An electronically controlled software water temperature correction method

Abstract

The present application relates to the technical field of electronic control engine, and discloses a kind of electronic control software water temperature correction method, comprising the following steps: step S101, the actual water temperature value of current time mining vehicle engine is obtained;Step S102, actual water temperature value is greater than or equal to preset water temperature threshold, enter step S102, otherwise return to step S101;Step S103, the running data of mining vehicle engine is obtained, and first running sequence is constructed;Step S104, first running sequence is input into water temperature correction model, and the value of output represents the target water temperature value of current time mining vehicle engine;The present application is fused by the transmission of the running data of mining vehicle engine in time, and the generation of real engine water temperature is generated under the premise of guaranteeing the normal work of engine by combining the adversarial training of neural network model, to avoid the interference of local water temperature to the driving behavior of driver.

Description

Technical Field

[0001] This invention relates to the field of electronically controlled engine technology, and more specifically, to an electronically controlled engine software water temperature correction method. Background Technology

[0002] As mining continues to deepen, mining trucks have become an indispensable tool in mining operations. As vehicles used to transport large quantities of materials such as ore and coal, mining truck engines generate a lot of heat, requiring a cooling system to ensure their normal operation.

[0003] Currently, the engine coolant temperature of mining trucks is typically converted into a coolant temperature report by a coolant temperature sensor and sent to the instrument panel, where it is displayed. However, actual operating conditions for mining trucks are quite complex, including high temperatures, high altitudes, and bumpy conditions. These complex conditions may cause localized overheating of the engine coolant, even though the actual coolant temperature does not affect the normal operation of the engine. In such cases, the coolant temperature sensor detects the localized overheating and sends an abnormal coolant temperature report to the instrument panel, which illuminates the overheating warning light. When the mining truck driver sees the overheating warning light illuminated, they may take measures such as slowing down or braking to check if there is a malfunction in the cooling system, which can seriously interfere with the driver's driving behavior. Summary of the Invention

[0004] This invention provides a water temperature correction method for electronic control software, which solves the technical problems mentioned in the background.

[0005] This invention provides a method for correcting water temperature in electronic control software, comprising the following steps:

[0006] Step S101: Obtain the actual water temperature value of the mining truck engine at the current time through the water temperature sensor;

[0007] Step S102: Determine whether the actual water temperature of the mining truck engine at the current time is greater than or equal to the preset water temperature threshold. If the actual water temperature is greater than or equal to the preset water temperature threshold, proceed to step S102; otherwise, return to step S101.

[0008] Step S103: Taking the current time as the end time point, shifting backward by a preset collection time period M as the starting time point, and acquiring the operating data of the mining truck engine according to a preset time interval m, and constructing a first operating sequence; the first operating sequence is represented as: A={a1…a N}, where N = M / m, a1…a N These represent the operating data of the mining truck engine from the 1st time point to the Nth time point, and the operating data of the mining truck engine from the start time point to the end time point, respectively.

[0009] in These represent the vehicle speed, engine speed, accelerator pedal pressure, brake pedal pressure, fuel injection quantity, engine oil temperature, actual coolant temperature, ambient temperature, load, and tire pressure of the mining truck at the Nth time point, respectively.

[0010] Step S104: Input the first running sequence into the water temperature correction model, and output the value representing the target water temperature value of the mining truck engine at the current time. The water temperature correction model includes: a first hidden layer, a logic layer, and a generator. The first hidden layer inputs the first running sequence and outputs a second running sequence. The first running sequence and the second running sequence have the same number of sequence units. The first hidden layer includes N time steps, 1≤i≤N. The i-th time step inputs the i-th sequence unit of the first running sequence and outputs a first vector at the i-th time step. The logic layer is used to extract the first vector of the N-th sequence unit of the second running sequence. The generator inputs the first vector and outputs the value representing the target water temperature value of the mining truck engine at the current time.

[0011] Furthermore, the preset water temperature threshold is a custom parameter.

[0012] Furthermore, both the preset data collection period M and the preset time interval m are custom parameters.

[0013] Furthermore, the calculation formula for the first hidden layer includes:

[0014] The intermediate vector h at the i-th time step i The calculation formula is as follows:

[0015] h i =σ(x i U1+h i-1 U2+b1), where h i-1 Let x represent the intermediate vector at time step (i-1), x represent the i-th sequence unit of the first sequence at time step (i-1), U1 and U2 represent the first and second weight parameters respectively, b1 represent the first bias parameter, and σ represent the sigmoid activation function.

[0016] The first vector at the i-th time step The calculation formula is as follows:

[0017] Where h i Let U represent the intermediate vector at the i-th time step, U3 represent the third weight parameter, b2 represent the second bias parameter, and ReLU represent the ReLU activation function.

