A method for early warning of diesel engine pollutant emissions

Abstract

This invention relates to the field of pollutant emission early warning technology, and discloses a method for early warning of diesel engine pollutant emissions. The method involves identifying the diesel engine to be warned, acquiring its data, and constructing a proportional simulation model of the diesel engine based on the data. Multiple monitoring grids are deployed on the proportional simulation model to acquire future operating data of the diesel engine, which is then simulated on the monitoring grids. Simulation values ​​are collected for each monitoring grid, and the predicted pollutant emissions of the diesel engine are calculated based on all simulation values. Based on the predicted pollutant emissions, a pollutant emission early warning is issued for the diesel engine. This method can accurately determine the pollutant emission situation of the diesel engine, issue real-time early warnings, and reduce the impact of pollutants on the environment and human health.

Description

Technical Field

[0001] This invention relates to the field of pollutant emission early warning technology, and more specifically, to a method for early warning of diesel engine pollutant emissions. Background Technology

[0002] With the acceleration of industrialization and the increasing congestion of urban traffic, diesel engines, as a highly efficient and economical power source, are widely used in various fields such as transportation, construction machinery, agricultural machinery, and power generation. However, while providing power, the combustion process of diesel engines inevitably produces various pollutants, including nitrogen oxides (NOx), particulate matter (PM), carbon monoxide (CO), and unburned hydrocarbons (HC).

[0003] There are generally two methods for determining the pollutants emitted by diesel engines. The first is the sensor-based field detection method, which directly measures and monitors emitted pollutants by installing various sensors in the diesel engine's exhaust system, measuring the concentrations of key pollutants such as carbon monoxide, hydrocarbons, nitrogen oxides, and particulate matter in real time. However, this method can be affected by sensor accuracy and changes in environmental conditions. The second method is the diesel engine combustion process fitting calculation method, which predicts emission characteristics under different operating conditions based on detailed simulations and calculations of the diesel engine combustion process. This method typically includes specific emissions and displacement calculations. Specific emissions refer to the pollutant emissions per unit power or unit workload, while displacement calculation refers to the total pollutant emissions over the entire operating cycle. However, this method requires an accurate combustion model and detailed input parameters, making the calculations complex and highly dependent on the accuracy of the model. Summary of the Invention

[0004] This invention provides a method for early warning of diesel engine pollutant emissions, which solves the technical problem in the prior art that it is impossible to accurately determine the pollutant emissions of diesel engines and issue emission warnings in real time.

[0005] To achieve the above objectives, the present invention provides a method for early warning of diesel engine pollutant emissions, comprising:

[0006] Identify the diesel engine to be warned, acquire the data information of the diesel engine to be warned, and construct a proportional simulation model of the diesel engine to be warned based on the data information;

[0007] Multiple monitoring grids are deployed on the scaled simulation model to obtain future operating data of the diesel engine to be warned. The future operating data is then simulated on the monitoring grids of the scaled simulation model. The future operating data includes: operating status data, vibration signal data, electrical data, and physical status data.

[0008] The simulation values ​​of each monitoring grid are collected, and the predicted pollutant emissions of the diesel engine to be warned are calculated based on all the simulation values.

[0009] Based on the predicted pollutant emissions, a pollutant emission warning should be issued for the diesel engine to be warned.

[0010] Furthermore, when acquiring the data information of the diesel engine to be warned and constructing a proportional simulation model of the diesel engine to be warned based on the data information, the process includes:

[0011] The number of neurons in the input layer, hidden layer, and output layer is determined based on a preset method, and the number of neurons in the input layer, the hidden layer, and the output layer are proportional. The number of neurons in the input layer is determined based on the number of features, the number of neurons in the output layer is determined based on the task type, and the number of neurons in the hidden layer is determined based on cross-validation.

[0012] Based on the proportional relationship, calculate the weights and biases corresponding to the input layer, hidden layer and output layer respectively;

[0013] A training dataset is obtained from the historical data repository of the diesel engine to be warned, and the neural network is trained based on the training dataset until the preset training termination condition is met to obtain the initial proportional simulation model.

