Heat supply system fault diagnosis method based on artificial intelligence technology and Internet of Things technology

Abstract

The invention discloses a heat supply system fault diagnosis method based on an artificial intelligence technology and an internet of things technology, and relates to the field of heat supply system fault diagnosis. The method aims at solving the problem that an existing heat supply system fault detection method is low in fault detection accuracy. The method comprises the steps of inputting a to-be-diagnosed heat supply equipment image into a heat supply equipment identification model, and obtaining a target detection frame position of the to-be-diagnosed heat supply equipment and a heat supply equipment type; obtaining a thermal image of the to-be-diagnosed heat supply equipment; mapping the position of the heating equipment in the image of the to-be-diagnosed heating equipment to the thermal image of the to-be-diagnosed heating equipment, and obtaining the center coordinate of a target detection frame in the thermal image, the ratio of the width of the target detection frame in the thermal image to the overall width of the thermal image, and the ratio of the height of the target detection frame in the thermal image to the overall height of the thermal image; and then judging whether the to-be-diagnosed heat supply equipment is abnormal or not by using the average temperature in the range of the target detection frame in the thermal image, if so, alarming, otherwise, ending the detection. The method is used for heat supply system fault detection.

Description

Technical Field

[0001] The present invention relates to the field of heating system fault diagnosis, and in particular to a heating system fault diagnosis method based on artificial intelligence technology and Internet of Things technology. Background Art

[0002] Global climate change and energy crisis urge countries to accelerate energy transformation, reduce dependence on fossil fuels, and promote the application of sustainable energy. In order to achieve these goals, the country has implemented energy-saving measures in many fields such as industry, construction, and transportation. As an important part of energy consumption, the heating system, especially in the cold northern regions, supervises its safe and normal operation and prevents energy waste, which has become an important energy-saving task. The heating system is composed of a variety of equipment, each of which plays an important role in the heating process. Core equipment such as boilers, radiators, heat exchangers and circulation pumps ensure the generation and transfer of heat, while control equipment such as thermostats and pressure controllers ensure the efficient and stable operation of the system under different working conditions. The heating system is highly interconnected, and any change in equipment will cause fluctuations in the overall system, especially the system reaction caused by temperature changes. Too high or too low temperature will have a negative impact on the operation, safety, energy efficiency and user comfort of the system. Too high temperature can easily lead to equipment damage, energy waste and safety hazards, while too low temperature may cause problems such as insufficient heating, system icing and corrosion. Therefore, maintaining the temperature of the heating system within a reasonable range and running stably can not only ensure the comfort of the living environment, but also improve system efficiency, prevent energy waste, and extend equipment life.

[0003] At present, heating system fault detection mainly relies on real-time monitoring by sensors, but the sensors require regular maintenance and calibration. At the same time, environmental changes may cause sensor inaccuracies, resulting in low accuracy of heating system fault detection. Summary of the invention

[0004] The purpose of the present invention is to solve the problem of low fault detection accuracy in existing heating system fault detection methods, and propose a heating system fault diagnosis method based on artificial intelligence technology and Internet of Things technology.

[0005] A method for fault diagnosis of a heating system based on artificial intelligence technology and Internet of Things technology. The specific process is as follows:

[0006] S1. Obtain an image of the heating equipment to be diagnosed, input the image of the heating equipment to be diagnosed into a heating equipment recognition model, and obtain the position of the heating equipment target detection frame and the type of the heating equipment in the image of the heating equipment to be diagnosed;

[0007] S2. Scan the heating equipment to be diagnosed using an infrared detector to obtain a thermal image of the heating equipment to be diagnosed;

[0008] S3, mapping the position of the target detection frame of the heating equipment in the image of the heating equipment to be diagnosed obtained in S1 to the thermal image of the heating equipment to be diagnosed obtained in S3, obtaining the center coordinates of the target detection frame in the thermal image, the ratio of the width of the target detection frame in the thermal image to the overall width of the thermal image, and the ratio of the height of the target detection frame in the thermal image to the overall height of the thermal image, and then obtaining the range of the target detection frame in the thermal image;

[0009] S4. Use the average temperature in the target detection frame in the thermal image to determine whether the heating equipment to be diagnosed has any abnormality. If an abnormality occurs, an alarm is issued; otherwise, the detection is terminated directly.

