Diesel engine performance prediction model training method and device, equipment, storage medium and product

Abstract

The invention discloses a diesel engine performance prediction model training method and device, equipment, a storage medium and a product. Comprising the steps of constructing an initial sample data set; the initial sample data set comprises diesel engine working condition parameter samples and diesel engine standard performance parameters corresponding to the diesel engine working condition parameter samples; performing model training on a pre-constructed network model by adopting the initial sample data set to obtain an initial diesel engine performance prediction model, and determining a model prediction error value of the initial diesel engine performance prediction model; if the model prediction error value is greater than a preset error threshold value, generating a target sample data set based on the diesel engine working condition parameter sample and a corresponding diesel engine standard performance parameter; and performing model training on the initial diesel engine performance prediction model by adopting the target sample data set until a preset model training ending condition is met, thereby obtaining a target diesel engine performance prediction model. According to the method, the performance calibration efficiency is greatly improved by constructing the high-precision diesel engine performance prediction model, and meanwhile the test cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of diesel engine test and experiment, and in particular to a diesel engine performance prediction model training method, device, equipment, storage medium and product. Background Art

[0002] Traditional diesel engines are widely used in many industries such as transportation, agriculture, and national defense due to their high thermal efficiency and reliable performance. In order to achieve the goal of energy conservation and emission reduction, it is necessary to carry out simultaneous optimization of multiple parameters of diesel engines and achieve simultaneous improvement of multiple performances through the coordination of multiple parameters.

[0003] However, different parameters have a strong nonlinear coupling relationship with various diesel engine performances, which makes optimization extremely difficult, and a large number of parameters will lead to an exponential increase in calibration scanning examples. Therefore, if the existing diesel engine calibration method is used to determine the best performance optimization parameters of the diesel engine, a large number of examples need to be tested in bench tests, which greatly increases the time and cost of performance optimization. Summary of the invention

[0004] The present invention provides a diesel engine performance prediction model training method, device, equipment, storage medium and product method, device, equipment and storage medium to improve the efficiency and accuracy of diesel engine optimal performance prediction while reducing the prediction test cost.

[0005] According to one aspect of the present invention, a diesel engine performance prediction model training method is provided, the method comprising:

[0006] Constructing an initial sample data set; the initial sample data set includes a diesel engine operating parameter sample and a diesel engine standard performance parameter corresponding to the diesel engine operating parameter sample;

[0007] Using the initial sample data set to perform model training on a pre-built network model to obtain an initial diesel engine performance prediction model, and determining a model prediction error value of the initial diesel engine performance prediction model;

[0008] If the model prediction error value is greater than a preset error threshold, a target sample data set is generated based on the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter;

[0009] The target sample data set is used to perform model training on the initial diesel engine performance prediction model until a preset model training end condition is met, thereby obtaining a target diesel engine performance prediction model.

[0010] According to another aspect of the present invention, a method for optimizing diesel engine performance is provided, the method comprising:

[0011] Acquire initial diesel engine operating parameters; the initial diesel engine operating parameters correspond to its own diesel engine performance parameters;

[0012] Generating target diesel engine operating parameters according to the initial diesel engine operating parameters and its own diesel engine performance parameters;

[0013] Using a target diesel engine performance prediction model to perform performance prediction on the target diesel engine operating parameters to obtain target performance parameters corresponding to the target diesel engine operating parameters;

[0014] Determining optimal diesel performance parameters and their corresponding optimal operating parameters according to target performance parameters of the target diesel engine operating parameters;

[0015] Wherein, the target diesel engine performance prediction model is generated by using the above diesel engine performance prediction model training method.

[0016] According to another aspect of the present invention, a diesel engine performance prediction model training device is provided, the device comprising:

[0017] A data set construction module is used to construct an initial sample data set; the initial sample data set includes a diesel engine operating parameter sample and a diesel engine standard performance parameter corresponding to the diesel engine operating parameter sample;

[0018] An initial model training module is used to perform model training on a pre-built network model using the initial sample data set to obtain an initial diesel engine performance prediction model, and determine a model prediction error value of the initial diesel engine performance prediction model;

[0019] A target data set generation module, configured to generate a target sample data set based on the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter if the model prediction error value is greater than a preset error threshold;

[0020] The target model training module is used to use the target sample data set to perform model training on the initial diesel engine performance prediction model until a preset model training end condition is met to obtain a target diesel engine performance prediction model.

[0021] According to another aspect of the present invention, a diesel engine performance optimization device is provided, the device comprising:

[0022] A parameter acquisition module, used to acquire initial diesel engine operating parameters; the initial diesel engine operating parameters correspond to its own diesel engine performance parameters;

[0023] A parameter generation module, used for generating target diesel engine operating parameters according to the initial diesel engine operating parameters and its own diesel engine performance parameters;

[0024] A performance prediction module, used to perform performance prediction on the target diesel engine operating parameters using a target diesel engine performance prediction model, and obtain target performance parameters corresponding to the target diesel engine operating parameters;

[0025] An optimal parameter determination module, used to determine the optimal diesel performance parameter and its corresponding optimal operating condition parameter according to the target performance parameter of the target diesel engine operating condition parameter;

[0026] Wherein, the target diesel engine performance prediction model is generated by using the above diesel engine performance prediction model training method.

[0027] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0028] at least one processor; and

[0029] a memory communicatively connected to the at least one processor; wherein,

[0030] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the diesel engine performance prediction model training and / or diesel engine performance optimization method described in any embodiment of the present invention.

[0031] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the diesel engine performance prediction model training and / or diesel engine performance optimization method described in any embodiment of the present invention when executed.

[0032] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the steps in the above-mentioned diesel engine performance prediction model training and / or diesel engine performance optimization method are implemented.

[0033] The technical solution of the embodiment of the present invention constructs an initial sample data set, uses the initial sample data set to perform model training on the pre-constructed network model, obtains an initial diesel engine performance prediction model, and determines the model prediction error value of the initial diesel engine performance prediction model; if the model prediction error value is greater than the preset error threshold, a target sample data set is generated based on the diesel engine operating parameter sample and its corresponding diesel engine standard performance parameter; the target sample data set is used to perform model training on the initial diesel engine performance prediction model until the preset model training end condition is met, and the target diesel engine performance prediction model is obtained. The above technical solution generates a sample data set again to perform secondary training on the model after the initial model training, and performs performance calibration on the high-precision diesel engine performance prediction model obtained through training, which greatly improves the efficiency of performance calibration and reduces the test cost. .