[0014] The network's predicted output is calculated through continuous matrix operations and activation function processing. The matrix operations include the hidden layer receiving input from the input layer and the output layer receiving input from the hidden layer, and performing linear transformations using a weight matrix and a bias vector. The activation function processing includes introducing nonlinear characteristics, with each neuron's linear output being processed by an activation function. The linear transformations of the input layer and the activation function processing are concatenated to obtain a first output. The first output is then passed to the hidden layer, and the linear transformations of the hidden layer and the activation function processing are concatenated to obtain a second output. The second output is then passed to the output layer, and the linear transformations of the output layer and the activation function processing are concatenated to obtain the final predicted output.

[0015] The difference between the predicted output and the true target value is quantified using a loss function to determine whether the initial proportional simulation model meets the performance requirements.

[0016] If the conditions are met, the proportional simulation model is obtained.

[0017] Furthermore, when deploying multiple monitoring grids on the scaled simulation model, this includes:

[0018] Obtain the model information of the scaled simulation model, wherein the model information includes the model volume and the model surface area;

[0019] Based on the model information, an initial deployment is carried out on the scaled simulation model to obtain a primary deployment strategy. The primary deployment strategy includes: selecting the initial deployment location of the monitoring equipment according to the model volume and the surface area, dividing the model surface into multiple monitoring grids, deploying monitoring equipment in each grid, and the number of monitoring equipment is greater than or equal to 1.

[0020] Determine the sensing reliability and sensing coverage of each monitoring grid;

[0021] The deployment reliability of the primary deployment strategy is calculated based on the perception reliability and perception coverage of each monitoring grid.

[0022] If the deployment confidence level is greater than or equal to the preset deployment confidence level, it is determined that no adjustment is needed to the primary deployment strategy;

[0023] If the deployment confidence is less than the preset deployment confidence, the primary deployment strategy is adjusted until the obtained deployment confidence is greater than or equal to the preset deployment confidence.

[0024] Furthermore, when calculating the deployment reliability of the primary deployment strategy based on the perceived reliability and perceived coverage of each monitoring grid, the following steps are included:

[0025] The deployment confidence of the primary deployment strategy is calculated using the following formula:

[0026] ;

[0027] Where S is the deployment confidence of the initial deployment strategy, n is the number of monitoring grids, and r1 k Let Δr1 be the sensing reliability corresponding to the k-th monitoring grid. k Let r2 be the standard sensing reliability corresponding to the k-th monitoring grid. k Let Δr2 be the sensing coverage corresponding to the k-th monitoring grid. k This represents the standard sensing coverage corresponding to the k-th monitoring grid.

[0028] Furthermore, when acquiring the future operating data of the diesel engine to be warned, and simulating the future operating data on the monitoring grid of the proportional simulation model, the process includes:

[0029] Obtain the adjustment range of the proportional simulation model and the diesel engine to be warned, corresponding to future working data;

[0030] The number of simulations of the future working data on the proportional simulation model is set according to the adjustment range;

[0031] Based on the number of simulations, the future working data is simulated on the monitoring grid of the proportional simulation model.

[0032] Furthermore, when collecting simulation values ​​for each monitoring grid and calculating the predicted pollutant emissions of the diesel engine to be warned based on all simulation values, the process includes:

[0033] Obtain a preset simulation value, and divide all simulation values ​​smaller than the preset simulation value into a first simulation value set;

[0034] All simulation values ​​that are greater than or equal to the preset simulation value are assigned to the second simulation value set.

[0035] Calculate the first simulation average value of the first simulation value set, and calculate the second simulation average value of the second simulation value set;

[0036] Count the number of first simulation values ​​in the first set of simulation values ​​that are greater than the average value of the first simulation; count the number of second simulation values ​​in the second set of simulation values ​​that are greater than the average value of the second simulation.

[0037] The predicted pollutant emissions of the diesel engine to be warned are calculated based on the first simulation average value, the second simulation average value, the number of first simulation values, and the number of second simulation values.

[0038] Further, when calculating the predicted pollutant emissions of the diesel engine to be warned based on the first simulation average value, the second simulation average value, the number of first simulation values, and the number of second simulation values, the process includes:

[0039] ;

[0040] Where Q is the predicted pollutant emissions of the diesel engine to be warned, t1 is the first calculation coefficient, w1 is the first simulation average value, Y1 is the number of first simulation values, Y2 is the number of second simulation values, and w2 is the second simulation average value.