[0010] Furthermore, the heating equipment identification model in S1 is obtained by:

[0011] Step 1: Obtain the original images of the heating system equipment and preprocess them. Use the preprocessed original images of the heating system equipment to form a data set, and then divide the data set into a training set, a validation set, and a test set:

[0012] Step 11: Obtaining original images of heating system equipment;

[0013] Step 12: preprocessing the original image of the heating system equipment to obtain the preprocessed original image of the heating system equipment;

[0014] Step 13: obtain the position of the target detection frame in the preprocessed original image of the heating system equipment, use the type of the heating system equipment as the label of the preprocessed original image of the heating system equipment, form a data set with the preprocessed original image of the heating system equipment, the preprocessed original image of the heating system equipment with the position of the target detection frame, and the label, and divide the data set into a training set, a validation set, and a test set;

[0015] The heating system equipment types include: boiler image, radiator image, heat exchanger image, and circulating pump image;

[0016] Step 2: Use the training set to train the equipment recognition network, use the test set and validation set to verify and test the trained equipment recognition network, and finally obtain the heating equipment recognition model.

[0017] Furthermore, the preprocessing of the original image of the heating system equipment in the steps one and two is specifically as follows:

[0018] First, data enhancement is performed on the original images of the heating system equipment;

[0019] Then, bilinear interpolation is used to scale the original image of the heating system equipment after data enhancement to obtain the preprocessed original image of the heating system equipment.

[0020] Further, the device identification network includes: a first convolution module, a second convolution module, a first C2f_add module, a second C2f_add module, a third convolution module, a third C2f_add module, a fourth C2f_add module, a fifth C2f_add module, a sixth C2f_add module, a first SCDown module, a seventh C2f_add module, an eighth C2f_add module, a ninth C2f_add module, a tenth C2f_add module, a second SCDown module, a third SCDown module, a fourth SCDown module, a fifth SCDown module, and a sixth SCDown module. Module, seventh SCDown module, eighth SCDown module, first C2fCIB_add module, SPPF module, PSA module, first Upsample module, first splicing module, first C2f module, second Upsample module, second splicing module, second C2f module, third C2f module, fourth convolution module, third splicing module, second C2fCIB_add module, third C2fCIB_add module, ninth SCDown module, fourth splicing module, third C2fCIB_add module, fourth C2fCIB_add module, header module, result output module;

[0021] The first convolution module includes: a first convolution layer, a first batch of normalization layers, and a first activation function layer; the input of the first convolution module is the preprocessed original image of the heating system equipment;

[0022] The first convolution layer is a convolution layer with a convolution kernel size of 3×3; the input of the first convolution layer is the preprocessed original image of the heating system equipment;

[0023] The input of the first batch of normalization layers is the output of the first convolutional layer;

[0024] The first activation function layer is a Silu activation function, and the input of the first activation function layer is the output of the first batch of normalization layers;

[0025] The second convolution module includes: a second convolution layer, a second batch normalization layer, and a second activation function layer; the input of the second convolution module is the output of the first convolution module;

[0026] The second convolution layer is a convolution layer with a convolution kernel size of 3×3;

[0027] The input of the second batch normalization layer is the output of the second convolutional layer;

[0028] The second activation function layer is a Silu activation function, and the input of the second activation function layer is the output of the second batch normalization layer;

[0029] The input of the first C2f_add module is the output of the second convolution module;

[0030] The input of the second C2f_add module is the output of the first C2f_add module;

[0031] The third convolution module includes: a third convolution layer, a third batch normalization layer, and a third activation function layer; the input of the third convolution module is the output of the second C2f_add module;

