Self-adaptive rotating speed control method, device and equipment for water pump and storage medium
The deep Q network reinforcement learning algorithm adjusts the pump speed in real time, solves the dynamic matching problem when filling the coolant, improves the engine cooling effect and energy efficiency, and extends the life of the parts.
Patent Information
- Application Number
- CN202510709381.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-29
- Publication Date
- 2025-07-18
AI Technical Summary
The prior art cannot dynamically adjust the pump speed according to real-time operating conditions when filling the coolant, resulting in large fluctuations in engine temperature, affecting performance and possibly causing damage, and high energy consumption, making adaptive optimization impossible.
The deep Q network reinforcement learning algorithm is adopted to collect the working parameters of the engine and water pump in real time, build a state space, select the appropriate speed adjustment action, and iteratively learn through reward signal feedback to optimize the pump speed control strategy.
It realizes real-time and accurate adjustment of the pump speed according to actual conditions, improves the engine cooling effect, reduces energy consumption, and extends the life of parts.
Smart Images
Figure CN120331950A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of new energy vehicle thermal management, and particularly relates to a method, device, equipment and storage medium for adaptive speed control of a water pump. Background Art
[0002] During vehicle operation, the engine generates a large amount of heat, which needs to be effectively dissipated through coolant to maintain the engine within an appropriate operating temperature range. The circulation of the coolant is mainly achieved by a water pump, and the reasonable control of the water pump speed has a crucial impact on aspects such as the cooling effect of the engine, energy consumption, and the service life of components.
[0003] Currently, the traditional water pump speed control methods are relatively simple. For example, one method is to increase the water pump speed during filling to accelerate the coolant circulation and promote gas discharge. However, this method has limitations: the speed increase is a fixed value and cannot be dynamically adjusted according to real-time working conditions (such as the amount of bubbles, temperature gradient), which may lead to excessive energy consumption or local gas resistance residue. Another method is to adopt an automatic switching between manual / auto modes to simplify the operation process. But this method also fails to solve the problems of dynamic exhaust and speed matching, and manual intervention for liquid replenishment is required.
[0004] During coolant filling, due to factors such as the filling volume, filling speed, and the working condition of the engine at that time, the traditional control methods cannot timely adjust the water pump speed according to these changes, easily resulting in large fluctuations in the engine temperature, affecting the engine performance, and even possibly causing problems such as engine overheating and damage. During coolant filling, if the water pump speed is too low, the coolant circulation speed is slow, and the heat generated by the engine cannot be taken away in time, causing the engine temperature to rise rapidly. This will not only reduce the engine efficiency but also may trigger faults such as thermal deformation and increased wear of engine components, and in severe cases, lead to major damages such as engine cylinder pulling and bearing seizure, shortening the service life of the engine. If the water pump speed is too high, although the heat can be quickly taken away, it will increase the engine load, resulting in a significant increase in energy consumption. At the same time, it will also cause the water pump itself and related components to bear excessive pressure and friction, accelerating the wear of these components, increasing the maintenance cost and the probability of failures, and may also lead to problems such as excessive pressure in the cooling system, causing components such as radiators and water pipes to burst and leak coolant.
[0005] Therefore, there is currently a lack of a solution that can adaptively adjust the water pump speed in real time and accurately according to the actual situation during coolant filling to improve the engine cooling effect, reduce energy consumption, and extend the service life of components. Summary of the Invention
[0006] The present application provides a method, device, equipment and storage medium for adaptive speed control of a water pump, which can adaptively adjust the water pump speed in real time and accurately according to the actual situation.