[0041] Furthermore, when determining whether to issue a pollutant emission warning to the diesel engine to be warned based on the predicted pollutant emissions, the process includes:

[0042] Obtain a pre-set predicted pollutant emission amount, and determine whether to issue a pollutant emission warning to the diesel engine to be warned based on the relationship between the predicted pollutant emission amount and the pre-set predicted pollutant emission amount;

[0043] If the predicted pollutant emissions are less than the preset predicted pollutant emissions, then it is determined that no pollutant emission warning will be issued to the diesel engine to be warned.

[0044] When the predicted pollutant emissions are greater than or equal to the preset predicted pollutant emissions, it is determined that a pollutant emission warning will be issued to the diesel engine to be warned.

[0045] Furthermore, after determining whether to issue a pollutant emission warning to the diesel engine to be warned based on the predicted pollutant emissions, the method further includes:

[0046] Extract all simulation values ​​from the second set of simulation values ​​and determine the cluster centers;

[0047] Determine the cluster distance from each simulated value to the cluster center, and perform curve fitting on all cluster distances based on a preset method to obtain a cluster distance fitting curve;

[0048] Determine the slope of the clustering distance fitting curve, and based on the slope-period mapping table, determine the next warning time for the diesel engine to be warned;

[0049] Identify the unfitted cluster distances and calculate the average cluster distance of all unfitted cluster distances;

[0050] The warning cycle adjustment coefficient of the diesel engine to be warned is set based on the average clustering distance, and the next warning time is adjusted based on the warning cycle adjustment coefficient to obtain the next target warning time;

[0051] Based on the next target warning time, periodic pollutant emission warnings are issued for the diesel engine to be warned.

[0052] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0053] This invention discloses a method for early warning of diesel engine pollutant emissions. The method involves identifying the diesel engine to be warned, acquiring its data, and constructing a proportional simulation model based on this data. Multiple monitoring grids are deployed on the proportional simulation model to acquire future operating data of the diesel engine. This future operating data is then simulated on the monitoring grids of the proportional simulation model. Simulation values ​​for each monitoring grid are collected, and the predicted pollutant emissions of the diesel engine are calculated based on all simulation values. Based on the predicted pollutant emissions, a method is used to determine whether to issue an early warning for the diesel engine. This method can accurately determine the pollutant emissions of the diesel engine, issue real-time warnings, and reduce the impact of pollutants on the environment and human health. Attached Figure Description

[0054] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0055] Figure 1 A flowchart illustrating a diesel engine pollutant emission early warning method according to an embodiment of the present invention is shown. Detailed Implementation

[0056] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0057] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0058] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0059] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0060] The following is a description of preferred embodiments of the present invention in conjunction with the accompanying drawings.

[0061] like Figure 1 As shown, an embodiment of the present invention discloses a method for early warning of diesel engine pollutant emissions, comprising:

[0062] S110: Identify the diesel engine to be warned, acquire the data information of the diesel engine to be warned, and construct a proportional simulation model of the diesel engine to be warned based on the data information;

[0063] In some embodiments of this application, the process of acquiring data information of the diesel engine to be warned and constructing a proportional simulation model of the diesel engine to be warned based on the data information includes:

[0064] The number of neurons in the input layer, hidden layer, and output layer is determined based on a preset method, and the number of neurons in the input layer, the hidden layer, and the output layer are proportional. The number of neurons in the input layer is determined based on the number of features, the number of neurons in the output layer is determined based on the task type, and the number of neurons in the hidden layer is determined based on cross-validation.

[0065] Based on the proportional relationship, calculate the weights and biases corresponding to the input layer, hidden layer and output layer respectively;

[0066] A training dataset is obtained from the historical data repository of the diesel engine to be warned, and the neural network is trained based on the training dataset until the preset training termination condition is met to obtain the initial proportional simulation model.

[0067] The network's predicted output is calculated through continuous matrix operations and activation function processing. The matrix operations include the hidden layer receiving input from the input layer and the output layer receiving input from the hidden layer, and performing linear transformations using a weight matrix and a bias vector. The activation function processing includes introducing nonlinear characteristics, with each neuron's linear output being processed by an activation function. The linear transformations of the input layer and the activation function processing are concatenated to obtain a first output. The first output is then passed to the hidden layer, and the linear transformations of the hidden layer and the activation function processing are concatenated to obtain a second output. The second output is then passed to the output layer, and the linear transformations of the output layer and the activation function processing are concatenated to obtain the final predicted output.