[0018] Furthermore, during the training of the water temperature correction model, a discriminator is connected to determine whether the target water temperature value output by the generator comes from the real sample or from the generator.

[0019] Furthermore, the calculation formulas for the generator and discriminator include:

[0020] The formula for calculating the generator's loss function GLoss is as follows:

[0021] GLoss = -log(D(G(B))), where B represents the first vector, G(B) represents the target water temperature value output by the generator from the input of the first vector, and D(G(B)) represents the probability value that the target water temperature value output by the generator is input to the target water temperature value output by the discriminator and belongs to the real sample.

[0022] The loss function DLoss of the discriminator is calculated as follows:

[0023] DLoss = -log(D(C)) - log(1 - D(G(B))), where B represents the first vector, C represents the target water temperature value corresponding to the real sample, D(C) represents the probability value of the discriminator judging the target water temperature value corresponding to the real sample as belonging to the real sample, G(B) represents the target water temperature value output by the generator from the first vector, and D(G(B)) represents the probability value of the target water temperature value output by the generator being input to the discriminator as belonging to the real sample.

[0024] Furthermore, the target water temperature values ​​corresponding to the real samples in the training dataset used to train the water temperature correction model are obtained through simulation experiments, including the following steps:

[0025] Step S201: Obtain the operating data of the mining truck engine and divide the mining truck engine into regions of size K*K matrix, and obtain the actual water temperature of each region through a water temperature sensor;

[0026] Step S202: Calculate the average actual water temperature of each area as the first water temperature value;

[0027] Step S203: If the engine speed in the operating data of the mining truck engine is greater than or equal to the preset engine speed threshold K or the oil temperature in the operating data of the mining truck engine is greater than or equal to the preset oil temperature threshold J, then the maximum value of the actual water temperature in one of the regions shall be taken as the target water temperature value; otherwise, the first water temperature value shall be taken as the target water temperature value.

[0028] Furthermore, the preset engine speed threshold K and the preset oil temperature threshold J are both custom parameters.

[0029] The beneficial effects of this invention are as follows: This invention generates a real engine water temperature while ensuring the normal operation of the engine by transmitting and fusing the operating data of the mining truck engine over time, and by combining it with the adversarial training of the neural network model, thereby avoiding interference with the driver's driving behavior due to excessively high local water temperature. Attached Figure Description

[0030] Figure 1 This is a flowchart of a water temperature correction method for electronic control software according to the present invention;

[0031] Figure 2 This is a schematic diagram of the resistance value and water temperature mapping curve of the water temperature sensor of the present invention;

[0032] Figure 3 This is a schematic diagram of the preset temperature-voltage value mapping curve of the present invention;

[0033] Figure 4 This is a flowchart of the present invention for obtaining the target water temperature value corresponding to the real sample through simulation experiments. Detailed Implementation

[0034] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0035] It should be noted that, unless otherwise defined, the technical or scientific terms used in one or more embodiments of the present invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in one or more embodiments of the present invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0036] like Figures 1-4 As shown, a water temperature correction method for electronic control software includes the following steps:

[0037] Step S101: Obtain the actual water temperature value of the mining truck engine at the current time through the water temperature sensor;

[0038] Step S102: Determine whether the actual water temperature of the mining truck engine at the current time is greater than or equal to the preset water temperature threshold. If the actual water temperature is greater than or equal to the preset water temperature threshold, proceed to step S102; otherwise, return to step S101.

[0039] Step S103: Using the current time as the end time point, shift the preset collection time period M forward as the start time point, and acquire the operating data of the mining truck engine according to the preset time interval m, and construct the first operating sequence.

[0040] The first running sequence is represented as: A = {a1…a2} N}, where N = M / m, a1…a N These represent the operating data of the mining truck engine from the 1st time point to the Nth time point, and the operating data of the mining truck engine from the start time point to the end time point, respectively.

[0041] in These represent the vehicle speed, engine speed, accelerator pedal pressure, brake pedal pressure, fuel injection quantity, engine oil temperature, actual coolant temperature, ambient temperature, load, and tire pressure of the mining truck at the Nth time point, respectively.

[0042] Step S104: Input the first running sequence into the water temperature correction model, and the output value represents the target water temperature value of the mining truck engine at the current time;

[0043] The water temperature correction model includes: a first hidden layer, a logic layer, and a generator;

[0044] The first hidden layer takes a first running sequence as input and outputs a second running sequence. The number of sequence units in the first running sequence and the second running sequence are the same.

[0045] The first hidden layer comprises N time steps, 1 ≤ i ≤ N. The i-th time step inputs the i-th sequence unit of the first running sequence and outputs the first vector at the i-th time step.