[0032] The third convolution layer is a convolution layer with a convolution kernel size of 3×3; the input of the third convolution layer is the output of the second C2f_add module;

[0033] The input of the third batch normalization layer is the output of the third convolutional layer;

[0034] The third activation function layer is a Silu activation function, and the input of the third activation function layer is the output of the third batch normalization layer;

[0035] The input of the third C2f_add module is the output of the 3 convolutional modules;

[0036] The input of the fourth C2f_add module is the output of the third C2f_add module;

[0037] The input of the fifth C2f_add module is the output of the fourth C2f_add module;

[0038] The input of the sixth C2f_add module is the output of the fifth C2f_add module;

[0039] The input of the first SCDown module is the output of the sixth C2f_add module;

[0040] The input of the seventh C2f_add module is the output of the first SCDown module;

[0041] The input of the eighth C2f_add module is the output of the seventh C2f_add module;

[0042] The input of the ninth C2f_add module is the output of the eighth C2f_add module;

[0043] The input of the tenth C2f_add module is the output of the ninth C2f_add module;

[0044] The input of the second SCDown module is the output of the tenth C2f_add module;

[0045] The input of the third SCDown module is the output of the second SCDown module;

[0046] The input of the fourth SCDown module is the output of the third SCDown module;

[0047] The input of the fifth SCDown module is the output of the fourth SCDown module;

[0048] The input of the sixth SCDown module is the output of the fifth SCDown module;

[0049] The input of the seventh SCDown module is the output of the sixth SCDown module;

[0050] The input of the eighth SCDown module is the output of the seventh SCDown module;

[0051] The input of the first C2fCIB_add module is the output of the eighth SCDown module;

[0052] The input of the SPPF module is the output of the first C2fCIB_add module;

[0053] The input of the PSA module is the output of the SPPF module;

[0054] The input of the first Upsample module is the output of the PSA module;

[0055] The first splicing module is used to splice the output of the first Upsample module and the output of the tenth C2f_add module to obtain a first splicing result;

[0056] The input of the first C2f module is the first splicing result;

[0057] The input of the second Upsample module is the output of the first C2f module;

[0058] The second splicing module is used to splice the output of the second Upsample module and the output of the sixth C2f_add module to obtain a second splicing result;

[0059] The input of the second C2f module is the second splicing result;

[0060] The input of the third C2f module is the output of the second C2f module; the third C2f module outputs a feature map X0;

[0061] The fourth convolution module includes: a fourth convolution layer, a fourth batch normalization layer, and a fourth activation function layer; the input of the fourth convolution module is the output of the third C2f module;

[0062] The fourth convolutional layer is a convolutional layer with a convolution kernel size of 3×3; the input of the fourth convolutional layer is the output of the third C2f module;

[0063] The fourth batch of normalization layer inputs are outputs of the fourth convolutional layer;

[0064] The fourth activation function layer is a Silu activation function, and the input of the fourth activation function layer is the output of the fourth batch normalization layer;

[0065] The third splicing module is used to splice the output of the first C2f module and the output of the fourth convolution module to obtain a third splicing result;

[0066] The input of the second C2fCIB_add module is the output of the third splicing module;

[0067] The input of the third C2fCIB_add module is the output of the second C2fCIB_add module; the third C2fCIB_add module outputs a feature map X1;

[0068] The input of the ninth SCDown module is the output of the third C2fCIB_add module;

[0069] The fourth splicing module is used to splice the output of the PSA module and the output of the ninth SCDown module to obtain a fourth splicing result;

[0070] The input of the third C2fCIB_add module is the fourth splicing result;

[0071] The fourth C2fCIB_add module input is the output of the third C2fCIB_add module; the fourth C2fCIB_add module outputs a feature map X2;

[0072] The input of the head module is feature map X0, feature map X1 and feature map X2 to obtain feature map X'0, feature map X'1 and feature map X'2;

[0073] The result output module is used to obtain the heating system equipment image and heating equipment type with the target detection box position through the processed feature map X0, the processed feature map X1 and the processed feature map X2.