[0068] The difference between the predicted output and the true target value is quantified using a loss function to determine whether the initial proportional simulation model meets the performance requirements.

[0069] If the conditions are met, the proportional simulation model is obtained.

[0070] In this embodiment, the data information includes operating status parameters, vibration signal parameters, electrical parameters, physical status parameters, etc.

[0071] In this embodiment, the number of neurons in the input layer, hidden layer, and output layer is determined based on a preset method (such as empirical rules, cross-validation, etc.). This ensures that the number of neurons in each layer is proportional to maintain the consistency of the network structure.

[0072] Determine the number of neurons in the input layer;

[0073] The number of neurons in the input layer is typically determined by the number of features in the dataset. If you have *n* features, then the input layer should have *n* neurons.

[0074] Determine the number of neurons in the output layer;

[0075] The number of neurons in the output layer is usually determined by the task type:

[0076] Binary classification problem: 1 neuron (usually using the Sigmoid activation function).

[0077] Multi-class classification problem: number of classes mm, number of neurons (usually using the Softmax activation function).

[0078] Regression problem: 1 neuron (linear activation function or no activation function).

[0079] Determine the number of neurons in the hidden layer;

[0080] The number of neurons in the hidden layer is determined by cross-validation.

[0081] To maintain the consistency of the network structure, the number of neurons in each layer must be proportional. For example, if the input layer has 100 neurons, the first hidden layer has 50 neurons, and the second hidden layer can have 25 neurons, so that the number of neurons in each layer is half that of the previous layer.

[0082] In this embodiment, weights and biases are calculated for each layer of neurons according to a proportional relationship. These weights and biases will be used for information transmission and processing in the neural network.

[0083] Weight initialization, including the following methods:

[0084] Random initialization: Use small random numbers (such as those following a standard normal distribution) to initialize the weights.

[0085] Xavier initialization: Especially suitable for Sigmoid and tanh activation functions, it avoids gradient vanishing or exploding by initializing the weight matrix to small, uniformly distributed values.

[0086] He initialization: Especially suitable for ReLU and its variant activation functions, it avoids gradient vanishing or exploding by initializing the weight matrix to small, uniformly distributed values.

[0087] Bias initialization; biases are typically initialized to zero or a small constant, such as 0.01.

[0088] In this embodiment, the neural network is trained using a training dataset. The training process continuously adjusts the weights and biases until preset training termination conditions are met (such as reaching a certain number of iterations, an error threshold, or validation set performance).

[0089] In this embodiment, after training is completed, an initial scaled simulation model is obtained. This model can be used for prediction and analysis.

[0090] In this embodiment, the network's predicted output is calculated through continuous matrix operations and activation function processing. This step involves the forward propagation of data.

[0091] Linear transformation: The neurons in each layer receive input from the previous layer and undergo a linear transformation through the weight matrix and bias vector.

[0092] Activation function processing: To introduce non-linearity, the linear output of each neuron is typically processed by an activation function. Common activation functions include Sigmoid, ReLU, and tanh.

[0093] Layer-by-layer propagation: The linear transformations and activation functions of each layer are chained together, starting from the input layer and proceeding to the output layer to obtain the final predicted output.

[0094] In this embodiment, a loss function is used to quantify the difference between the predicted output and the true target value. Loss functions include Mean Squared Error (MSE) and Cross-Entropy Loss, etc. The specific loss function can be selected based on the actual situation. The smaller the value of the loss function, the closer the model's predictive ability is to the actual situation. If the value of the loss function is sufficiently small, or the model performs well on other performance metrics, then the model is considered to meet the performance requirements. If unsatisfactory, the network structure or training parameters are adjusted, and the model is retrained.

[0095] The beneficial effects of the above technical solution are: the present invention can provide a guarantee for the early warning of emissions from diesel engines under warning by training a proportional simulation model, and accurately realize the early warning.

[0096] S120: Deploy multiple monitoring grids on the scaled simulation model to obtain future operating data of the diesel engine to be warned, and simulate the future operating data on the monitoring grids of the scaled simulation model. The future operating data includes: operating status data, vibration signal data, electrical data, and physical status data.

