within the normal temperature range Take the smaller constant value. When When the oil pressure is less than or equal to zero, that is, when the oil passage pressure is insufficient to maintain the minimum critical oil film thickness, or when When the preset temperature limit is exceeded, a safety penalty function is activated. It increases exponentially, specifically:

[0073] ;

[0074] in, Represents an exponential function; This is the penalty gain coefficient when the pressure is insufficient. This is the penalty gain coefficient when the temperature exceeds the limit. This is the preset upper limit value for oil temperature; Before substituting into the above formula, a decimal scaling normalization method is used to perform dimensionless processing. and The difference is also processed using the decimal scaling normalization method to make it dimensionless, so that the independent variable of the exponential function is a dimensionless value. and All are dimensionless gain coefficients.

[0075] Furthermore, when Greater than zero and Safety penalty function when within the normal temperature range The smaller constant value chosen is a positive number greater than zero to ensure that the safety penalty term still applies to the objective function under normal operating conditions. This generates a non-zero contribution, enabling the sequential quadratic programming algorithm to maintain a continuous focus on safety margins while optimizing energy consumption; when Less than or equal to zero or Exceed When the independent variable of the exponential function changes from zero to a positive value, the safety penalty function... The value jumps from the constant value to an exponential value greater than 1, thus forming a significant penalty weight in the objective function, driving the sequential quadratic programming algorithm to prioritize satisfying the oil supply safety constraint.

[0076] By exponentially amplifying the safety penalty function, under conditions of insufficient oil pressure or excessive temperature, the sequential quadratic programming algorithm automatically abandons the goal of optimal energy consumption and prioritizes ensuring the safety of oil supply.

[0077] In this embodiment of the application, in order to adapt to the efficiency characteristics of the oil pump in different operating stages, the efficiency mapping relationship between the oil pump speed and the power consumption is stored in the electronic oil pump control unit by lookup table interpolation. The efficiency mapping relationship is obtained by bench calibration before leaving the factory and includes the corresponding data of speed and power consumption under multiple oil temperature conditions and multiple flow demand conditions.

[0078] Step 4: Based on the target oil pump speed and the target solenoid valve opening, generate control commands and distribute them to each actuator to complete the coordinated control.

[0079] The target oil pump speed is converted into a speed control command for the oil pump drive circuit, and the target solenoid valve opening is converted into an opening control command for the solenoid valve drive circuit. The above control commands are then sent to the corresponding actuator drive circuits for execution.

[0080] In this embodiment, to improve the speed at which control commands follow the target value, feedforward compensation processing is included in addition to step 4. Specifically, based on the mapping relationship between the reverse torque of the oil pump and its rotational speed and pressure, and according to the current estimated oil viscosity and the target pressure, the feedforward rotational speed increment is obtained through table lookup and linear interpolation algorithms. Increase the feedforward speed. The compensated target pump speed is obtained by superimposing this onto the target pump speed, and then converted into a speed control command. The above inverse mapping relationship is pre-stored in the electronic pump control unit in the form of a lookup table. The inverse mapping relationship table stores the feedforward speed increment values ​​corresponding to multiple combinations of viscosity and target pressure.

[0081] Furthermore, the above-mentioned current viscosity estimate of the oil is obtained as follows: after each degradation compensation step (steps 501 to 503) is executed, based on the ratio offset calculated at that time. From the pre-stored ratio offset The corresponding viscosity estimate is retrieved from the mapping table between the viscosity estimate and the viscosity estimate, and the viscosity estimate is updated in the viscosity estimate register of the electronic oil pump control unit. During the interval between the two degradation compensation steps, the feedforward compensation process uses the most recent viscosity estimate stored in the viscosity estimate register to participate in the table lookup calculation.

[0082] In this embodiment of the application, for a system employing a parallel structure of a main oil pump and an auxiliary oil pump, in order to eliminate the hydraulic coupling effect caused by the common rail oil circuit between the main oil pump and the auxiliary oil pump, the following decoupling process is further included in step 4:

[0083] Step 401: Based on the hydraulic characteristic parameters of the main oil pump and the auxiliary oil pump, the relative gain matrix (RGA) analysis method is used to calculate the cross-influence coefficient of the main oil pump speed change on the actual displacement of the auxiliary oil pump. The relative gain matrix analysis method takes the system's steady-state gain matrix as input and outputs the relative gain values ​​between each control channel. The relative gain value is taken from the corresponding control channel.