[0074] Further, the head module includes: a feature map X0 processing unit, a feature map X1 processing unit, and a feature map X2 processing unit;

[0075] The feature map X0 processing unit includes: a fifth convolution subunit, a sixth convolution subunit, a seventh convolution subunit, an eighth convolution subunit, a ninth convolution subunit, a tenth convolution subunit, an eleventh convolution subunit, a twelfth convolution subunit, and a fifth concatenation subunit; the input of the feature map X0 processing unit is the feature map X0;

[0076] The fifth convolution subunit includes: a fifth convolution layer, a fifth batch normalization layer, and a fifth activation function layer;

[0077] The fifth convolution layer is a convolution layer with a convolution kernel of 3×3; the input of the fifth convolution layer is the feature map X0;

[0078] The input of the fifth batch normalization layer is the output of the fifth convolutional layer;

[0079] The fifth activation function layer is a silu activation function, and the input of the fifth activation function layer is the output of the fifth batch normalization layer;

[0080] The sixth convolution subunit is the same as the fifth convolution subunit; the input of the sixth convolution subunit is the output of the fifth convolution subunit;

[0081] The seventh convolution subunit is a convolution layer with a convolution kernel of 1×1; the input of the seventh convolution subunit is the output of the sixth convolution subunit

[0082] The 8th convolution subunit is the same as the 5th convolution subunit; the input of the 8th convolution subunit is the feature map X1;

[0083] The 9th convolution subunit includes: a 9th convolution layer, a 9th batch normalization layer, and a 9th activation function layer; the input of the 9th convolution subunit is the output of the 8th convolution subunit;

[0084] The ninth convolution layer is a convolution layer with a convolution kernel of 1×1; the input of the ninth convolution layer is the output of the eighth convolution subunit;

[0085] The input of the ninth batch normalization layer is the output of the ninth convolutional layer;

[0086] The ninth activation function layer is a silu activation function, and the input of the ninth activation function layer is the output of the ninth batch normalization layer;

[0087] The 10th convolution subunit is the same as the 5th convolution subunit; the input of the 10th convolution subunit is the output of the 9th convolution subunit;

[0088] The 11th convolution subunit is the same as the 9th convolution subunit; the input of the 11th convolution subunit is the output of the 10th convolution subunit;

[0089] The 12th convolution subunit is a convolution layer with a convolution kernel of 1×1; the input of the 12th convolution subunit is the output of the 11th convolution subunit;

[0090] The fifth concatenation subunit is used to concatenate the output of the seventh convolution subunit and the output of the twelfth convolution subunit to obtain a feature map X'0;

[0091] The structure of the feature map X1 processing unit is the same as that of the feature map X0 processing unit; the input of the feature map X1 processing unit is the feature map X1, and the output is the feature map X'1;

[0092] The structure of the feature map X2 processing unit is the same as that of the feature map X0 processing unit; the input of the feature map X2 processing unit is the feature map X2, and the output is the feature map X'2.

[0093] Furthermore, the result output module includes: an image mapping unit and a non-maximum suppression unit;

[0094] The image mapping unit is used to scale the feature map X'0, the feature map X'1 and the feature map X'2 to the same size as the original image of the heating system equipment and then map them to the original image of the heating system equipment to obtain the original image of the heating system equipment after mapping the feature map X'0, the feature map X'1 and the feature map X'2;

[0095] The non-maximum suppression unit is used to perform non-maximum suppression processing on the original image of the heating system equipment after mapping the feature map X'0, the feature map X'1 and the feature map X'2 to obtain the heating system equipment image with the target detection frame position and the heating equipment type.

[0096] Furthermore, the center coordinates of the target detection frame in the thermal image in S3 are obtained by:

[0097]

[0098] Among them, (x heat ,y heat ) is the center coordinate of the target detection box in the thermal image, (x noraml ,y normal ) is the center coordinate of the heating equipment target detection box in the heating equipment image to be diagnosed, W heat is the thermal image width, W heat is the thermal image height, W normal is the image width of the heating equipment to be diagnosed, H normal It is the image height of the heating equipment to be diagnosed.