[0007] In a first aspect, an embodiment of the present application provides a method for adaptively controlling the rotational speed of a water pump. The method for adaptively controlling the rotational speed of the water pump includes the following steps:
[0008] During the coolant filling process, the operating parameters of the engine and the current rotational speed of the water pump are collected in real time;
[0009] Integrate the collected data to construct a state space reflecting the real-time states of the engine and the water pump during coolant filling;
[0010] Based on the state space, select a suitable water pump rotational speed adjustment action from a pre-defined action space. After executing the selected action, according to the responses of the engine and the cooling system, feedback a reward signal for evaluating the executed action;
[0011] According to the current states of the engine and the water pump, the executed action, and the reward signal, perform iterative learning to optimize the water pump rotational speed control strategy under different states.
[0012] Combined with the first aspect, in an embodiment, based on the state space, selecting a suitable water pump rotational speed adjustment action from a pre-defined action space includes:
[0013] Based on the deep Q-network reinforcement learning algorithm, calculate the Q values of each action according to the state space;
[0014] Select the action with the maximum Q value as the decision result according to the Q value.
[0015] Combined with the first aspect, in an embodiment, the iterative learning according to the current states of the engine and the water pump, the executed action, and the reward signal to optimize the water pump rotational speed control strategy under different states includes:
[0016] Construct a training sample according to the current states of the engine and the water pump, the executed action, and the reward signal;
[0017] Input the training sample into the deep Q-network for training, and calculate the loss function through the backpropagation algorithm;
[0018] Adjust the weights and biases of the deep Q-network according to the loss function to minimize the loss function, and obtain the optimal water pump rotational speed control strategy under different states through iterative learning.
[0019] Combined with the first aspect, in an embodiment, before integrating the collected data, it further includes:
[0020] Preprocess the collected data to remove outliers and noise, and normalize the preprocessed data.
[0021] In combination with the first aspect, in one embodiment, the operating parameters of the engine include engine speed, engine load, engine temperature, coolant temperature, coolant level, filling flow rate of the coolant, and filling pressure.
[0022] In combination with the first aspect, in one embodiment, integrating the collected data to construct a state space reflecting the real-time states of the engine and the water pump during coolant filling includes:
[0023] Combining the current speed of the water pump, engine speed, engine load, engine temperature, coolant temperature, coolant level, filling flow rate of the coolant, and filling pressure into a state vector to reflect the real-time states of the engine and the water pump during coolant filling.
[0024] In combination with the first aspect, in one embodiment, the evaluation dimensions of the reward signal include the change situation of the engine temperature, the change situation of the energy consumption, the completion progress of the coolant filling, and the operating conditions of the engine system components.
[0025] In a second aspect, an embodiment of the present application provides a water pump adaptive speed control device, and the water pump adaptive speed control device includes:
[0026] A data acquisition module, which is used to collect the operating parameters of the engine and the current speed of the water pump in real time during the coolant filling process;
[0027] A state construction module, which is used to integrate the collected data to construct a state space reflecting the real-time states of the engine and the water pump during coolant filling;
[0028] A decision module, which selects a suitable water pump speed adjustment action from a pre-defined action space based on the state space,
[0029] A feedback module, which is used to feedback a reward signal for evaluating the executed action according to the responses of the engine and the cooling system after executing the selected action;
[0030] A learning module, which performs iterative learning according to the current states of the engine and the water pump, the executed action, and the reward signal to optimize the water pump speed control strategy under different states.
[0031] In a third aspect, an embodiment of the present application provides a water pump adaptive speed control device, and the water pump adaptive speed control device includes a processor, a memory, and a water pump adaptive speed control program stored on the memory and executable by the processor. When the water pump adaptive speed control program is executed by the processor, the steps of the above-mentioned water pump adaptive speed control method are implemented.
[0032] Fourth aspect, a computer-readable storage medium has a water pump adaptive speed control program stored thereon. When the water pump adaptive speed control program is executed by a processor, the steps of the above-mentioned water pump adaptive speed control method are implemented.