[0084] Furthermore, the steady-state gain matrix of the above system is obtained as follows: During the bench calibration phase, step excitations are applied to the main oil pump speed and the auxiliary oil pump speed, respectively. The changes in the main oil pump flow rate and the auxiliary oil pump flow rate under steady-state conditions are recorded. The steady-state response gains of each input channel to each output channel are arranged according to the channel correspondence to form a steady-state gain matrix. The relative gain matrix analysis method calculates the relative gain values ​​between each control channel based on the steady-state gain matrix. Take the relative gain value of the main oil pump speed input channel to the auxiliary oil pump flow output channel.

[0085] Step 402: When the target speed of the main oil pump changes, based on the cross-influence coefficient... and the change in the target speed of the main oil pump The speed compensation amount of the auxiliary oil pump is calculated according to the following formula. :

[0086] ;

[0087] in, This is the speed compensation amount for the auxiliary oil pump, expressed in revolutions per minute (rpm). The cross-influence coefficient is a dimensionless value. This represents the target speed change of the main oil pump, expressed in revolutions per minute (rpm). The negative sign indicates that the compensation direction is opposite to the direction of the coupled disturbance, meaning that the coupling effect is offset through reverse compensation. The speed compensation amount... The target speed of the auxiliary oil pump is superimposed on the target speed of the auxiliary oil pump to obtain the decoupled target speed of the auxiliary oil pump.

[0088] Furthermore, the aforementioned cross-influence coefficients The hydraulic characteristic parameters are obtained based on the calibration under steady-state operating conditions and stored as fixed values ​​in the electronic oil pump control unit. When the hydraulic characteristics of the system deviate significantly due to changes in oil viscosity or pipeline aging, the relative gain matrix analysis can be re-executed and the cross-influence coefficients updated. The method of storing values ​​corrects the accuracy of decoupling compensation.

[0089] Step 403: Convert the decoupled auxiliary oil pump target speed into a corresponding speed control command and send it to the auxiliary oil pump drive circuit for execution.

[0090] Through the above decoupling process, when the main oil pump speed increases, leading to a rise in common rail pressure, and the auxiliary oil pump inlet pressure increases, causing a decrease in the actual displacement of the auxiliary oil pump, the electronic oil pump control unit uses the cross-influence coefficient... Positive compensation is applied to the given speed of the auxiliary oil pump to counteract the coupling effect caused by pressure fluctuations.

[0091] In this embodiment of the application, in order to compensate for the impact of oil viscosity drift and oil pump mechanical wear on control accuracy, the following degradation compensation steps are also included in the control cycle consisting of steps 1 to 4:

[0092] Step 501: When the vehicle is in a low-load cruise steady-state condition, collect the current drive current value of the oil pump. and current speed value Calculate the ratio of current to rotational speed. ,in The unit is ampere. The unit is revolutions per minute. The unit is ampere-minute per revolution.

[0093] Step 502: Set the current ratio Ratio to factory standard Compare and calculate the ratio offset. ,in The reference ratio, calibrated under standard oil viscosity and new pump conditions during low-load cruise steady-state operation before the vehicle leaves the factory, is stored as a fixed value in the electronic oil pump control unit.

[0094] Step 503: Based on ratio offset The mapping relationship between the preset offset and compensation parameters is used to obtain the corresponding control parameter compensation value. When the ratio offset is positive and the offset amplitude is within the first offset interval, it is determined that the viscosity is increasing. The corresponding feedforward torque compensation coefficient increment is obtained and updated to the inverse mapping relationship table used for feedforward compensation in step 4. When the ratio offset is positive and the offset amplitude is within the second offset interval greater than the upper limit of the first offset interval, it is determined that the oil pump volumetric efficiency is decreasing. The corresponding speed and flow rate curve slope correction value is obtained and updated to the oil pump efficiency mapping relationship in step 3.