[0034] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 is a flow chart of a diesel engine performance prediction model training method provided according to Embodiment 1 of the present invention;

[0037] Figure 2 is a flow chart of a diesel engine performance prediction model training method provided according to Embodiment 2 of the present invention;

[0038] Figure 3 is a flow chart of a diesel engine performance optimization method provided according to Embodiment 3 of the present invention;

[0039] Figure 4 is a flow chart of a diesel engine performance optimization method provided according to a fourth embodiment of the present invention;

[0040] Figure 5 2 is a schematic diagram of the structure of a diesel engine performance prediction model training device provided according to Embodiment 5 of the present invention;

[0041] Figure 6 is a structural schematic diagram of a diesel engine performance optimization device provided according to Embodiment 6 of the present invention;

[0042] Figure 7 It is a structural schematic diagram of an electronic device for implementing the diesel engine performance prediction model training and / or diesel engine performance optimization method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0044] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0045] Embodiment 1

[0046] Figure 1 This is a flow chart of a diesel engine performance prediction model training method provided in the first embodiment of the present invention. This embodiment is applicable to the case of model training of a diesel engine performance prediction model used for diesel engine performance prediction. The method can be executed by a diesel engine performance prediction model training device. The diesel engine performance prediction model training device can be implemented in the form of hardware and / or software. The diesel engine performance prediction model training device can be configured in an electronic device. Figure 1 As shown, the method includes:

[0047] S110, constructing an initial sample data set; the initial sample data set includes a diesel engine operating parameter sample and a diesel engine standard performance parameter corresponding to the diesel engine operating parameter sample.

[0048] The diesel engine operating parameters may include diesel engine parameters under certain operating conditions. For example, the diesel engine operating parameters may include at least one of two injection timings, injection distribution ratio, injection pressure, intake pressure and intake temperature of the diesel engine.

[0049] The standard performance parameters of the diesel engine corresponding to the diesel engine operating parameters may include the diesel engine's fuel consumption rate, nitrogen oxide emissions, and soot emissions. For the diesel engine operating parameters under any operating condition, the diesel engine cylinder pressure, nitrogen oxide emissions, and soot emissions under that condition can be obtained based on bench tests. The fuel consumption rate can be determined based on the diesel engine cylinder pressure.

[0050] The number of samples included in the initial sample data set can be pre-set by relevant technicians according to actual needs. For example, the number of samples can be 40. It should be noted that for any diesel engine operating parameter, the corresponding parameter value range is different. For example, the parameter value range of the injection distribution ratio is [x1, y1], and the parameter value range of the injection pressure is [x2, y2]. In the process of randomly generating sample data, it is necessary to randomly generate sample data according to the parameter value ranges of different diesel engine operating parameters.

[0051] Exemplarily, the initial sample data set is generated by randomly generating the values ​​of each diesel engine operating parameter within the parameter value range corresponding to the corresponding diesel engine operating parameter, thereby obtaining a preset number of diesel engine operating parameter samples. For example, if the preset number of samples is 40, 40 groups of diesel engine operating parameter samples can be generated. For any group of diesel engine operating parameter samples, a bench test is performed based on the parameter values ​​corresponding to the two injection times, injection distribution ratio, injection pressure, intake pressure and intake temperature in the sample, and the standard performance parameters of the diesel engine corresponding to the group of diesel engine operating parameter samples are obtained, specifically including fuel consumption rate, nitrogen oxide emissions and soot emissions.

[0052] Optionally, the diesel engine operating condition parameter samples may be generated based on a Sobel sample generation method and based on parameter value ranges of different diesel engine operating condition parameters.

[0053] S120, using the initial sample data set to perform model training on the pre-built network model to obtain an initial diesel engine performance prediction model, and determining a model prediction error value of the initial diesel engine performance prediction model.

[0054] The network model may be preset by relevant technical personnel. For example, the network model may be a Gaussian Process Regression (GPR) model.

[0055] Exemplarily, the initial sample data set is input into a pre-built Gaussian process regression model for model training to obtain the prediction results of the model output; based on the diesel engine standard performance parameters of the diesel engine operating parameter samples in the initial sample data set and the prediction results of the model output, the model prediction error value of the initial diesel engine performance prediction model can be determined.

[0056] In order to further improve the model training accuracy of the initial diesel engine performance prediction model, the initial sample data set can be divided into a training set, a validation set, and a test set, and the model training, validation, and testing can be performed based on the corresponding sample data sets.

[0057] In an optional embodiment, the pre-built network model is trained using an initial sample data set to obtain an initial diesel engine performance prediction model, and a model prediction error value of the initial diesel engine performance prediction model is determined, including:

[0058] Step a1: divide the initial sample data set into a sample training set and a sample test set based on a preset division ratio.

[0059] The preset division ratio can be preset by relevant technical personnel. For example, the initial sample data set can be divided into a sample training set, a sample verification set, and a sample test set at a division ratio of 70%, 15%, and 15%. Alternatively, the processed sample data set can be divided into a sample training set and a sample test set only at a division ratio of 80% and 20%.

[0060] Step a2: input the sample training set into the pre-built network model for model training to obtain an initial diesel engine performance prediction model.

[0061] Specifically, the sample training set is input into the pre-built network model to obtain the predicted performance parameters of the model output; the network model is trained according to the standard performance parameters and predicted performance parameters of the diesel engine until a pre-set iteration number threshold is reached, and the training is stopped to obtain the initial diesel engine performance prediction model.

[0062] Step a3: input the sample test set into the initial diesel engine performance prediction model to obtain the diesel engine prediction performance parameters output by the model.

[0063] Step a4: determining a model prediction error value of an initial diesel engine performance prediction model according to the diesel engine standard performance parameters and the diesel engine predicted performance parameters of the diesel engine operating parameter samples in the sample test set.

[0064] According to the diesel engine standard performance parameters and the diesel engine predicted performance parameters of the diesel engine operating parameter samples in the sample test set, the error value between the two is determined, so as to obtain the sample error values ​​corresponding to each diesel engine operating parameter sample; the average value of the sample error values ​​of each diesel engine operating parameter sample is determined as the model prediction error value of the initial diesel engine performance prediction model.

[0065] S130: If the model prediction error value is greater than a preset error threshold, a target sample data set is generated based on the diesel engine operating parameter samples and their corresponding diesel engine standard performance parameters.

[0066] The preset error threshold may be preset according to actual needs. For example, the preset error value may be set to 10%.

[0067] For example, if the model prediction error value is greater than the preset error threshold, it means that the model accuracy of the initial diesel engine performance prediction model is not enough and cannot be directly used for diesel engine performance prediction. The initial diesel engine performance prediction model needs to be trained again. Therefore, it is necessary to further accumulate the sample size to increase the sample size. Specifically, the target sample data set can be generated based on the diesel engine oil operating parameter samples, the corresponding diesel engine standard performance parameters, and the parameter value intervals corresponding to different diesel engine operating parameter samples.

[0068] If the model prediction error value is not greater than the preset error threshold, it means that the model accuracy of the initial diesel engine performance prediction model is high and can be directly used for diesel engine performance prediction.

[0069] S140, using the target sample data set to perform model training on the initial diesel engine performance prediction model until a preset model training end condition is met, thereby obtaining a target diesel engine performance prediction model.