[0033] The beneficial effects brought by the technical solutions provided in the embodiments of the present application at least include:
[0034] In the water pump adaptive speed control method in the present application, during the coolant filling process, the working parameters of the engine and the current speed of the water pump are collected in real time; the collected data is integrated to construct a state space reflecting the real-time states of the engine and the water pump during coolant filling; based on the state space, a suitable water pump speed adjustment action is selected from a pre-defined action space. After executing the selected action, according to the responses of the engine and the cooling system, a reward signal for evaluating the executed action is fed back; according to the current states of the engine and the water pump, the executed action, and the reward signal, iterative learning is performed to optimize the water pump speed control strategy under different states.
[0035] Based on reinforcement learning, the present application adaptively adjusts the water pump speed in real time and accurately according to the actual situation during coolant filling, thereby improving the engine cooling effect, reducing energy consumption, and extending the service life of components. Description of the Drawings
[0036] Figure 1 is a processing flow chart of the water pump adaptive speed control method of the present application;
[0037] Figure 2 is a schematic flow diagram of an embodiment of the water pump adaptive speed control method of the present application;
[0038] Figure 3 is Figure 2 a schematic flow diagram of step S4 in
[0039] Figure 4 is a structural block diagram of an embodiment of the water pump adaptive speed control device of the present application;
[0040] Figure 5 is a schematic hardware structure diagram of the water pump adaptive speed control device involved in the solution of the embodiment of the present application. Detailed Embodiments
[0041] To enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.
[0042] To solve the problems in the prior art, this application provides a method for adaptive speed control of a water pump. Refer to Figure 1 as shown, which mainly includes the following links:
[0043] 1. Data collection: During the coolant filling process, the working parameters of the engine are collected in real time, including but not limited to data such as engine speed, load, temperature, coolant level, coolant temperature, filling flow rate, and filling pressure; at the same time, information such as the current speed and working status of the water pump is collected.
[0044] 2. State construction: The collected data is processed and integrated to construct the state space of the intelligent agent, which can comprehensively and accurately reflect the real-time states of the engine and the water pump during coolant filling.
[0045] 3. Action selection: Based on the reinforcement learning algorithm, the intelligent agent selects a suitable water pump speed adjustment action from the action space according to the current state. The action space includes different speed adjustment amplitudes and directions.
[0046] 4. Execute action and environmental feedback: Execute the selected action, that is, adjust the water pump speed, and observe the responses of the engine and the cooling system to obtain the reward signal feedback from the environment. The reward signal is determined according to factors such as whether the engine temperature is stable within a reasonable range, the change in energy consumption, the completion progress of coolant filling, and the operating conditions of system components. If the engine temperature is stable, the energy consumption is reduced, the filling is completed smoothly, and the components are normal, a positive reward is given; otherwise, if the engine temperature is too high or too low, the energy consumption is too high, the filling is abnormal, or the components are damaged, a negative reward is given.
[0047] 5. Learning and updating: The intelligent agent updates the parameters of the policy network and the value network according to the reward signal and the current state by using the reinforcement learning algorithm to improve the accuracy and optimization degree of the next decision, and continuously iterates and learns, so that the intelligent agent can gradually find the optimal water pump speed control strategy under different states.
[0048] To make the purpose, technical solutions, and advantages of this application clearer, the embodiments of this application will be further described in detail below in conjunction with the accompanying drawings.
[0049] In a first aspect, an embodiment of the present application provides a water pump adaptive speed control method.
[0050] In one embodiment, referring to Figure 2 , Figure 2 This is a flow chart of an embodiment of the method for adaptive speed control of a water pump according to the present application. Figure 2 As shown, the water pump adaptive speed control method includes:
[0051] S1. During the process of adding coolant, the operating parameters of the engine and the current speed of the water pump are collected in real time;
[0052] Specifically, during the coolant filling process, the engine operating parameters are collected in real time through various sensors arranged in the engine and cooling system. For example, the engine speed is collected by the engine speed sensor, the engine load is obtained by the load sensor, the engine temperature and coolant temperature are measured by the temperature sensor, the coolant level is monitored by the liquid level sensor, and the data such as the filling flow rate and filling pressure of the coolant are collected by the filling flow sensor and the filling pressure sensor respectively; at the same time, the current speed and working status of the water pump are collected through the sensor provided by the water pump, where the working status of the water pump refers to the operating status of its various components, that is, whether the operation is normal.