[0046] The logic layer is used to extract the first vector of the Nth sequence unit of the second running sequence;

[0047] The generator takes a first vector as input and outputs a value representing the target water temperature of the mining truck engine at the current time.

[0048] In one embodiment of the present invention, the preset water temperature threshold is a custom parameter. The preset water temperature threshold should be lower than the upper limit of the normal operating temperature of the mining truck engine. For example, the preset water temperature threshold is set to be 50 degrees Celsius lower than the upper limit of the normal operating temperature of the mining truck engine.

[0049] In one embodiment of the present invention, the preset collection time period M and the preset time interval m are both custom parameters. Preferably, the preset collection time period M is set to 30 seconds and the preset time interval m is set to 1 second.

[0050] In one embodiment of the present invention, the water temperature sensor is composed of an NTC (negative temperature coefficient) thermistor, such as... Figure 2 As shown, the lower the engine coolant temperature of the mining truck, the higher the resistance value of the coolant temperature sensor; conversely, the higher the engine coolant temperature, the lower the resistance value. The coolant temperature sensor converts this resistance value into a corresponding voltage value and sends it to the ECU (Electronic Control Unit). Figure 3 As shown, the ECU obtains the actual water temperature of the mining truck engine based on the voltage value and the preset temperature-voltage value mapping curve, and sends the actual water temperature to the instrument panel for display.

[0051] It should be noted that the preset temperature-voltage value mapping curve was obtained by the manufacturer of the mining truck engine through multiple simulation experiments. For example, different water temperature values ​​were obtained by using a thermometer, and the corresponding voltage values ​​of the water temperature sensor were recorded. The above steps were repeated to obtain multiple data points and draw a scatter plot. Each data point corresponds to a water temperature value and a voltage value. Finally, the envelope was fitted using the least squares method as the preset temperature-voltage value mapping curve.

[0052] In one embodiment of the present invention, the calculation formula for the first hidden layer includes:

[0053] The intermediate vector h at the i-th time step i The calculation formula is as follows:

[0054] h i =σ(x i U1+h i-1 U2+b1), where h i-1 Let x represent the intermediate vector at time step i-1. i This represents the i-th sequence unit of the first sequence at the i-th time step of the input; U1 and U2 represent the first and second weight parameters, respectively; b1 represents the first bias parameter; σ represents the sigmoid activation function.

[0055] The first vector at the i-th time step The calculation formula is as follows:

[0056] Where h iLet U represent the intermediate vector at the i-th time step, U3 represent the third weight parameter, b2 represent the second bias parameter, and ReLU represent the ReLU activation function.

[0057] In one embodiment of the present invention, during the training process of the water temperature correction model, a discriminator is connected to determine whether the target water temperature value output by the generator comes from the real sample or from the generator.

[0058] In one embodiment of the present invention, the calculation formulas for the generator and the discriminator include:

[0059] The formula for calculating the generator's loss function GLoss is as follows:

[0060] GLoss = -log(D(G(B))), where B represents the first vector, G(B) represents the target water temperature value output by the generator from the input of the first vector, and D(G(B)) represents the probability value that the target water temperature value output by the generator is input to the target water temperature value output by the discriminator and belongs to the real sample.

[0061] The loss function DLoss of the discriminator is calculated as follows:

[0062] DLoss = -log(D(C)) - log(1 - D(G(B))), where B represents the first vector, C represents the target water temperature value corresponding to the real sample, D(C) represents the probability value of the discriminator judging the target water temperature value corresponding to the real sample as belonging to the real sample, G(B) represents the target water temperature value output by the generator from the first vector, and D(G(B)) represents the probability value of the target water temperature value output by the generator being input to the discriminator as belonging to the real sample.

[0063] In one embodiment of the present invention, the first hidden layer may also be constructed based on a long short-term memory recurrent neural network model, which will not be elaborated here.

[0064] It should be noted that adversarial training between the generator and the discriminator can force the generator to generate target water temperature values ​​that are closer to those of the real samples.

[0065] In one embodiment of the present invention, such as Figure 4 As shown, the target water temperature values ​​corresponding to the real samples in the training dataset used to train the water temperature correction model are obtained through simulation experiments, including the following steps:

[0066] Step S201: Obtain the operating data of the mining truck engine and divide the mining truck engine into regions of size K*K matrix, and obtain the actual water temperature of each region through a water temperature sensor;

[0067] Step S202: Calculate the average actual water temperature of each area as the first water temperature value;

[0068] Step S203: If the engine speed in the operating data of the mining truck engine is greater than or equal to the preset engine speed threshold K or the oil temperature in the operating data of the mining truck engine is greater than or equal to the preset oil temperature threshold J, then the maximum value of the actual water temperature in one of the regions shall be taken as the target water temperature value; otherwise, the first water temperature value shall be taken as the target water temperature value.