[0070] Among them, the end condition of model training can be that the number of model iteration training times reaches a preset number threshold, or the loss value tends to be stable or reaches a set loss value threshold, which is not limited in this embodiment.

[0071] Optionally, the target sample data set may be used to train the initial diesel engine performance prediction model, or the target sample data set and the initial sample data set may be used together to train the initial diesel engine performance prediction model, which is not limited in this embodiment.

[0072] The technical solution of the embodiment of the present invention constructs an initial sample data set, uses the initial sample data set to perform model training on the pre-constructed network model, obtains an initial diesel engine performance prediction model, and determines the model prediction error value of the initial diesel engine performance prediction model; if the model prediction error value is greater than the preset error threshold, a target sample data set is generated based on the diesel engine operating parameter sample and its corresponding diesel engine standard performance parameter; the target sample data set is used to perform model training on the initial diesel engine performance prediction model until the preset model training end condition is met, and the target diesel engine performance prediction model is obtained. The above technical solution generates a sample data set again to perform secondary training on the model after the initial model training, and performs performance calibration on the high-precision diesel engine performance prediction model obtained through training, which greatly improves the efficiency of performance calibration and reduces the test cost. .

[0073] Embodiment 2

[0074] Figure 2 This is a flow chart of a diesel engine performance prediction model training method provided in the second embodiment of the present invention. This embodiment is based on the above embodiment and is optimized and improved.

[0075] Furthermore, the step of "generating a target sample data set based on the diesel engine operating parameter samples and their corresponding diesel engine standard performance parameters" is refined into "generating a first sample data set based on the sample number of the diesel engine operating parameter samples and a preset error threshold; and generating a second sample data set based on the diesel engine operating parameter samples and their corresponding diesel engine standard performance parameters; generating a target sample data set based on the first sample data set and the second sample data set." to improve the method of generating the target sample data set.

[0076] It should be noted that for the parts not described in detail in the embodiments of the present invention, reference can be made to the descriptions of other embodiments. Figure 2 As shown, the method comprises the following specific steps:

[0077] S210, constructing an initial sample data set; the initial sample data set includes a diesel engine operating parameter sample and a diesel engine standard performance parameter corresponding to the diesel engine operating parameter sample.

[0078] S220, using the initial sample data set to perform model training on the pre-built network model to obtain an initial diesel engine performance prediction model, and determining a model prediction error value of the initial diesel engine performance prediction model.

[0079] S230: If the model prediction error value is greater than a preset error threshold, a first sample data set is generated according to the sample quantity of the diesel engine operating parameter samples and based on the preset error threshold.

[0080] Exemplarily, based on the parameter value range corresponding to the diesel engine operating parameter, a first sample data set having the same number of samples as the diesel engine operating parameter samples can be randomly generated.

[0081] It is understandable that in order to further improve the accuracy of the generated first sample data set, the first sample data set may be generated in a targeted manner.

[0082] In an optional embodiment, according to the sample quantity of the diesel engine operating parameter samples and based on a preset error threshold, a first sample data set is generated, including:

[0083] Step b1: input the diesel engine operating parameter sample into the initial diesel engine performance prediction model to obtain the diesel engine predicted performance parameters output by the model.

[0084] Step b2: determining the sample error value of the diesel engine operating parameter sample according to the diesel engine standard performance parameters and the diesel engine predicted performance parameters of the diesel engine operating parameter sample.

[0085] The deviation between the standard performance parameters of the diesel engine and the predicted performance parameters of the diesel engine is taken as the sample error value of the corresponding diesel engine operating condition parameter sample.

[0086] Step b3: Determine a reference operating parameter sample according to the sample error value of the diesel engine operating parameter sample and based on a preset error threshold.

[0087] Specifically, the diesel engine operating parameter samples whose sample error values ​​are greater than a preset error threshold are used as reference operating parameter samples. The preset error threshold can be preset by relevant technicians, for example, can be set to 10%.

[0088] Step b3: Generate a first sample data set with the same number of samples as the diesel engine operating parameter samples according to the parameter value interval range of the reference operating parameter samples.

[0089] Exemplarily, if the number of samples of the diesel engine operating condition parameter is 20, 20 first parameter samples are randomly generated based on the reference value range as samples in the first sample data set.

[0090] S240: Generate a second sample data set according to the diesel engine operating parameter sample and its corresponding diesel engine standard performance parameter.

[0091] In an optional embodiment, a second sample data set is generated according to the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter, including:

[0092] Step c1: determining a dominating relationship or a non-dominating relationship between the diesel engine operating parameter samples according to the diesel engine standard performance parameters of the diesel engine operating parameter samples.

[0093] Specifically, the diesel engine operating parameter samples are taken as individuals, and the diesel engine standard performance parameters of the diesel engine operating parameter samples are taken as individual target values. In this embodiment, the target values ​​are fuel consumption rate, nitrogen oxide emissions and soot emissions, respectively.

[0094] Compare the target values ​​of any two individuals Mi and Mj, and take the one with the lower target value as an example. In a three-dimensional space, if the three target values ​​of individual Mi are all lower than the three target values ​​of Mj, then Mi is considered to dominate Mj. If at least one of the three target values ​​of Mi is lower than Mj, and the other target values ​​are higher than Mj, then Mi and Mi are considered to be in a non-dominant relationship.

[0095] Step c2: determining candidate operating condition parameter samples according to the dominating relationship and non-dominating relationship between the diesel engine operating condition parameter samples.

[0096] Specifically, according to the dominance relationship and non-dominance relationship between individuals, the number of each diesel engine operating condition parameter sample individual dominated by other individuals is determined, and the diesel engine operating condition parameter sample individuals are sorted from small to large according to the number of dominance. Among them, the individual with a dominance number of 0 has a sequence value of 1, and then the sequence value increases with the dominance number. The sample corresponding to the individual with a sequence value of 1 is determined as the candidate operating condition parameter sample.

[0097] It should be noted that the candidate operating condition parameter sample set with a sequence value of 1 can be called a Pareto surface set, which is not dominated by any individual and is regarded as the current best individual set.

[0098] Step c3: determining the crowding distance value of the candidate operating condition parameter sample according to the diesel engine standard performance parameter of the candidate operating condition parameter sample.

[0099] Among them, the crowding distance value of the candidate operating condition parameter samples can be used to characterize the degree of crowding between samples. In order to ensure the diversity between samples, samples with larger crowding distance values ​​can be selected preferentially.

[0100] In an optional embodiment, the crowding distance value of the candidate operating parameter sample is determined based on the diesel engine standard performance parameters of the candidate operating parameter sample, including: determining adjacent operating parameter samples that have an adjacency relationship with the candidate operating parameter sample; determining a target rectangular area based on the parameter values ​​of the diesel engine standard performance parameters of the adjacent operating parameter samples; and determining the crowding distance value of the candidate operating parameter sample based on the area side length of the target rectangular area.