[0053] S2, integrating the collected data to construct a state space reflecting the real-time state of the engine and the water pump when the coolant is added;
[0054] It is worth mentioning that before the collected data is integrated, the data in step S1 will be preprocessed, that is, firstly, obviously unreasonable outliers are removed, for example, the engine temperature suddenly exceeds the normal range, which may be caused by sensor failure, and it is removed; then, a filtering algorithm is used to remove noise interference to make the data smoother and more stable; then, the data is normalized, and data of different ranges and dimensions are uniformly mapped to the interval [0, 1] to facilitate the processing and calculation of subsequent algorithms and improve the data availability and model training effect.
[0055] Then, the preprocessed data is integrated to construct the state space of the intelligent agent. For example, the current speed of the water pump, engine speed, engine load, engine temperature, coolant temperature, coolant level, coolant filling flow and filling pressure are combined into a state vector, which can fully and accurately reflect the real-time status of the engine and water pump when the coolant is filled, providing a basis for the decision-making of the intelligent agent.
[0056] S3, based on the state space, selecting a suitable water pump speed adjustment action from a predefined action space, and after executing the selected action, feeding back a reward signal for evaluating the executed action according to the response of the engine and the cooling system;
[0057] Specifically, the agent selects a suitable water pump speed adjustment action from the action space based on the Deep Q-Network (DQN) reinforcement learning algorithm according to the current state. The action space is predefined and includes different speed adjustment amplitudes, such as increasing by 100 revolutions per minute, increasing by 200 revolutions per minute, decreasing by 100 revolutions per minute, decreasing by 200 revolutions per minute, and maintaining the current speed unchanged, etc. The agent selects the action with the maximum Q value as the current decision-making action by querying the Q-value table.
[0058] Execute the selected action, that is, adjust the water pump speed by controlling the motor drive device of the water pump. For example, if the action of increasing by 100 revolutions per minute is selected, a corresponding control signal is sent to the motor drive device to increase the water pump speed. Then, continuously observe the responses of the engine and the cooling system, and use sensors to monitor in real time the changes in parameters such as the engine temperature, coolant temperature, and energy consumption. Determine the reward signal of the environmental feedback according to the changes in these parameters. If the engine temperature remains stable within a reasonable range (such as 80°C - 90°C) during the filling process, and the energy consumption does not increase significantly, and the coolant filling is successfully completed, a positive reward is given, such as a reward value of +5; conversely, if the engine temperature is too high exceeding 95°C, or the energy consumption increases significantly, or the filling is abnormal (such as too slow or too fast filling speed), a negative reward is given, such as a reward value of -3. Of course, the size of the reward value can be reasonably set as needed, and this embodiment does not limit it here. At the same time, record the current state, the executed action, and the obtained reward signal for subsequent learning and policy update.
[0059] S4. Perform iterative learning according to the current states of the engine and the water pump, the executed action, and the reward signal to optimize the water pump speed control strategy under different states.
[0060] See Figure 3 As shown, in this embodiment, step S4 includes:
[0061] S41. Construct a training sample according to the current states of the engine and the water pump, the executed action, and the reward signal;
[0062] S42. Input the training sample into the deep Q network for training, and calculate the loss function through the backpropagation algorithm;
[0063] S43. Adjust the weights and biases of the deep Q network according to the loss function to minimize the loss function, and obtain the optimal water pump speed control strategy under different states through iterative learning.
[0064] Specifically, the agent updates the parameters of the policy network and the value network using the learning algorithm of the deep Q-network based on the reward signal and the current state. First, training samples are constructed according to the recorded state, action, and reward signal. Then, the training samples are input into the deep Q-network for training. The loss function (such as the mean squared error loss function) is calculated through the backpropagation algorithm, and the weights and biases of the network are adjusted according to the loss function to minimize the loss function and improve the accuracy of Q-value estimation and the optimization degree of the policy. Through continuous iterative learning, the agent can gradually find the optimal water pump speed control strategy under different states, thereby realizing the adaptive optimization control of the water pump speed during coolant filling.