[0097] In some embodiments of this application, when deploying multiple monitoring grids on the scaled simulation model, the following are included:

[0098] Obtain the model information of the scaled simulation model, wherein the model information includes the model volume and the model surface area;

[0099] Based on the model information, an initial deployment is carried out on the scaled simulation model to obtain a primary deployment strategy. The primary deployment strategy includes: selecting the initial deployment location of the monitoring equipment according to the model volume and the surface area, dividing the model surface into multiple monitoring grids, deploying monitoring equipment in each grid, and the number of monitoring equipment is greater than or equal to 1.

[0100] Determine the sensing reliability and sensing coverage of each monitoring grid;

[0101] The deployment reliability of the primary deployment strategy is calculated based on the perception reliability and perception coverage of each monitoring grid.

[0102] If the deployment confidence level is greater than or equal to the preset deployment confidence level, it is determined that no adjustment is needed to the primary deployment strategy;

[0103] If the deployment confidence is less than the preset deployment confidence, the primary deployment strategy is adjusted until the obtained deployment confidence is greater than or equal to the preset deployment confidence.

[0104] In this embodiment, the initial deployment location of the monitoring equipment is selected based on the model's volume and surface area. The model surface is divided into multiple monitoring grids, and monitoring equipment is deployed in each grid. The number of monitoring devices is greater than or equal to one. The type and number of devices to be deployed in each grid are determined, depending on the monitoring target and the expected monitoring accuracy.

[0105] In this embodiment, the sensing reliability of each monitoring grid is analyzed, that is, the probability that the devices within that grid can accurately detect the target. The sensing coverage of each monitoring grid is determined, that is, the percentage of the grid covered by monitoring devices.

[0106] In this embodiment, the preset deployment confidence level is a threshold as the minimum acceptable deployment confidence level.

[0107] The beneficial effects of the above technical solution are: the present invention calculates the deployment reliability of the primary deployment strategy based on the perception reliability and perception coverage of each monitoring grid, and determines whether to adjust the primary deployment strategy based on the deployment reliability and the preset deployment reliability. The aim is to achieve the best monitoring effect through continuous evaluation and adjustment, so as to ensure the efficiency and reliability of the monitoring system.

[0108] In some embodiments of this application, calculating the deployment confidence of the primary deployment strategy based on the perceived reliability and perceived coverage of each monitoring grid includes:

[0109] The deployment confidence of the primary deployment strategy is calculated using the following formula:

[0110] ;

[0111] Where S is the deployment confidence of the initial deployment strategy, n is the number of monitoring grids, and r1 k Let Δr1 be the sensing reliability corresponding to the k-th monitoring grid. k Let r2 be the standard sensing reliability corresponding to the k-th monitoring grid. k Let Δr2 be the sensing coverage corresponding to the k-th monitoring grid. k This represents the standard sensing coverage corresponding to the k-th monitoring grid.

[0112] In some embodiments of this application, when acquiring future operating data of the diesel engine to be warned and simulating the future operating data on the monitoring grid of the proportional simulation model, the following steps are included:

[0113] Obtain the adjustment range of the proportional simulation model and the diesel engine to be warned, corresponding to future working data;

[0114] The number of simulations of the future working data on the proportional simulation model is set according to the adjustment range;

[0115] ;

[0116] Where P is the number of simulations, and A(b, c) is the adjustment range of the proportional simulation model and the diesel engine to be warned corresponding to future working data. e Let A(b, c) be the right boundary value of the adjustment range corresponding to future working data for the proportional simulation model and the diesel engine to be warned. f The left boundary value of the adjustment range corresponding to future working data for the proportional simulation model and the diesel engine to be warned;

[0117] Based on the number of simulations, the future working data is simulated on the monitoring grid of the proportional simulation model.

[0118] In this embodiment, the adjustment range refers to the predicted range of various operating condition changes that the diesel engine may encounter during future operation. These ranges are typically determined by factors such as equipment workload and environmental conditions. Data sources for the adjustment range include historical operating data, equipment performance parameters, and anticipated changes in the operating environment. By analyzing and predicting historical and real-time data, the adjustment range for future operating data can be obtained.

[0119] In this embodiment, for example, the load variation range may be 5%-100%, the temperature variation range may be -20℃ to 40℃, and the speed variation range may be 600rpm to 2400rpm, etc. The above are just examples and are not specific limitations.