[0101] Specifically, in a three-dimensional space, the three diesel engine standard performance parameters of the candidate operating parameter samples are used as the three-dimensional space coordinate system to generate the coordinate points of the candidate operating parameter samples in the three-dimensional space. For example, the coordinate point of the candidate operating parameter sample A in the three-dimensional space is (xa, ya, za), where xa is the fuel consumption rate in the diesel engine standard performance parameters, ya is the nitrogen oxide emission in the diesel engine standard performance parameters, and za is the carbon soot emission in the diesel engine standard performance parameters. According to the diesel engine standard performance parameters of the candidate operating parameter samples, the position coordinates of the candidate operating parameter samples in the three-dimensional space are generated. According to the position coordinates of each candidate operating parameter sample in the three-dimensional space, the adjacent operating parameter samples of any candidate operating parameter sample are determined. Among them, for any candidate operating parameter sample, its adjacent operating parameter samples can be operating parameter samples with a close distance to it.

[0102] Specifically, for any candidate operating condition parameter sample, the sample distance between the candidate operating condition parameter sample and other candidate operating condition parameter samples is determined according to the position coordinates of each candidate operating condition parameter sample in the three-dimensional space, and two other candidate operating condition parameter samples with smaller distance values ​​are selected from each sample distance as adjacent operating condition parameter samples of the candidate operating condition parameter sample.

[0103] According to the parameter values ​​of the diesel engine standard performance parameters of the adjacent operating parameter samples, the target rectangular area is determined. Among them, the rectangular area can be a rectangular area surrounded by two adjacent operating parameter samples. Continuing with the above example, the parameter values ​​of the candidate operating parameter sample A are xa, ya and za. In three-dimensional space, according to two coordinate points, a uniquely determined rectangular area can be constructed, namely the target rectangular area. According to the parameter value, the rectangular side length of the target rectangular area can be determined. The average value of the different rectangular side lengths is determined as the crowding distance value of the candidate operating parameter sample. For example, if the side lengths of the target rectangular area are side length m1, side length m1, side length m2 and side length m2, respectively, and the four side lengths are the same in pairs; then the crowding distance value is (m1+m2) / 2.

[0104] Step c4: Determine the guiding operating condition parameter sample according to the crowding distance value of the candidate operating condition parameter sample.

[0105] To ensure sample diversity, samples with larger crowding distance values ​​are usually selected to facilitate subsequent optimization search. Specifically, according to the crowding distance values ​​corresponding to the candidate operating condition parameter samples, a certain number of candidate operating condition parameter samples with larger crowding distance values ​​are selected as guiding operating condition parameter samples.

[0106] Step c5: generating second operating condition parameter samples equal in number to the guiding operating condition parameter samples based on the guiding operating condition parameter samples according to the other operating condition parameter samples in the diesel engine operating condition parameter samples except the guiding operating condition parameter samples.

[0107] A sample is randomly selected from the guiding operating condition parameter samples as the global optimal solution. The particle swarm optimization algorithm (PSO) is used to generate the second operating condition parameter samples with the same number of samples as the guiding operating condition parameter samples. Specifically, the sample size and evolution direction of the second operating condition parameter samples are obtained by multiplying the position coordinate difference between the other operating condition parameter samples and the randomly selected global optimal sample by a preset fixed coefficient.

[0108] Step c6: Determine the standard performance parameters of the diesel engine of the second operating condition parameter sample based on the bench test.

[0109] A bench test is performed on the second operating condition parameter sample to obtain standard performance parameters of the diesel engine of the second operating condition parameter sample.

[0110] Step c7: Generate a second sample data set including a second operating condition parameter sample and its corresponding diesel engine standard performance parameter.

[0111] S250: Generate a target sample data set according to the first sample data set and the second sample data set.

[0112] S260: Use the target sample data set to perform model training on the initial diesel engine performance prediction model until a preset model training end condition is met to obtain a target diesel engine performance prediction model.

[0113] The technical solution of this embodiment generates a first sample data set based on the sample number of the diesel engine operating parameter samples and a preset error threshold, and generates a second sample data set based on the diesel engine operating parameter samples and their corresponding diesel engine standard performance parameters, and generates a target sample data set based on the first sample data set and the second sample data set, thereby achieving accurate and rapid supplementation and calibration of samples, thereby facilitating the subsequent rapid positioning of the optimal engine performance and parameter matching for a certain operating point.

[0114] Embodiment 3

[0115] Figure 3 This is a flow chart of a diesel engine performance prediction model training method provided by the third embodiment of the present invention. This embodiment is applicable to the case of diesel engine performance prediction and optimization. The method can be executed by a diesel engine performance optimization device. The diesel engine performance optimization device can be implemented in the form of hardware and / or software. The diesel engine performance optimization device can be configured in an electronic device. Figure 3 As shown, the method includes:

[0116] S310, obtaining initial diesel engine operating parameters; the initial diesel engine operating parameters correspond to its own diesel engine performance parameters.

[0117] The initial diesel engine operating condition parameters may be diesel engine operating condition parameters pre-generated based on their own value ranges, and may specifically include at least one of the two injection times, injection distribution ratio, injection pressure, intake pressure, and intake temperature of the diesel engine.

[0118] The diesel engine performance parameters corresponding to the initial diesel engine operating parameters may include fuel consumption rate, nitrogen oxide emission and soot emission, which can be predicted by the target diesel engine performance prediction model described in the above embodiment or generated based on bench tests.

[0119] S320: Generate target diesel engine operating parameters according to the initial diesel engine operating parameters and its own diesel engine performance parameters.

[0120] In an optional embodiment, the target diesel engine operating parameters are generated according to the initial diesel engine operating parameters and its own diesel engine performance parameters, including:

[0121] Step d1, determining the dominating relationship or non-dominating relationship between the initial diesel engine operating condition parameters according to the diesel engine performance parameters of the initial diesel engine operating condition parameters themselves.

[0122] Specifically, the initial diesel engine operating parameters are taken as individuals, and the diesel engine standard performance parameters of the initial diesel engine operating parameters themselves are taken as individual target values. In this embodiment, the target values ​​are fuel consumption rate, nitrogen oxide emissions and soot emissions, respectively.

[0123] Compare the target values ​​of any two individuals Mi and Mj, and take the one with the lower target value as an example. In a three-dimensional space, if the three target values ​​of individual Mi are all lower than the three target values ​​of Mj, then Mi is considered to dominate Mj. If at least one of the three target values ​​of Mi is lower than Mj, and the other target values ​​are higher than Mj, then Mi and Mi are considered to be in a non-dominant relationship.

[0124] Step d2: Determine candidate diesel engine operating parameters based on the dominating relationship and the non-dominating relationship between the initial diesel engine operating parameters.