[0095] Furthermore, the boundary values ​​between the first and second offset intervals were obtained through bench testing, wherein the first offset interval corresponds to the ratio offset when the oil viscosity increases within the normal aging range. The typical amplitude range, the second offset range corresponds to the ratio offset when the volumetric efficiency of the oil pump decreases observably due to mechanical wear. The typical amplitude range has a maximum offset protection threshold at the upper limit of both ranges. When the ratio offsets... A fault alarm is triggered when the upper limit of the second offset interval is exceeded, and automatic compensation updates will no longer be performed.

[0096] It should be noted that the execution frequency of the above degradation compensation steps is lower than that of the regular control cycle consisting of steps 1 to 4. The degradation compensation steps are only triggered after the vehicle enters a low-load cruise steady-state condition and the steady-state duration exceeds the preset steady-state determination duration, so as to ensure the steady-state representativeness of the collected data.

[0097] Example 2

[0098] A multi-condition high-efficiency collaborative control system for electronic oil pumps in new energy vehicles includes:

[0099] The multi-source signal acquisition and fusion module is used to acquire multi-source heterogeneous signals from multiple domain controllers, perform adaptive weighted fusion processing on the multi-source heterogeneous signals, and obtain a fused state vector.

[0100] The working condition identification and flow prediction module is used to input the fused state vector into the working condition identification and prediction process, identify the current working condition mode type, and predict the lubrication and cooling flow demand at each moment within the preset prediction time window based on the current working condition mode type, and generate the target flow demand envelope.

[0101] The multi-objective optimization solution module is used to solve the multi-objective optimization problem with oil pump speed and solenoid valve opening as decision variables, total system energy consumption and lubrication and cooling safety redundancy as optimization objectives, and the target flow demand envelope as flow constraint condition, so as to obtain the target oil pump speed and target solenoid valve opening.

[0102] The control command generation and distribution module is used to convert the target oil pump speed into a speed control command for the oil pump drive circuit, convert the target solenoid valve opening degree into an opening control command for the solenoid valve drive circuit, and send the control commands to the corresponding actuator drive circuits for execution.

[0103] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles, characterized in that, include: Multi-source heterogeneous signals from multiple domain controllers are collected, and adaptive weighted fusion processing is performed on the multi-source heterogeneous signals to obtain a fused state vector; The fused state vector includes filtered and confidence-weighted oil pressure estimates, temperature change trend values, and driving intention feature values. The fused state vector is input into the working condition identification and prediction process to identify the current working condition mode type and predict the lubrication and cooling flow requirements at each moment within a preset prediction time window based on the current working condition mode type, thereby generating a target flow requirement envelope; the target flow requirement envelope is a time series curve describing the upper and lower limits of the required flow at each moment within the prediction time window. Using the oil pump speed and solenoid valve opening as decision variables, the total system energy consumption and lubrication and cooling safety redundancy as optimization objectives, and the target flow demand envelope as flow constraint, a multi-objective optimization problem is solved to obtain the target oil pump speed and target solenoid valve opening. The target oil pump speed is converted into a speed control command for the oil pump drive circuit, and the target solenoid valve opening is converted into an opening control command for the solenoid valve drive circuit. The control commands are then sent to the corresponding actuator drive circuits for execution.

2. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, The multi-source heterogeneous signals include three categories: the first category is driving intention signals, including the accelerator pedal depth change rate and brake energy recovery intensity signals from the vehicle controller; the second category is thermal load signals, including cell temperature data from the battery management system and power device temperature estimates from the motor controller; and the third category is actuator status signals, including the current oil pump speed, the current solenoid valve opening, and oil passage pressure data obtained after high-frequency sampling by the pressure sensor. The adaptive weighted fusion processing includes: performing a fast Fourier transform on the oil pressure data to obtain the pressure pulsation spectrum; The three types of signals are time-aligned using the local clock of the electronic oil pump control unit as the reference time axis. A linear interpolation algorithm is used to compensate for time deviations caused by communication delays or inconsistent sampling periods to obtain a synchronization signal sequence. Based on the synchronization signal sequence, a Kalman filter algorithm is used for fusion calculation to obtain the fused state vector.

3. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, In the Kalman filter algorithm, the weight coefficients of each signal component are dynamically adjusted according to the vehicle's driving state. Specifically, the wheel speed disturbance value is calculated based on the wheel speed sensor signals from the vehicle's electronic stability system. When the wheel speed disturbance value exceeds a preset smoothness threshold, it is determined that the vehicle is driving on a bumpy road. The corresponding diagonal elements in the observation noise covariance matrix corresponding to the pressure sensor fluctuation component are increased, thus weakening the Kalman gain's response to the pressure sensor fluctuation component. At the same time, the corresponding diagonal elements in the observation noise covariance matrix corresponding to the virtual pressure estimate are decreased, thus enhancing the Kalman gain's response to the virtual pressure estimate. The virtual pressure estimate is calculated based on the current speed of the oil pump and the pre-stored mapping relationship between speed and pressure; the smoothness threshold is a fixed threshold obtained by pre-calibrating wheel speed disturbance data collected under various road conditions.

4. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, Identifying the current operating condition mode type includes: inputting multi-dimensional feature data from the fused state vector into a fuzzy neural network and outputting the membership distribution of the current operating condition mode type; the input layer of the fuzzy neural network receives each feature component from the fused state vector, and the hidden layer maps each feature component to a membership value in the range of each fuzzy linguistic variable using a Gaussian membership function; the fuzzy rule features are extracted after weighted combination of the fuzzy membership values ​​of each feature component; the output layer outputs the membership values ​​corresponding to each operating condition mode type after normalization using a Softmax activation function; the mean and standard deviation parameters of the Gaussian membership function are updated using the Adam optimization algorithm during the network training phase; and the current operating condition mode is determined based on the operating condition mode type with the highest membership degree in the membership distribution.

5. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, Predicting the lubrication and cooling flow rate demand at each moment within a preset prediction time window includes: extracting the motor torque sequence from the fused state vector within a preset sampling period in the past; inputting the motor torque sequence into a short-term flow rate demand predictor based on a long short-term memory network; combining the flow rate baseline parameters corresponding to the current operating mode; and outputting the predicted lubricating oil flow rate demand at each moment within the preset prediction time window. The flow rate baseline parameters are the lubricating oil demand benchmark flow rate values ​​pre-calibrated in bench tests for each operating mode. The flow rate baseline parameters and the motor torque sequence are concatenated together to serve as the input feature vector of the short-term flow rate demand predictor. The short-term flow rate demand predictor uses a mean squared error loss function, takes the historical motor torque sequence as input samples, and uses the measured lubricating oil flow rate data within the corresponding time period as supervised labels for supervised training.

6. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, Generating the target flow demand envelope includes: calculating the upper and lower flow limits for each time moment based on the predicted flow value at each time moment and the redundancy coefficient corresponding to the current operating mode; the upper flow limit for each time moment is equal to the product of the redundancy coefficient plus 1 and the predicted flow value at that time moment, and the lower flow limit for each time moment is equal to the product of 1 minus the redundancy coefficient and the predicted flow value at that time moment; the redundancy coefficient is a value greater than zero and less than 1, and is retrieved from a pre-stored mapping table of operating modes and redundancy coefficients according to the current operating mode; when identified as a high-power fast charging pre-operating mode, the lower flow limit is set to be no less than the corresponding flow value of the auxiliary cooling oil pump at the standby cruise speed; when identified as a high-torque low-speed motor operating mode, the predicted flow value is replaced with the fixed flow value of the oil pump in the maximum flow open-loop control mode.

7. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, Solving the multi-objective optimization problem involves: constructing an objective function, which is the cumulative value of weighted energy consumption and safety penalty terms at all times within the prediction time window. The objective function includes a weighted sum of three terms: the first term is the electrical power consumption of the oil pump at the corresponding speed and flow rate; the second term is the throttling heat loss power of the solenoid valve at the corresponding opening degree; and the third term is the safety penalty function. The safety penalty function takes a constant value greater than zero when the difference between the oil passage pressure and the minimum critical pressure required to maintain the oil film thickness is greater than zero and the oil temperature is within the normal temperature range. When the difference is less than or equal to zero or the oil temperature exceeds the preset upper limit of oil temperature, the value of the safety penalty function is an exponential function value with the weighted sum of the pressure deficiency and temperature over-limit after dimensionless processing as the exponent; the set constraints include the oil pump speed being within the allowable speed range, the solenoid valve opening being within the allowable opening range, and the actual flow rate at each moment not being lower than the lower limit of flow rate; the objective function is iteratively solved under the constraints using a sequential quadratic programming algorithm, and the solution result of the previous control cycle is used as the initial point to start the iteration in each control cycle.

8. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, It also includes feedforward compensation processing and decoupling processing; the feedforward compensation processing includes: based on the mapping relationship between the reverse torque of the oil pump and its speed and pressure, according to the current oil viscosity estimate and target pressure value, obtaining the feedforward speed increment through table lookup and linear interpolation algorithm, and superimposing the feedforward speed increment onto the target oil pump speed to obtain the compensated target oil pump speed; the decoupling processing includes: based on the hydraulic characteristic parameters of the main oil pump and the auxiliary oil pump, using the relative gain matrix analysis method to calculate the cross-influence coefficient of the main oil pump speed change on the actual displacement of the auxiliary oil pump; when the target speed of the main oil pump changes, the speed compensation amount of the auxiliary oil pump is equal to the negative of the product of the cross-influence coefficient and the change in the target speed of the main oil pump; the speed compensation amount is superimposed onto the target speed of the auxiliary oil pump to obtain the decoupled target speed of the auxiliary oil pump.

9. The multi-condition high-efficiency collaborative control method for electronic oil pumps in new energy vehicles according to claim 1, characterized in that, It also includes a degradation compensation step, which is triggered after the vehicle is in a low-load cruise steady-state condition and the steady-state duration exceeds a preset steady-state determination duration, including: The current drive current and current speed of the oil pump are collected, and the ratio of current to speed is calculated. The ratio is compared with the factory reference ratio to obtain the ratio offset. When the ratio offset is positive and the offset is within the first offset interval, it is determined that the viscosity has increased. The corresponding feedforward torque compensation coefficient increment is obtained and updated to the inverse mapping table used by the feedforward compensation processing. When the ratio offset is positive and the offset is within the second offset interval greater than the upper limit of the first offset interval, it is determined that the volumetric efficiency of the oil pump has decreased. The corresponding speed and flow curve slope correction value is obtained and updated to the oil pump efficiency mapping relationship in the multi-objective optimization problem. When the ratio offset exceeds the upper limit of the second offset interval, a fault alarm is triggered. The current viscosity estimate of the oil is obtained from the pre-stored mapping table based on the ratio offset calculated in this operation.

10. A multi-condition high-efficiency collaborative control system for an electronic oil pump in a new energy vehicle, comprising a multi-condition high-efficiency collaborative control method for an electronic oil pump in a new energy vehicle according to any one of claims 1-9, characterized in that, include: The multi-source signal acquisition and fusion module is used to acquire multi-source heterogeneous signals from multiple domain controllers, perform adaptive weighted fusion processing on the multi-source heterogeneous signals, and obtain a fusion state vector. The working condition identification and flow prediction module is used to input the fused state vector into the working condition identification and prediction process, identify the current working condition mode type, and predict the lubrication and cooling flow demand at each moment within a preset prediction time window based on the current working condition mode type, and generate the target flow demand envelope. The multi-objective optimization solution module is used to solve the multi-objective optimization problem by taking the oil pump speed and solenoid valve opening as decision variables, the total system energy consumption and lubrication and cooling safety redundancy as optimization objectives, and the target flow demand envelope as flow constraint conditions, so as to obtain the target oil pump speed and target solenoid valve opening. The control command generation and distribution module is used to convert the target oil pump speed into a speed control command for the oil pump drive circuit, convert the target solenoid valve opening degree into an opening control command for the solenoid valve drive circuit, and send the control commands to the corresponding actuator drive circuits for execution.

Images