[0120] The beneficial effects of the above technical solution are: This invention simulates future working data on the monitoring grid of a proportional simulation model based on the number of simulations, enabling a comprehensive evaluation of the diesel engine's performance and reliability. By setting the number of simulations, the accuracy and coverage of the simulation can be improved. This helps optimize the design and operation strategies of the diesel engine.

[0121] S130: Collect the simulation values ​​of each monitoring grid, and calculate the predicted pollutant emissions of the diesel engine to be warned based on all the simulation values;

[0122] In some embodiments of this application, when collecting simulation values ​​for each monitoring grid and calculating the predicted pollutant emissions of the diesel engine to be warned based on all simulation values, the process includes:

[0123] Obtain a preset simulation value, and divide all simulation values ​​smaller than the preset simulation value into a first simulation value set;

[0124] All simulation values ​​that are greater than or equal to the preset simulation value are assigned to the second simulation value set.

[0125] Calculate the first simulation average value of the first simulation value set, and calculate the second simulation average value of the second simulation value set;

[0126] Count the number of first simulation values ​​in the first set of simulation values ​​that are greater than the average value of the first simulation; count the number of second simulation values ​​in the second set of simulation values ​​that are greater than the average value of the second simulation.

[0127] The predicted pollutant emissions of the diesel engine to be warned are calculated based on the first simulation average value, the second simulation average value, the number of first simulation values, and the number of second simulation values.

[0128] In this embodiment, the preset simulation value is set based on specific simulation requirements and objectives. It is used to divide the simulation value and is a preset value. The preset simulation value is preferably 8.

[0129] The beneficial effects of the above technical solution are: the present invention calculates the predicted pollutant emissions of the diesel engine to be warned based on the first simulation average value, the second simulation average value, the number of first simulation values ​​and the number of second simulation values, and provides the calculation of the predicted pollutant emissions of the diesel engine to be warned, which can lay the foundation for emission warning, provide warning data, avoid prediction errors, and avoid affecting the normal operation of the diesel engine.

[0130] In some embodiments of this application, when calculating the predicted pollutant emissions of the diesel engine to be warned based on the first simulation average value, the second simulation average value, the number of first simulation values, and the number of second simulation values, the following steps are included:

[0131] ;

[0132] Where Q is the predicted pollutant emissions of the diesel engine to be warned, t1 is the first calculation coefficient, w1 is the first simulation average value, Y1 is the number of first simulation values, Y2 is the number of second simulation values, and w2 is the second simulation average value.

[0133] S140: Based on the predicted pollutant emissions, determine whether to issue a pollutant emission warning to the diesel engine to be warned.

[0134] In some embodiments of this application, determining whether to issue a pollutant emission warning to the diesel engine to be warned based on the predicted pollutant emissions includes:

[0135] Obtain a pre-set predicted pollutant emission amount, and determine whether to issue a pollutant emission warning to the diesel engine to be warned based on the relationship between the predicted pollutant emission amount and the pre-set predicted pollutant emission amount;

[0136] If the predicted pollutant emissions are less than the preset predicted pollutant emissions, then it is determined that no pollutant emission warning will be issued to the diesel engine to be warned.

[0137] When the predicted pollutant emissions are greater than or equal to the preset predicted pollutant emissions, it is determined that a pollutant emission warning will be issued to the diesel engine to be warned.

[0138] In this embodiment, the preset predicted pollutant emission amount is a value used to measure whether the pollutant emission meets the relevant requirements. It can be set according to the actual situation of the diesel engine to meet the emission standards. Here, it is preferably 8g / kwh.

[0139] The beneficial effects of the above technical solution are: it can accurately determine the pollutant emissions of diesel engines, issue early warnings in real time, and reduce the impact of pollutants on the environment and human health.

[0140] In some embodiments of this application, after determining whether to issue a pollutant emission warning to the diesel engine to be warned based on the predicted pollutant emissions, the method further includes:

[0141] Extract all simulation values ​​from the second set of simulation values ​​and determine the cluster centers;

[0142] Determine the cluster distance from each simulated value to the cluster center, and perform curve fitting on all cluster distances based on a preset method to obtain a cluster distance fitting curve;

[0143] Determine the slope of the clustering distance fitting curve, and based on the slope-period mapping table, determine the next warning time for the diesel engine to be warned;

[0144] Identify the unfitted cluster distances and calculate the average cluster distance of all unfitted cluster distances;

[0145] The warning cycle adjustment coefficient of the diesel engine to be warned is set based on the average clustering distance, and the next warning time is adjusted based on the warning cycle adjustment coefficient to obtain the next target warning time;

[0146] Based on the next target warning time, periodic pollutant emission warnings are issued for the diesel engine to be warned.