[0065] In summary, the water pump adaptive speed control method in this application collects the working parameters of the engine and the current speed of the water pump in real time during coolant filling; integrates the collected data to construct a state space reflecting the real-time state of the engine and the water pump during coolant filling; based on the state space, selects a suitable water pump speed adjustment action from the pre-defined action space, and after executing the selected action, feedbacks a reward signal for evaluating the executed action according to the responses of the engine and the cooling system; performs iterative learning according to the current states of the engine and the water pump, the executed action, and the reward signal to optimize the water pump speed control strategy under different states.
[0066] This application adaptively adjusts the water pump speed in real time and accurately according to the actual situation during coolant filling based on reinforcement learning, thereby improving the engine cooling effect, reducing energy consumption, and extending the service life of components.
[0067] In a second aspect, an embodiment of this application also provides a water pump adaptive speed control device.
[0068] In one embodiment, referring to Figure 4 , Figure 4 is a schematic diagram of the functional modules of an embodiment of the water pump adaptive speed control device of this application. As Figure 4 shown, the water pump adaptive speed control device includes: a data acquisition module, a data processing module, a state construction module, a decision-making module, an execution module, a feedback module, a learning module, a monitoring and warning module, and an emergency handling module.
[0069] The specific functions of the above modules are introduced below:
[0070] Data acquisition module: It includes a variety of sensors, such as temperature sensors, pressure sensors, flow sensors, rotational speed sensors, etc. The temperature sensors are distributed at positions such as the engine cylinder block and coolant pipelines to collect the engine temperature and coolant temperature; the pressure sensors are installed at the coolant filling pipeline and the pump outlet to collect the filling pressure and pump outlet pressure; the flow sensor is used to measure the filling flow rate of the coolant and the circulating flow rate of the pump; the rotational speed sensors respectively monitor the engine speed and pump speed. These sensors transmit the collected data to the data processing module in real time.
[0071] Data processing module: Receives the data transmitted from the data acquisition module, performs preprocessing, removes outliers and noise, and normalizes the data to improve data quality, and then transmits the processed data to the state construction module.
[0072] State construction module: Integrates the data transmitted from the data processing module, constructs the state space of the agent, generates a state vector, and sends the state vector to the decision-making module.
[0073] Decision-making module: Based on the deep Q-network reinforcement learning algorithm, selects a suitable pump speed adjustment action from the predefined action space according to the received state vector. The decision-making module includes a policy network and a Q-value calculation module. The policy network calculates the Q-values of each action according to the state vector, and the Q-value calculation module selects the action with the maximum Q-value as the decision result according to the Q-values and sends the action instruction to the execution module.
[0074] Execution module: Receives the action instruction sent by the decision-making module, and adjusts the pump speed by controlling the motor drive device of the pump to achieve the control of the pump.
[0075] Feedback module: Monitors the operating status of the engine and cooling system in real time, obtains parameters such as the engine temperature, coolant temperature, and energy consumption collected by the sensors, judges the working conditions of the engine and cooling system according to these parameters, determines the reward signal of the environmental feedback, and sends the reward signal to the learning module. At the same time, records the current state, executed action, and reward signal for subsequent learning and analysis.
[0076] Learning module: Updates the parameters of the policy network and value network according to the reward signal and the current state transmitted by the feedback module using the learning algorithm of the deep Q-network. The learning module includes a training sample generation module and a network training module. The training sample generation module constructs training samples according to the recorded state, action, and reward signal, and the network training module inputs the training samples into the deep Q-network for training, and adjusts the weights and biases of the network through the backpropagation algorithm to improve the accuracy of Q-value estimation and the optimization degree of the policy, and continuously iterates and learns to enable the system to achieve adaptive optimization control of the pump speed during coolant filling.