[0099] Furthermore, the ratio of the width of the target detection frame in the thermal image to the overall width of the thermal image is obtained by:

[0100]

[0101] Among them, w heat is the ratio of the width of the target detection box in the thermal image to the overall width of the thermal image, w normal It is the ratio of the width of the heating equipment target detection frame in the heating equipment image to be diagnosed to the width of the heating equipment image to be diagnosed.

[0102] Furthermore, the ratio of the height of the target detection frame in the thermal image to the overall height of the thermal image is specifically:

[0103]

[0104] Among them, h heat is the ratio of the height of the target detection box in the thermal image to the overall height of the thermal image, h normal It is the ratio of the height of the heating equipment target detection frame in the heating equipment image to be diagnosed to the height of the heating equipment image to be diagnosed.

[0105] Furthermore, the step S4 uses the average temperature in the target detection frame in the thermal image to determine whether the heating equipment to be diagnosed is abnormal. If abnormal, an alarm is issued; otherwise, the detection is terminated directly. Specifically,

[0106] S401, obtaining the average temperature in the range of the target detection frame in the thermal image, specifically:

[0107]

[0108] Among them, T aver is the average temperature within the range of the target detection box in the thermal image, T j is the temperature value of the jth pixel in the range of the target detection frame in the thermal image, M is the total number of pixels in the range of the target detection frame in the thermal image, and j is the pixel number in the range of the target detection frame in the thermal image;

[0109] S402: judging whether the heating equipment to be diagnosed is abnormal according to the type of heating equipment and the average temperature in the range of the target detection frame in the thermal image; if abnormal, giving an alarm; otherwise, ending the detection, specifically:

[0110] If the heating equipment type is a boiler, when 70℃≤T aver ≤90℃ means that the boiler has no heating failure and the test is ended; if T aver <70℃, it means the boiler temperature is too low, and a low boiler temperature alarm is issued; if T aver >90℃, it means the boiler temperature is too high and a boiler temperature alarm is issued;

[0111] If the heating equipment type is a radiator, when 60℃≤T aver ≤80℃ indicates that the radiator has no heating failure and the test ends; if T aver <60℃, it means the radiator temperature is too low, and a low radiator temperature alarm is issued; if T aver >80℃, it means the radiator temperature is too high and a radiator temperature alarm is issued;

[0112] If the heating equipment type is a heat exchanger, when 50℃≤T aver ≤90℃ indicates that the heat exchanger has no heating failure and the test is terminated; if T aver <50℃, it means the heat exchanger temperature is too low, and a heat exchanger low temperature alarm is issued; if T aver >90℃, it means the heat exchanger temperature is too high and a heat exchanger temperature alarm is issued;

[0113] If the heating equipment type is a circulating pump, when 30℃≤T aver ≤60℃ indicates that the circulation pump has no heating failure and the test ends; if T aver <30℃, it means the circulating pump temperature is too low, and a circulating pump low temperature alarm is issued; if T aver >60℃, it means the circulation pump temperature is too high and a circulation pump high temperature alarm is issued.

[0114] The beneficial effects of the present invention are:

[0115] The present invention uses a heating equipment identification model to detect the heating system, and then combines thermal imaging to obtain the current equipment temperature of the heating system, so as to determine whether the heating equipment in the heating system has a fault, thereby improving the detection speed and accuracy of heating equipment faults in the heating system. The present invention performs different fault warnings for the four types of heating equipment in the heating system, namely the boiler, radiator, heat exchanger, and circulating pump, and can make targeted heating adjustments according to the faults of different heating equipment. After detecting the target heating equipment, the present invention combines thermal technology to analyze the equipment temperature in real time and give corresponding warnings, thereby reducing energy waste and extending the life of the equipment. The present invention monitors the abnormality of the heating equipment and can promptly handle or repair the abnormality of the heating equipment, avoiding problems such as heating cessation due to sudden equipment failure. BRIEF DESCRIPTION OF THE DRAWINGS