[0147] In this embodiment, the preset methods include least squares method, polynomial fitting method, etc., and the specific method can be selected according to the requirements.

[0148] In this embodiment, the early warning cycle adjustment coefficient of the diesel engine to be warned is set based on the average clustering distance-adjustment coefficient mapping table, and the product of the adjustment coefficient and the next early warning cycle is used as the next target early warning time.

[0149] The beneficial effects of the above technical solution are: by setting the next target warning time, the present invention can realize the periodic warning of the diesel engine to be warned, which can avoid the warning being untimely and also avoid unnecessary work.

[0150] In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in any suitable manner in one or more embodiments or examples.

[0151] Although the invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the embodiments disclosed in this invention can be combined with each other in any way. The fact that not all of these combinations are described in this specification is merely for the sake of brevity and resource conservation.

[0152] It will be understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for early warning of diesel engine pollutant emissions, characterized in that, include: Identify the diesel engine to be warned, acquire the data information of the diesel engine to be warned, and construct a proportional simulation model of the diesel engine to be warned based on the data information; Multiple monitoring grids are deployed on the scaled simulation model to obtain future operating data of the diesel engine to be warned. The future operating data is then simulated on the monitoring grids of the scaled simulation model. The future operating data includes: operating status data, vibration signal data, electrical data, and physical status data. The simulation values ​​of each monitoring grid are collected, and the predicted pollutant emissions of the diesel engine to be warned are calculated based on all the simulation values. Based on the predicted pollutant emissions, determine whether to issue a pollutant emission warning to the diesel engine to be warned; When collecting simulation values ​​for each monitoring grid and calculating the predicted pollutant emissions of the diesel engine to be warned based on all simulation values, the following steps are included: Obtain a preset simulation value, and divide all simulation values ​​smaller than the preset simulation value into a first simulation value set; All simulation values ​​that are greater than or equal to the preset simulation value are assigned to the second simulation value set. Calculate the first simulation average value of the first simulation value set, and calculate the second simulation average value of the second simulation value set; Count the number of first simulation values ​​in the first set of simulation values ​​that are greater than the average value of the first simulation; count the number of second simulation values ​​in the second set of simulation values ​​that are greater than the average value of the second simulation. The predicted pollutant emissions of the diesel engine to be warned are calculated based on the first simulation average value, the second simulation average value, the number of first simulation values, and the number of second simulation values. When calculating the predicted pollutant emissions of the diesel engine to be warned based on the first simulation average value, the second simulation average value, the number of first simulation values, and the number of second simulation values, the following steps are included: Where Q is the predicted pollutant emissions of the diesel engine to be warned, t1 is the first calculation coefficient, w1 is the first simulation average value, Y1 is the number of first simulation values, Y2 is the number of second simulation values, and w2 is the second simulation average value.

2. The diesel engine pollutant emission early warning method according to claim 1, characterized in that, When acquiring data information of the diesel engine to be warned and constructing a proportional simulation model of the diesel engine to be warned based on the data information, the process includes: The number of neurons in the input layer, hidden layer, and output layer is determined based on a preset method, and the number of neurons in the input layer, the hidden layer, and the output layer are proportional. The number of neurons in the input layer is determined based on the number of features, the number of neurons in the output layer is determined based on the task type, and the number of neurons in the hidden layer is determined based on cross-validation. Based on the proportional relationship, calculate the weights and biases corresponding to the input layer, hidden layer and output layer respectively; A training dataset is obtained from the historical data repository of the diesel engine to be warned, and the neural network is trained based on the training dataset until the preset training termination condition is met to obtain the initial proportional simulation model. The network's predicted output is calculated through continuous matrix operations and activation function processing. The matrix operations include the hidden layer receiving input from the input layer and the output layer receiving input from the hidden layer, and performing linear transformations using a weight matrix and a bias vector. The activation function processing includes introducing nonlinear characteristics, with each neuron's linear output being processed by an activation function. The linear transformation of the input layer and the activation function processing are concatenated to obtain a first output. The first output is passed to the hidden layer, and the linear transformation of the hidden layer and the activation function processing are concatenated to obtain a second output. The second output is passed to the output layer, and the linear transformation of the output layer and the activation function processing are concatenated to obtain the final predicted output. The difference between the predicted output and the true target value is quantified using a loss function to determine whether the initial proportional simulation model meets the performance requirements. If the conditions are met, the proportional simulation model is obtained.