[0077] Monitoring and early warning module: Real-time monitor the key parameters of the engine and cooling system, and set the normal range thresholds of the parameters. For example, the normal range of the engine temperature is set to 80°C - 90°C, and the normal range of the coolant level is set to 30% - 80% of the range of the liquid level sensor, etc. If it is found that the parameter exceeds the normal range, immediately trigger a warning signal and send the warning information to the emergency handling module.
[0078] Emergency handling module: Receive the warning information transmitted by the monitoring and early warning module, and take corresponding emergency measures according to the warning type. For example, when the engine temperature is too high and exceeds 95°C, limit the further increase of the water pump speed, and at the same time reduce the engine load to reduce the heat generated by the engine; if the coolant filling is abnormal, such as too high or too low filling pressure, stop the coolant filling, and conduct fault troubleshooting and repair.
[0079] Thirdly, the embodiment of the present application provides a water pump adaptive speed control device, and the water pump adaptive speed control device can be a device with data processing functions such as a personal computer (PC), a laptop computer, a server, etc.
[0080] Refer to Figure 5 , Figure 5 which is a schematic diagram of the hardware structure of the water pump adaptive speed control device involved in the solution of the embodiment of the present application. In the embodiment of the present application, the water pump adaptive speed control device may include a processor, a memory, a communication interface, and a communication bus.
[0081] Among them, the communication bus can be of any type and is used to interconnect the processor, the memory, and the communication interface.
[0082] The communication interface includes interfaces such as input / output (I / O) interfaces, physical interfaces, and logical interfaces for realizing the interconnection of internal devices of the water pump adaptive speed control device, and interfaces for realizing the interconnection of the water pump adaptive speed control device with other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, an optical fiber interface, an ATM interface, etc.; the user device can be a display (Display), a keyboard (Keyboard), etc.
[0083] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical memory, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.
[0084] The processor can be a general-purpose processor, which can call the water pump adaptive speed control program stored in the memory and execute the water pump adaptive speed control method provided by the embodiments of the present application. For example, the general-purpose processor can be a central processing unit (CPU). Among them, the method executed when the water pump adaptive speed control program is called can refer to the various embodiments of the water pump adaptive speed control method of the present application, which will not be elaborated here.
[0085] Those skilled in the art can understand that Figure 5 the hardware structure shown in does not constitute a limitation to the present application, and may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.
[0086] In a fourth aspect, the embodiments of the present application further provide a readable storage medium.
[0087] The readable storage medium of the present application stores a water pump adaptive speed control program, where when the water pump adaptive speed control program is executed by a processor, the steps of the water pump adaptive speed control method as described above are implemented.
[0088] Among them, the method implemented when the water pump adaptive speed control program is executed can refer to the various embodiments of the water pump adaptive speed control method of the present application, which will not be elaborated here.
[0089] It should be noted that the serial numbers of the above embodiments of the present application are only for description and do not represent the superiority or inferiority of the embodiments.
[0090] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware. However, in many cases, the former is a better implementation method. Based on such an understanding, the technical solution of the present application, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above and includes several instructions to enable a terminal device to execute the methods described in various embodiments of the present application.
[0091] The terms "including" and "having" and any variations thereof in the description of the specification, claims and drawings of the present application are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products or devices. The descriptions of the terms "first", "second", "third", etc. are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit that "first", "second" and "third" are of different types.
[0092] In the description of the embodiments of the present application, "exemplary", "for example" or "for instance" are used to indicate examples, illustrations or explanations. Any embodiment or design solution described as "exemplary", "for example" or "for instance" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of the words "exemplary", "for example" or "for instance" is intended to present the relevant concepts in a specific manner.
[0093] In the description of the embodiments of the present application, unless otherwise specified, " / " means "or". For example, A / B may mean A or B; "and / or" in the text is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B may mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present application, "a plurality of" means two or more than two.