[0116] Figure 1 The following is the overall flow chart;

[0117] Figure 2(a) shows the boiler image;

[0118] Figure 2(b) shows the radiator image;

[0119] Figure 2(c) is an image of the heat exchanger;

[0120] Figure 2(d) is the image of the circulation pump;

[0121] Figure 3(a) is a diagram of the C2f_add module structure;

[0122] Figure 3(b) is a structural diagram of the SCDown module;

[0123] Figure 3(c) is a diagram of the C2fCIB_add module structure;

[0124] Figure 3(d) is a diagram of the CIB module structure;

[0125] Figure 3(e) is a block diagram of the SPPF module;

[0126] Figure 3(f) is a structural diagram of the PSA module;

[0127] Figure 4(a) is a diagram of the overall network structure;

[0128] Figure 4(b) is a structural diagram of the head module. DETAILED DESCRIPTION

[0129] Specific implementation method 1: Figure 1 As shown, the specific process of the heating system fault diagnosis method based on artificial intelligence technology and Internet of Things technology in this embodiment is as follows:

[0130] S1. Obtain an image of the heating equipment to be diagnosed, input the image of the heating equipment to be diagnosed into a heating equipment recognition model, and obtain the position of the heating equipment target detection frame and the type of the heating equipment in the image of the heating equipment to be diagnosed;

[0131] The heating equipment identification model is obtained in the following way:

[0132] Step 1: Obtain the original images of the heating system equipment and preprocess them. Use the preprocessed original images of the heating system equipment to form a data set, and then divide the data set into a training set, a validation set, and a test set:

[0133] Step 11: Get the original image of the heating system equipment:

[0134] The categories of the original images of the heating equipment include: boiler images, radiator images, heat exchanger images, and circulating pump images;

[0135] The original image resolution of the heating equipment is 1080×810. The image data comes from the School of Civil Engineering of Heilongjiang University. There are 4,000 images in total, 1,000 for each type of equipment. The equipment examples are as follows: Figure 2(a)-Figure 2(d) shown.

[0136] Step 1 and 2: preprocessing the original image of the heating system equipment;

[0137] First, data enhancement is performed on the original image of the heating system equipment, including: 1) random horizontal or vertical flipping, 2) random cropping, and 3) Mosaic data enhancement;

[0138] The data augmentation method in this step reduces the risk of subsequent network overfitting.

[0139] Then, the original image of the heating system equipment after data enhancement is scaled by bilinear interpolation to obtain the original image of the heating system equipment after preprocessing;

[0140] The size of the preprocessed original image of the heating system equipment is 640×480×3;

[0141] Step 13: obtain the position of the target detection frame in the preprocessed original image of the heating system equipment, set a label for the preprocessed original image of the heating system equipment, form a data set with the preprocessed original image of the heating system equipment, the preprocessed original image of the heating system equipment with the position of the target detection frame, and the label, and divide the data set into a training set, a validation set, and a test set in a ratio of 6:2:2;

[0142] The label is the category of the original image of the heating system equipment;

[0143] Step 2: Use the training set to train the device recognition network to obtain a trained device recognition network;

[0144] As shown in Figure 4(a), the fault detection network includes: a first convolution module, a second convolution module, a first C2f_add module, a second C2f_add module, a third convolution module, a third C2f_add module, a fourth C2f_add module, a fifth C2f_add module, a sixth C2f_add module, a first SCDown module, a seventh C2f_add module, an eighth C2f_add module, a ninth C2f_add module, a tenth C2f_add module, a second SCDown module, a third SCDown module, a fourth SCDown module, a fifth SCDown module, and a sixth SCDo wn module, the seventh SCDown module, the eighth SCDown module, the first C2fCIB_add module, the SPPF module, the PSA module, the first Upsample module, the first splicing module, the first C2f module, the second Upsample module, the second splicing module, the second C2f module, the third C2f module, the fourth convolution module, the third splicing module, the second C2f module, the third C2fCIB_add module, the ninth SCDown module, the fourth splicing module, the third C2fCIB_add module, the fourth C2fCIB_add module, the header module and the result output module;