[0125] Specifically, according to the dominance relationship and non-dominance relationship between individuals, the number of individuals dominance of each initial diesel engine operating condition parameter individual by other individuals is determined, and the initial diesel engine operating condition parameter individuals are sorted from small to large according to the number of dominance. Among them, the individual with a dominance number of 0 has a sequence value of 1, and then the sequence value increases in sequence with the dominance number. The operating condition parameter corresponding to the individual with a sequence value of 1 is determined as the candidate diesel engine operating condition parameter.

[0126] It should be noted that the candidate diesel engine operating condition parameter set with a sequence value of 1 can be called a Pareto surface set, which is not dominated by any individual and is regarded as the current best individual set.

[0127] Step d3: determining the crowding distance value of the candidate diesel engine operating parameter according to the diesel engine performance parameter of the candidate diesel engine operating parameter.

[0128] Among them, the crowding distance values ​​of the candidate diesel engine operating condition parameters can be used to characterize the degree of crowding between individual parameters. In order to ensure the diversity between individual parameters, parameters with larger crowding distance values ​​can be preferentially selected.

[0129] In an optional embodiment, the crowding distance value of the candidate diesel engine operating condition parameter is determined based on the diesel engine performance parameters of the candidate diesel engine operating condition parameter, including: determining adjacent operating condition parameters that have an adjacency relationship with the candidate diesel engine operating condition parameter; determining a target rectangular area based on the parameter values ​​of the diesel engine performance parameters of the adjacent operating condition parameters; and determining the crowding distance value of the candidate diesel engine operating condition parameter based on the area side length of the target rectangular area.

[0130] Specifically, in a three-dimensional space, the three diesel engine performance parameters of the candidate diesel engine operating parameters are used as the three-dimensional space coordinate system to generate the coordinate points of the candidate diesel engine operating parameters in the three-dimensional space. For example, the coordinate point of the candidate diesel engine operating parameter B in the three-dimensional space is (xb, yb, zb), where xb is the fuel consumption rate in the diesel engine performance parameters, yb is the nitrogen oxide emission in the diesel engine performance parameters, and zb is the carbon smoke emission in the diesel engine performance parameters. According to the diesel engine performance parameters of the candidate diesel engine operating parameters, the position coordinates of the candidate diesel engine operating parameters in the three-dimensional space are generated. According to the position coordinates of each candidate diesel engine operating parameter in the three-dimensional space, the adjacent operating parameters of any candidate diesel engine operating parameter are determined. Among them, for any candidate diesel engine operating parameter, its adjacent operating parameters can be operating parameters with a close distance to it.

[0131] Specifically, for any candidate diesel engine operating condition parameter, the sample distance between the candidate diesel engine operating condition parameter and other candidate diesel engine operating condition parameters is determined according to the position coordinates of each candidate diesel engine operating condition parameter in the three-dimensional space, and two other candidate diesel engine operating condition parameters with smaller distance values ​​are selected from each sample distance as adjacent operating condition parameters of the candidate diesel engine operating condition parameter.

[0132] According to the parameter values ​​of the diesel engine performance parameters of the adjacent operating parameters, the target rectangular area is determined. Among them, the rectangular area can be a rectangular area surrounded by two adjacent operating parameters. Continuing with the above example, the parameter values ​​of the candidate diesel engine operating parameter B are xb, yb and zb. In three-dimensional space, according to two coordinate points, a uniquely determined rectangular area can be constructed, namely the target rectangular area. According to the parameter value, the rectangular side length of the target rectangular area can be determined. The average value of the side lengths of different rectangles is determined as the crowding distance value of the candidate diesel engine operating parameter. For example, if the side lengths of the target rectangular area are side length m1, side length m1, side length m2 and side length m2, respectively, and the four side lengths are the same in pairs; then the crowding distance value is (m1+m2) / 2.

[0133] Step d4: Determine the guiding operating condition parameters according to the crowding distance values ​​of the candidate diesel engine operating condition parameters.

[0134] Specifically, according to the crowding distance values ​​corresponding to the candidate diesel engine operating condition parameters, a certain number of candidate diesel engine operating condition parameters with larger crowding distance values ​​are selected as the guiding operating condition parameters.

[0135] Step d5: according to other operating condition parameters in the initial diesel engine operating condition parameters except the guiding operating condition parameters, based on the guiding operating condition parameters, generate target diesel engine operating condition parameters equal to the number of the guiding operating condition parameters.

[0136] A parameter is randomly selected from the guided operating parameters as the global optimal solution. The particle swarm optimization algorithm PSO is used to generate target diesel engine operating parameters equal to the number of parameters of the guided operating parameters. Specifically, the parameter size and evolution direction of the target diesel engine operating parameters are obtained by multiplying the position coordinate difference between other operating parameters and the randomly selected global optimal parameter by a preset fixed coefficient.

[0137] S330: Use the target diesel engine performance prediction model to perform performance prediction on the target diesel engine operating parameters to obtain target performance parameters corresponding to the target diesel engine operating parameters.

[0138] The target diesel engine operating condition parameters are input into a pre-trained target diesel engine performance prediction model to perform performance prediction, and the target performance parameters corresponding to the target diesel engine operating condition parameters are obtained.

[0139] The target diesel engine performance prediction model is generated by using the diesel engine performance prediction model training method described in the above embodiment.

[0140] S340, determining the optimal diesel performance parameters and the corresponding optimal operating condition parameters according to the target performance parameters of the target diesel engine operating condition parameters.

[0141] Find the optimal diesel performance parameters from the target performance parameters, and use the target diesel engine operating parameters corresponding to the optimal diesel performance parameters as the optimal operating parameters.

[0142] The technical solution of the embodiment of the present invention generates target diesel engine operating parameters based on the initial diesel engine operating parameters and its own diesel engine performance parameters; uses a target diesel engine performance prediction model to perform performance prediction on the target diesel engine operating parameters to obtain target performance parameters corresponding to the target diesel engine operating parameters; and determines optimal diesel performance parameters and their corresponding optimal operating parameters based on the target performance parameters of the target diesel engine operating parameters, thereby achieving accurate prediction and optimization of diesel engine performance and improving the efficiency of diesel engine performance prediction and optimization.

[0143] Embodiment 4

[0144] Figure 4 This is a flow chart of a diesel engine performance optimization method provided in Embodiment 4 of the present invention. This embodiment further provides a preferred embodiment based on the above embodiments.

[0145] like Figure 4 As shown, the method comprises the following steps:

[0146] S40, constructing an initial sample data set, and calibrating each diesel engine operating parameter sample in the initial sample data set based on a bench test to obtain a diesel engine standard performance parameter corresponding to each sample;

[0147] S41, using the initial sample data set to perform model training on the pre-built GPR network model to obtain a diesel engine performance prediction model;

[0148] S42, if the model error of the diesel engine performance prediction model is greater than 10%, execute S43-S45; if the model error of the diesel engine performance prediction model is not greater than 10%, execute S44;

[0149] S43A, parameter error interval corresponding to the diesel engine operating condition parameter sample with large locking error;

[0150] S43B, randomly generating a first sample data set of the same quantity (or 50% quantity) as the diesel engine operating condition parameter samples according to the parameter error interval;

[0151] S44A, sorting the samples according to the non-dominant relationship between the diesel engine operating parameter samples;

[0152] S44B, according to the results of ranking the sample individuals, determine the optimal individual under the current Pareto surface;

[0153] S44C. Based on the PSO algorithm, generate a second sample data set with the same quantity (or 50% of the quantity) as the diesel engine operating parameter samples.