[0094] In some processes described in the embodiments of the present application, a plurality of operations or steps appear in a specific order. However, it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or may be executed in parallel. The serial numbers of the operations are only used to distinguish different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.
[0095] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.
[0096] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.
Claims
1. An adaptive speed control method for a water pump, characterized in that The described adaptive pump speed control method includes: During the coolant filling process, the working parameters of the engine and the current speed of the pump are collected in real time; Integrate the collected data to construct a state space reflecting the real-time states of the engine and the pump during coolant filling; Based on the state space, select a suitable pump speed adjustment action from a pre-defined action space. After executing the selected action, according to the responses of the engine and the cooling system, feedback a reward signal for evaluating the executed action; According to the current states of the engine and the pump, the executed action, and the reward signal, perform iterative learning to optimize the pump speed control strategy under different states.
2. The water pump adaptive speed control method according to claim 1, wherein, Based on the state space, selecting a suitable pump speed adjustment action from a pre-defined action space includes: Based on the deep Q-network reinforcement learning algorithm, calculate the Q-values of each action according to the state space; Select the action with the maximum Q-value as the decision result according to the Q-values.
3. The water pump adaptive speed control method according to claim 2, characterized in that The performing iterative learning according to the current states of the engine and the pump, the executed action, and the reward signal to optimize the pump speed control strategy under different states includes: Construct a training sample according to the current states of the engine and the pump, the executed action, and the reward signal; Input the training sample into the deep Q-network for training, and calculate the loss function through the backpropagation algorithm; Adjust the weights and biases of the deep Q-network according to the loss function to minimize the loss function, and through iterative learning, obtain the optimal pump speed control strategy under different states.
4. The water pump adaptive speed control method according to claim 1, wherein Before integrating the collected data, it further includes: Preprocess the collected data to remove outliers and noise, and normalize the preprocessed data.
5. The adaptive pump speed control method according to claim 1, characterized in that: The working parameters of the engine include engine speed, engine load, engine temperature, coolant temperature, coolant level, filling flow rate and filling pressure of the coolant.
6. The water pump adaptive speed control method according to claim 5, wherein, The integrating the collected data to construct a state space reflecting the real-time states of the engine and the pump during coolant filling includes: Combine the current speed of the pump, engine speed, engine load, engine temperature, coolant temperature, coolant level, filling flow rate and filling pressure of the coolant into a state vector to reflect the real-time states of the engine and the pump during coolant filling.
7. The adaptive pump speed control method according to claim 1, characterized in that: The evaluation dimensions of the reward signal include the change of engine temperature, the change of energy consumption, the completion progress of coolant filling, and the operating conditions of engine system components.
8. An adaptive speed control device for a water pump, characterized in that, The adaptive pump speed control device includes: A data acquisition module, which is used to collect the working parameters of the engine and the current speed of the pump in real time during the coolant filling process; A state construction module, which is used to integrate the collected data to construct a state space reflecting the real-time states of the engine and the pump during coolant filling; A decision-making module, which selects a suitable pump speed adjustment action from a pre-defined action space based on the state space, A feedback module, which is used to feedback a reward signal for evaluating the executed action according to the responses of the engine and the cooling system after the selected action is executed; A learning module, which performs iterative learning according to the current states of the engine and the water pump, the executed action, and the reward signal to optimize the water pump speed control strategy under different states.
9. An adaptive speed control device for a water pump, characterized in that, The water pump adaptive speed control device includes a processor, a memory, and a water pump adaptive speed control program stored on the memory and executable by the processor. When the water pump adaptive speed control program is executed by the processor, the steps of the water pump adaptive speed control method according to any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium, characterized in that, A water pump adaptive speed control program is stored on the computer-readable storage medium. When the water pump adaptive speed control program is executed by a processor, the steps of the water pump adaptive speed control method according to any one of claims 1 to 7 are implemented.