[0145] The first convolution module includes: a first convolution layer, a first batch of normalization layers, and a first activation function layer; the first convolution module outputs a preprocessed original image of the heating system equipment with a size of 320×240×48;

[0146] The first convolution layer is a convolution layer with a convolution kernel size of 3×3, a stride of 2, and a padding of 1, and the input of the first convolution layer is the preprocessed original image of the heating system equipment;

[0147] The input of the first batch of normalization layers is the output of the first convolutional layer;

[0148] The first activation function layer is a Silu activation function, and the input of the first activation function layer is the output of the first batch of normalization layers;

[0149] The second convolution module includes: a second convolution layer, a second batch normalization layer, and a second activation function layer; the output of the second convolution module is a feature map of a size of 160×120×96; the input of the second convolution module is the output of the first convolution module;

[0150] The second convolutional layer is a convolutional layer with a convolution kernel size of 3×3, a stride of 2, and a padding of 1;

[0151] The input of the second batch normalization layer is the output of the second convolutional layer;

[0152] The second activation function layer is a Silu activation function, and the input of the second activation function layer is the output of the second batch normalization layer;

[0153] The input of the first C2f_add module is the output of the second convolution module;

[0154] The input of the second C2f_add module is the output of the first C2f_add module, and the output feature map is 160×120×96 in size;

[0155] The third convolution module includes: a third convolution layer, a third batch normalization layer, and a third activation function layer; the third convolution module outputs a feature map with a size of 80×60×192;

[0156] The third convolutional layer is a convolutional layer with a convolution kernel size of 3×3, a stride of 2, and a padding of 1, and the input is the output of the second C2f_add module;

[0157] The input of the third batch normalization layer is the output of the third convolutional layer;

[0158] The third activation function layer is a Silu activation function, and the input of the third activation function layer is the output of the third batch normalization layer;

[0159] The third C2f_add module is the same as the first C2f_add module, and the input of the third C2f_add module is the output of the 3 convolution modules;

[0160] The fourth C2f_add module is the same as the first C2f_add module, and the input of the fourth C2f_add module is the output of the third C2f_add module;

[0161] The fifth C2f_add module is the same as the first C2f_add module, and the input of the fifth C2f_add module is the output of the fourth C2f_add module;

[0162] The sixth C2f_add module is the same as the first C2f_add module, and the input of the sixth C2f_add module is the output of the fifth C2f_add module; the sixth C2f_add module outputs a feature map with a specification of 80×60×192;

[0163] The input of the first SCDown module is the output of the sixth C2f_add module, and the first SCDown module outputs a feature map with a specification of 40×30×384;

[0164] The seventh C2f_add module is the same as the first C2f_add module, and its input is the output of the first SCDown module;

[0165] The eighth C2f_add module is the same as the first C2f_add module, and the input is the output of the seventh C2f_add module;

[0166] The ninth C2f_add module is the same as the first C2f_add module, and its input is the output of the eighth C2f_add module;

[0167] The tenth C2f_add module is the same as the first C2f_add module, and its input is the output of the ninth C2f_add module. The tenth C2f_add module outputs a feature map X1 with a specification of 40×30×384;

[0168] The input of the second SCDown module is the output of the tenth C2f_add module;

[0169] The input of the third SCDown module is the output of the second SCDown module;

[0170] The input of the fourth SCDown module is the output of the third SCDown module;

[0171] The input of the fifth SCDown module is the output of the fourth SCDown module;

[0172] The input of the sixth SCDown module is the output of the fifth SCDown module;

[0173] The input of the seventh SCDown module is the output of the sixth SCDown module;

[0174] The input of the eighth SCDown module is the output of the seventh SCDown module, and the output specification of the eighth SCDown module is a feature map of 20×15×576;