[0154] S45, generating a target sample data set according to the first sample data set and the second sample data set, and continuing to train the diesel engine performance prediction model using the target sample data set until the model error is less than 10%;

[0155] S46, according to the non-dominant relationship between the diesel engine operating parameter samples, the samples are sorted by individual merit, and new diesel engine operating parameter individuals are generated based on the PSO algorithm;

[0156] S47, using the diesel engine performance prediction model to perform performance prediction on the newly generated diesel engine operating condition parameter individuals until the optimal diesel engine performance parameters are obtained.

[0157] S48. Use experiments to verify the optimal result obtained above.

[0158] In the diesel engine bench test, the particle swarm optimization algorithm is introduced to achieve efficient calibration of the test process, helping to quickly find the optimal engine performance and parameter matching scheme for a certain operating point. At the same time, the particle swarm optimization algorithm can predict and optimize the performance of similar diesel engines.

[0159] The technical solution of the above embodiment uses the particle swarm optimization algorithm to achieve efficient calibration of diesel engines and simultaneous improvement of multiple target performance. On the one hand, this method combines the test results with the optimization algorithm to achieve directional search for optimal parameters and real-time guidance on the optimization direction, thereby accelerating the calibration process; on the other hand, in the optimization process, the existing sample results are continuously used to train the Gaussian process regression model, and sample data is supplemented in a directional manner based on the feedback of the prediction results, which not only reduces the demand for sample size, but also shortens the training cycle of the Gaussian process regression model, realizes the transition from test optimization to model optimization, further reduces the test cost, and accelerates the diesel engine calibration process.

[0160] Embodiment 5

[0161] Figure 5 The structure diagram of a diesel engine performance prediction model training device provided in Embodiment 5 of the present invention is shown in FIG. A diesel engine performance prediction model training device provided in an embodiment of the present invention can be applied to the case of model training of a diesel engine performance prediction model used for diesel engine performance prediction. The diesel engine performance prediction model training device can be implemented in the form of hardware and / or software, such as Figure 5 As shown, the diesel engine performance prediction model training device can be configured in an electronic device, and specifically includes: a data set construction module 501, an initial model training module 502, a target data set generation module 503 and a target model training module 504. Among them,

[0162] The data set construction module 501 is used to construct an initial sample data set; the initial sample data set includes a diesel engine operating parameter sample and a diesel engine standard performance parameter corresponding to the diesel engine operating parameter sample;

[0163] An initial model training module 502 is used to perform model training on a pre-built network model using the initial sample data set to obtain an initial diesel engine performance prediction model, and determine a model prediction error value of the initial diesel engine performance prediction model;

[0164] A target data set generation module 503 is used to generate a target sample data set based on the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter if the model prediction error value is greater than a preset error threshold;

[0165] The target model training module 504 is used to perform model training on the initial diesel engine performance prediction model using the target sample data set until a preset model training end condition is met to obtain a target diesel engine performance prediction model.

[0166] The technical solution of the embodiment of the present invention constructs an initial sample data set, uses the initial sample data set to perform model training on the pre-constructed network model, obtains an initial diesel engine performance prediction model, and determines the model prediction error value of the initial diesel engine performance prediction model; if the model prediction error value is greater than the preset error threshold, a target sample data set is generated based on the diesel engine operating parameter sample and its corresponding diesel engine standard performance parameter; the target sample data set is used to perform model training on the initial diesel engine performance prediction model until the preset model training end condition is met, and the target diesel engine performance prediction model is obtained. The above technical solution generates a sample data set again to perform secondary training on the model after the initial model training, and performs performance calibration on the high-precision diesel engine performance prediction model obtained through training, which greatly improves the efficiency of performance calibration and reduces the test cost. .

[0167] Optionally, the initial model training module 502 includes:

[0168] A sample division unit, used for dividing the initial sample data set into a sample training set and a sample test set based on a preset division ratio;

[0169] A model training unit, used for inputting the sample training set into a pre-built network model for model training to obtain an initial diesel engine performance prediction model;

[0170] A model prediction unit, used for inputting the sample test set into the initial diesel engine performance prediction model to obtain the diesel engine prediction performance parameters output by the model;

[0171] The prediction error determination unit is used to determine the model prediction error value of the initial diesel engine performance prediction model according to the diesel engine standard performance parameters and the diesel engine predicted performance parameters of the diesel engine operating parameter samples in the sample test set.

[0172] Optionally, the target data set generation module 503 includes:

[0173] a first sample generating unit, configured to generate a first sample data set according to the sample quantity of the diesel engine operating parameter sample and based on the preset error threshold; and

[0174] A second sample generating unit, used for generating a second sample data set according to the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter;

[0175] The target sample generating unit is configured to generate a target sample data set according to the first sample data set and the second sample data set.

[0176] Optionally, the first sample generating unit is specifically configured to:

[0177] Inputting the diesel engine operating condition parameter sample into the initial diesel engine performance prediction model to obtain the diesel engine predicted performance parameter output by the model;

[0178] Determining a sample error value of the diesel engine operating parameter sample according to the diesel engine standard performance parameter and the diesel engine predicted performance parameter of the diesel engine operating parameter sample;

[0179] Determining a reference operating condition parameter sample according to the sample error value of the diesel engine operating condition parameter sample and based on the preset error threshold;

[0180] According to the parameter value interval range of the reference operating condition parameter samples, a first sample data set having the same number of samples as the diesel engine operating condition parameter samples is generated.

[0181] Optionally, the second sample generating unit includes:

[0182] a relationship generating subunit, configured to determine a dominating relationship or a non-dominating relationship between the diesel engine operating parameter samples according to the diesel engine standard performance parameters of the diesel engine operating parameter samples;

[0183] A candidate sample determination subunit, configured to determine a candidate operating condition parameter sample according to a dominant relationship and a non-dominant relationship between the diesel engine operating condition parameter samples;

[0184] a crowding distance determination subunit, configured to determine a crowding distance value of a candidate operating condition parameter sample according to a standard performance parameter of a diesel engine of the candidate operating condition parameter sample;

[0185] A guide sample determination subunit, used to determine a guide operating condition parameter sample according to the crowding distance value of the candidate operating condition parameter sample;

[0186] A second operating condition sample generating subunit, configured to generate, based on the guiding operating condition parameter samples, other operating condition parameter samples among the diesel engine operating condition parameter samples except the guiding operating condition parameter samples, second operating condition parameter samples of the same number as the guiding operating condition parameter samples;

[0187] A second performance parameter generating subunit, for determining a standard performance parameter of the diesel engine of the second operating condition parameter sample based on a bench test;

[0188] The second sample generating subunit is used to generate a second sample data set including the second operating condition parameter sample and its corresponding diesel engine standard performance parameter.