3. The diesel engine pollutant emission early warning method according to claim 1, characterized in that, When deploying multiple monitoring grids on the scaled simulation model, the following are included: Obtain the model information of the scaled simulation model, wherein the model information includes the model volume and the model surface area; Based on the model information, an initial deployment is carried out on the scaled simulation model to obtain a primary deployment strategy. The primary deployment strategy includes: selecting the initial deployment location of the monitoring equipment according to the model volume and the surface area, dividing the model surface into multiple monitoring grids, deploying monitoring equipment in each grid, and the number of monitoring equipment is greater than or equal to 1. Determine the sensing reliability and sensing coverage of each monitoring grid; The deployment reliability of the primary deployment strategy is calculated based on the perception reliability and perception coverage of each monitoring grid. If the deployment confidence level is greater than or equal to the preset deployment confidence level, it is determined that no adjustment is needed to the primary deployment strategy; If the deployment confidence is less than the preset deployment confidence, the primary deployment strategy is adjusted until the obtained deployment confidence is greater than or equal to the preset deployment confidence.

4. The diesel engine pollutant emission early warning method according to claim 3, characterized in that, When calculating the deployment reliability of the primary deployment strategy based on the perceived reliability and perceived coverage of each monitoring grid, the following is included: The deployment confidence of the primary deployment strategy is calculated using the following formula: Where S represents the deployment confidence of the initial deployment strategy, n represents the number of monitoring grids, and r1 k Let Δr1 be the sensing reliability corresponding to the k-th monitoring grid. k Let r2 be the standard sensing reliability corresponding to the k-th monitoring grid. k Let Δr2 be the sensing coverage corresponding to the k-th monitoring grid. k This represents the standard sensing coverage corresponding to the k-th monitoring grid.

5. The diesel engine pollutant emission early warning method according to claim 1, characterized in that, When acquiring the future operating data of the diesel engine to be warned, and simulating the future operating data on the monitoring grid of the proportional simulation model, the following steps are included: Obtain the adjustment range of the proportional simulation model and the diesel engine to be warned, corresponding to future working data; The number of simulations of the future working data on the proportional simulation model is set according to the adjustment range; Based on the number of simulations, the future working data is simulated on the monitoring grid of the proportional simulation model.

6. The diesel engine pollutant emission early warning method according to claim 1, characterized in that, When determining whether to issue a pollutant emission warning to the diesel engine to be warned based on the predicted pollutant emissions, the following steps are included: Obtain a pre-set predicted pollutant emission amount, and determine whether to issue a pollutant emission warning to the diesel engine to be warned based on the relationship between the predicted pollutant emission amount and the pre-set predicted pollutant emission amount; If the predicted pollutant emissions are less than the preset predicted pollutant emissions, then it is determined that no pollutant emission warning will be issued to the diesel engine to be warned. When the predicted pollutant emissions are greater than or equal to the preset predicted pollutant emissions, it is determined that a pollutant emission warning will be issued to the diesel engine to be warned.

7. The diesel engine pollutant emission early warning method according to claim 1, characterized in that, After determining whether to issue a pollutant emission warning to the diesel engine to be warned based on the predicted pollutant emissions, the process further includes: Extract all simulation values ​​from the second set of simulation values ​​and determine the cluster centers; Determine the cluster distance from each simulated value to the cluster center, and perform curve fitting on all cluster distances based on a preset method to obtain a cluster distance fitting curve; Determine the slope of the clustering distance fitting curve, and based on the slope-period mapping table, determine the next warning time for the diesel engine to be warned; Identify the unfitted cluster distances and calculate the average cluster distance of all unfitted cluster distances; The warning cycle adjustment coefficient of the diesel engine to be warned is set based on the average clustering distance, and the next warning time is adjusted based on the warning cycle adjustment coefficient to obtain the next target warning time; Based on the next target warning time, periodic pollutant emission warnings are issued for the diesel engine to be warned.

Images