[0069] In one embodiment of the present invention, the preset engine speed threshold K and the preset oil temperature threshold J are both custom parameters. Preferably, the default value of K is the upper limit of the engine speed for normal operation, and the default value of J is the upper limit of the oil temperature for normal operation.

[0070] The embodiments of this example have been described above. However, this example is not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of this example, and all of them are within the protection scope of this example.

Claims

1. A method for correcting water temperature using electronic control software, characterized in that, Includes the following steps: Step S101: Obtain the actual water temperature value of the mining truck engine at the current time through the water temperature sensor; Step S102: Determine whether the actual water temperature of the mining truck engine at the current time is greater than or equal to the preset water temperature threshold. If the actual water temperature is greater than or equal to the preset water temperature threshold, proceed to step S103; otherwise, return to step S101. Step S103: Taking the current time as the end time point, shifting backward by a preset collection time period M as the starting time point, and acquiring the operating data of the mining truck engine according to a preset time interval m, and constructing a first operating sequence; the first operating sequence is represented as: Where N = M / m, These represent the operating data of the mining truck engine from the 1st time point to the Nth time point, and the operating data of the mining truck engine from the start time point to the end time point, respectively. ,in These represent the vehicle speed, engine speed, accelerator pedal pressure, brake pedal pressure, fuel injection quantity, engine oil temperature, actual coolant temperature, ambient temperature, load, and tire pressure of the mining truck at the Nth time point, respectively. Step S104: Input the first running sequence into the water temperature correction model, and the output value represents the target water temperature value of the mining truck engine at the current time. The water temperature correction model includes: a first hidden layer, a logic layer, and a generator. The first hidden layer takes a first running sequence as input and outputs a second running sequence. The first and second running sequences have the same number of sequence units. The first hidden layer includes N time steps, 1≤i≤N. The i-th time step takes the i-th sequence unit of the first running sequence as input and outputs a first vector at the i-th time step. The logic layer is used to extract the first vector from the N-th sequence unit of the second running sequence. The generator takes the first vector as input and outputs a value representing the target water temperature of the mining truck engine at the current time. During the training of the water temperature correction model, a discriminator is connected to determine whether the target water temperature value output by the generator comes from the real sample or from the generator. The calculation formulas for the generator and discriminator include: Loss function of generator The calculation formula is as follows: Where B represents the first vector, This represents the target water temperature value output by the generator from the input first vector. This represents the probability that the target water temperature value output by the generator, when input into the target water temperature value output by the discriminator, belongs to a real sample. loss function of discriminator The calculation formula is as follows: Where B represents the first vector, and C represents the target water temperature value corresponding to the actual sample. This represents the probability that the discriminator determines the target water temperature value corresponding to the real sample to be a real sample. This represents the target water temperature value output by the generator from the input first vector. This represents the probability that the target water temperature value output by the generator, when input into the target water temperature value output by the discriminator, belongs to a real sample. The target water temperature values ​​corresponding to the real samples in the training dataset used to train the water temperature correction model are obtained through simulation experiments, including the following steps: Step S201: Obtain the operating data of the mining truck engine and divide the mining truck engine into regions of size L*L matrix, and obtain the actual water temperature of each region through a water temperature sensor; Step S202: Calculate the average actual water temperature of each area as the first water temperature value; Step S203: If the engine speed in the operating data of the mining truck engine is greater than or equal to the preset engine speed threshold K or the oil temperature in the operating data of the mining truck engine is greater than or equal to the preset oil temperature threshold J, then the maximum value of the actual water temperature in one of the regions shall be taken as the target water temperature value; otherwise, the first water temperature value shall be taken as the target water temperature value.

2. The method for correcting water temperature in electronic control software according to claim 1, characterized in that, The preset water temperature threshold is a custom parameter.

3. The method for correcting water temperature in electronic control software according to claim 1, characterized in that, Both the preset data collection period M and the preset time interval m are custom parameters.

4. The method for correcting water temperature in electronic control software according to claim 1, characterized in that, The calculation formula for the first hidden layer includes: The intermediate vector at the i-th time step The calculation formula is as follows: ,in Let represent the intermediate vector at the (i-1)th time step. This represents the i-th sequence unit of the first sequence at the i-th time step. and These represent the first weight parameter and the second weight parameter, respectively. Indicates the first bias parameter; Represents the sigmoid activation function The first vector at the i-th time step The calculation formula is as follows: ,in Let represent the intermediate vector at the i-th time step. This represents the third weighting parameter. This represents the second bias parameter, and ReLU represents the ReLU activation function.

5. The method for correcting water temperature in electronic control software according to claim 1, characterized in that, Both the preset engine speed threshold K and the preset oil temperature threshold J are custom parameters.

Images