[0189] Optionally, a crowding distance determination subunit is used to:

[0190] Determining adjacent operating condition parameter samples having an adjacency relationship with the candidate operating condition parameter samples;

[0191] Determining a target rectangular area according to the parameter values ​​of the diesel engine standard performance parameters of the adjacent operating condition parameter samples;

[0192] The crowding distance value of the candidate operating condition parameter sample is determined according to the area side length of the target rectangular area.

[0193] The diesel engine performance prediction model training device provided in the embodiment of the present invention can execute the diesel engine performance prediction model training method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0194] Embodiment 6

[0195] Figure 6 The structure diagram of a diesel engine performance optimization device provided in the sixth embodiment of the present invention. The diesel engine performance optimization device provided in the embodiment of the present invention can be applied to the situation of diesel engine performance prediction and optimization. The diesel engine performance optimization device can be implemented in the form of hardware and / or software, such as Figure 6 As shown, the diesel engine performance optimization device can be configured in an electronic device, and specifically includes: a parameter acquisition module 601, a parameter generation module 602, a performance prediction module 603 and an optimal parameter determination module 604.

[0196] The parameter acquisition module 601 is used to acquire initial diesel engine operating parameters; the initial diesel engine operating parameters correspond to its own diesel engine performance parameters;

[0197] A parameter generating module 602, for generating target diesel engine operating parameters according to the initial diesel engine operating parameters and its own diesel engine performance parameters;

[0198] The performance prediction module 603 is used to perform performance prediction on the target diesel engine operating parameters using the target diesel engine performance prediction model to obtain target performance parameters corresponding to the target diesel engine operating parameters;

[0199] An optimal parameter determination module 604 is used to determine the optimal diesel performance parameter and its corresponding optimal operating condition parameter according to the target performance parameter of the target diesel engine operating condition parameter;

[0200] Wherein, the target diesel engine performance prediction model is generated by using the diesel engine performance prediction model training method described in the above embodiment.

[0201] Optionally, the parameter generation module 602 includes:

[0202] a relationship determination unit, configured to determine a dominant relationship or a non-dominant relationship between the initial diesel engine operating condition parameters according to the diesel engine performance parameters of the initial diesel engine operating condition parameters themselves;

[0203] A candidate operating condition parameter generating unit, configured to determine candidate diesel engine operating condition parameters according to a dominant relationship and a non-dominant relationship between the initial diesel engine operating condition parameters;

[0204] a crowding distance generating unit, configured to determine a crowding distance value of a candidate diesel engine operating condition parameter according to a diesel engine performance parameter of the candidate diesel engine operating condition parameter;

[0205] a guiding parameter generating unit, configured to determine the guiding operating condition parameter according to the crowding distance value of the candidate diesel engine operating condition parameter;

[0206] The operating condition parameter generating unit is used to generate target diesel engine operating condition parameters equal to the number of the guiding operating condition parameters based on the guiding operating condition parameters according to other operating condition parameters in the initial diesel engine operating condition parameters except the guiding operating condition parameters.

[0207] Optional, crowding distance generation unit, specifically used to:

[0208] Determining adjacent operating condition parameters having an adjacency relationship with the candidate diesel engine operating condition parameters;

[0209] Determining a target rectangular area according to parameter values ​​of diesel engine performance parameters of the adjacent operating condition parameters;

[0210] The crowding distance value of the candidate diesel engine operating condition parameter is determined according to the area side length of the target rectangular area.

[0211] The diesel engine performance optimization device provided in the embodiment of the present invention can execute the diesel engine performance optimization method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0212] Embodiment 7

[0213] Figure 7 A schematic diagram of an electronic device 70 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0214] like Figure 7 As shown, the electronic device 70 includes at least one processor 71, and a memory connected to the at least one processor 71, such as a read-only memory (ROM) 72, a random access memory (RAM) 73, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 71 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 72 or the computer program loaded from the storage unit 78 to the random access memory (RAM) 73. In the RAM 73, various programs and data required for the operation of the electronic device 70 can also be stored. The processor 71, the ROM 72, and the RAM 73 are connected to each other via a bus 74. An input / output (I / O) interface 75 is also connected to the bus 74.

[0215] A number of components in the electronic device 70 are connected to the I / O interface 75, including: an input unit 76, such as a keyboard, a mouse, etc.; an output unit 77, such as various types of displays, speakers, etc.; a storage unit 78, such as a disk, an optical disk, etc.; and a communication unit 79, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 79 allows the electronic device 70 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0216] The processor 71 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 71 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 71 executes the various methods and processes described above, such as diesel engine performance prediction model training and / or diesel engine performance optimization methods.

[0217] In some embodiments, the diesel engine performance prediction model training and / or diesel engine performance optimization method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 78. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 70 via the ROM 72 and / or the communication unit 79. When the computer program is loaded into the RAM 73 and executed by the processor 71, one or more steps of the diesel engine performance prediction model training and / or diesel engine performance optimization method described above may be performed. Alternatively, in other embodiments, the processor 71 may be configured to perform the diesel engine performance prediction model training and / or diesel engine performance optimization method in any other appropriate manner (e.g., by means of firmware).

[0218] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0219] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0220] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0221] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0222] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0223] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0224] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0225] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A diesel engine performance prediction model training method, characterized in that: include: Construct an initial sample dataset; The initial sample data set includes diesel engine operating parameter samples and diesel engine standard performance parameters corresponding to the diesel engine operating parameter samples; Using the initial sample data set to perform model training on a pre-built network model to obtain an initial diesel engine performance prediction model, and determining a model prediction error value of the initial diesel engine performance prediction model; If the model prediction error value is greater than a preset error threshold, a target sample data set is generated based on the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter; The target sample data set is used to perform model training on the initial diesel engine performance prediction model until a preset model training end condition is met, thereby obtaining a target diesel engine performance prediction model.

2. The method according to claim 1, characterized in that The using the initial sample data set to perform model training on the pre-built network model to obtain an initial diesel engine performance prediction model, and determining a model prediction error value of the initial diesel engine performance prediction model, includes: Dividing the initial sample data set into a sample training set and a sample test set based on a preset division ratio; Inputting the sample training set into a pre-built network model for model training to obtain an initial diesel engine performance prediction model; Inputting the sample test set into the initial diesel engine performance prediction model to obtain the diesel engine prediction performance parameters output by the model; The model prediction error value of the initial diesel engine performance prediction model is determined according to the diesel engine standard performance parameters and the diesel engine predicted performance parameters of the diesel engine operating parameter samples in the sample test set.

3. The method according to claim 1, characterized in that The generating of a target sample data set based on the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter comprises: generating a first sample data set according to the sample quantity of the diesel engine operating parameter samples and based on the preset error threshold; and, generating a second sample data set according to the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter; A target sample data set is generated according to the first sample data set and the second sample data set.

4. The method according to claim 3, characterized in that The step of generating a first sample data set based on the number of samples of the diesel engine operating parameter samples and the preset error threshold comprises: Inputting the diesel engine operating condition parameter sample into the initial diesel engine performance prediction model to obtain the diesel engine predicted performance parameter output by the model; Determining a sample error value of the diesel engine operating parameter sample according to the diesel engine standard performance parameter and the diesel engine predicted performance parameter of the diesel engine operating parameter sample; Determining a reference operating condition parameter sample according to the sample error value of the diesel engine operating condition parameter sample and based on the preset error threshold; According to the parameter value interval range of the reference operating condition parameter samples, a first sample data set having the same number of samples as the diesel engine operating condition parameter samples is generated.

5. The method according to claim 3, characterized in that: The generating of a second sample data set according to the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter comprises: determining a dominating relationship or a non-dominating relationship between the diesel engine operating condition parameter samples according to the diesel engine standard performance parameters of the diesel engine operating condition parameter samples; Determining candidate operating condition parameter samples according to the dominant relationship and the non-dominant relationship between the diesel engine operating condition parameter samples; Determining a crowding distance value of the candidate operating condition parameter sample according to the diesel engine standard performance parameter of the candidate operating condition parameter sample; Determining a guiding operating condition parameter sample according to the crowding distance value of the candidate operating condition parameter sample; According to other operating condition parameter samples among the diesel engine operating condition parameter samples except the guiding operating condition parameter samples, based on the guiding operating condition parameter samples, generating second operating condition parameter samples of the same number as the number of samples of the guiding operating condition parameter samples; Determining standard performance parameters of the diesel engine of the second operating condition parameter sample based on a bench test; A second sample data set including the second operating condition parameter samples and their corresponding diesel engine standard performance parameters is generated.

6. The method according to claim 5, characterized in that The step of determining the crowding distance value of the candidate operating condition parameter sample according to the diesel engine standard performance parameter of the candidate operating condition parameter sample comprises: Determining adjacent operating condition parameter samples having an adjacency relationship with the candidate operating condition parameter samples; Determining a target rectangular area according to the parameter values ​​of the diesel engine standard performance parameters of the adjacent operating condition parameter samples; The crowding distance value of the candidate operating condition parameter sample is determined according to the area side length of the target rectangular area.

7. A diesel engine performance optimization method, characterized in that: include: Obtaining initial diesel engine operating parameters; The initial diesel engine operating condition parameters correspond to its own diesel engine performance parameters; Generating target diesel engine operating parameters according to the initial diesel engine operating parameters and its own diesel engine performance parameters; Using a target diesel engine performance prediction model to perform performance prediction on the target diesel engine operating parameters to obtain target performance parameters corresponding to the target diesel engine operating parameters; Determining optimal diesel performance parameters and their corresponding optimal operating parameters according to target performance parameters of the target diesel engine operating parameters; Wherein, the target diesel engine performance prediction model is generated by using the diesel engine performance prediction model training method described in any one of claims 1-6.

8. The method according to claim 7, characterized in that The generating target diesel engine operating parameters according to the initial diesel engine operating parameters and its own diesel engine performance parameters comprises: determining a dominating relationship or a non-dominating relationship between the initial diesel engine operating condition parameters according to the diesel engine performance parameters of the initial diesel engine operating condition parameters themselves; Determining candidate diesel engine operating condition parameters according to the dominating relationship and the non-dominating relationship between the initial diesel engine operating condition parameters; Determining a crowding distance value of the candidate diesel engine operating condition parameter according to the diesel engine performance parameter of the candidate diesel engine operating condition parameter; Determining a guiding operating condition parameter according to the crowding distance value of the candidate diesel engine operating condition parameter; According to other operating condition parameters in the initial diesel engine operating condition parameters except the guide operating condition parameters, target diesel engine operating condition parameters equal in number to the guide operating condition parameters are generated based on the guide operating condition parameters.

9. The method according to claim 8, characterized in that The step of determining the crowding distance value of the candidate diesel engine operating condition parameter according to the diesel engine performance parameter of the candidate diesel engine operating condition parameter comprises: Determining adjacent operating condition parameters having an adjacency relationship with the candidate diesel engine operating condition parameters; Determining a target rectangular area according to parameter values ​​of diesel engine performance parameters of the adjacent operating condition parameters; The crowding distance value of the candidate diesel engine operating condition parameter is determined according to the area side length of the target rectangular area.

10. A diesel engine performance prediction model training device, characterized in that: include: Dataset construction module, used to construct the initial sample dataset; The initial sample data set includes diesel engine operating parameter samples and diesel engine standard performance parameters corresponding to the diesel engine operating parameter samples; An initial model training module is used to perform model training on a pre-built network model using the initial sample data set to obtain an initial diesel engine performance prediction model, and determine a model prediction error value of the initial diesel engine performance prediction model; A target data set generation module, configured to generate a target sample data set based on the diesel engine operating condition parameter sample and its corresponding diesel engine standard performance parameter if the model prediction error value is greater than a preset error threshold; The target model training module is used to use the target sample data set to perform model training on the initial diesel engine performance prediction model until a preset model training end condition is met to obtain a target diesel engine performance prediction model.

11. A diesel engine performance optimization device, characterized in that: include: A parameter acquisition module is used to obtain initial diesel engine operating parameters; The initial diesel engine operating condition parameters correspond to its own diesel engine performance parameters; A parameter generation module, used for generating target diesel engine operating parameters according to the initial diesel engine operating parameters and its own diesel engine performance parameters; A performance prediction module, used to perform performance prediction on the target diesel engine operating parameters using a target diesel engine performance prediction model, and obtain target performance parameters corresponding to the target diesel engine operating parameters; An optimal parameter determination module, used to determine the optimal diesel performance parameter and its corresponding optimal operating condition parameter according to the target performance parameter of the target diesel engine operating condition parameter; Wherein, the target diesel engine performance prediction model is generated by using the diesel engine performance prediction model training method described in any one of claims 1-6.

12. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the diesel engine performance prediction model training described in any one of claims 1-6, and / or, perform the diesel engine performance optimization method described in any one of claims 7-9.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which are used to enable a processor to implement the diesel engine performance prediction model training described in any one of claims 1-6 and / or the diesel engine performance optimization method described in any one of claims 7-9 when executed.

14. A computer program product, characterized in that The computer program product comprises a computer program, which, when executed by a processor, implements the diesel engine performance prediction model training according to any one of claims 1-6, and / or implements the diesel engine performance optimization method according to any one of claims 7-9.

Images