[0175] The input of the first C2fCIB_add module is the output of the eighth SCDown module, and the output specification is a feature map of 20×15×576;

[0176] The input of the SPPF module is the output of the first C2fCIB_add module, and the output specification is a feature map of 20×15×576;

[0177] The input of the PSA module is the output of the SPPF module, and the output specification is a feature map of 20×15×576;

[0178] The input of the first Upsample module is the output of the PSA module, and the output specification is a feature map of 40×30×576;

[0179] The first splicing module is used to splice the output of the first Upsample module and the output of the tenth C2f_add module to obtain a first splicing result; the first splicing result is a feature map with a specification of 40×30×960;

[0180] The input of the first C2f module is the first splicing result; there is no residual connection layer in the first C2f module, and the rest of the structure is the same as the first C2f_add module structure;

[0181] The input of the second Upsample module is the output of the first C2f module; the second Upsample module outputs a feature map with a specification of 80×60×384;

[0182] The second splicing module is used to splice the output of the second Upsample module and the output of the sixth C2f_add module to obtain a second splicing result; the second splicing result is a feature map with a specification of 80×60×576;

[0183] The input of the second C2f module is the second splicing result;

[0184] The input of the third C2f module is the output of the second C2f module, and the third C2f module outputs a feature map X0 with a specification of 80×60×192;

[0185] The fourth convolution module includes: a fourth convolution layer, a fourth batch normalization layer, and a fourth activation function layer; the fourth convolution module outputs a feature map with a specification of 40×30×192;

[0186] The fourth convolution layer is a convolution layer with a convolution kernel size of 3×3, a stride of 2, and a padding of 1. The input of the fourth convolution module is the output of the third C2f module.

[0187] The fourth batch of normalization layer inputs are outputs of the fourth convolutional layer;

[0188] The fourth activation function layer is a Silu activation function, and the input of the fourth activation function layer is the output of the fourth batch normalization layer;

[0189] The third splicing module is used to splice the output of the first C2f module and the output of the fourth convolution module to obtain a third splicing result; the third splicing module outputs a feature map with a specification of 40×30×576;

[0190] The input of the second C2fCIB_add module is the output of the third splicing module;

[0191] The input of the third C2fCIB_add module is the output of the second C2fCIB_add module, and the third C2fCIB_add module outputs a feature map X1 with a specification of 40×30×384;

[0192] The input of the ninth SCDown module is the output of the third C2fCIB_add module, and the ninth SCDown module outputs a feature map with a specification of 20×15×384;

[0193] The fourth splicing module is used to splice the output of the PSA module and the output of the ninth SCDown module to obtain a fourth splicing result; the fourth splicing module outputs a feature map with a specification of 20×15×960;

[0194] The input of the third C2fCIB_add module is the fourth splicing result;

[0195] The input of the fourth C2fCIB_add module is the output of the third C2fCIB_add module; the fourth C2fCIB_add module outputs a feature map X2 with a specification of 20×15×576;

[0196] As shown in FIG4( b ), the head module includes: a feature map X0 processing unit, a feature map X1 processing unit, and a feature map X2 processing unit;

[0197] The feature map X0 processing unit includes: a 5th convolution subunit, a 6th convolution subunit, a 7th convolution subunit, an 8th convolution subunit, a 9th convolution subunit, a 10th convolution subunit, an 11th convolution subunit, a 12th convolution subunit, and a 5th splicing subunit;

[0198] The fifth convolution subunit includes: a fifth convolution layer, a fifth batch normalization layer, and a fifth activation function layer;

[0199] The fifth convolutional layer is a convolutional layer with a convolution kernel of 3×3, a stride of 1, and a padding of 1; the input of the fifth convolutional layer is the feature map X0;

[0200] The input of the fifth batch normalization layer is the output of the fifth convolutional layer;

[0201] The fifth activation function layer is a silu activation function, and the input of the fifth activation function layer is the output of the fifth batch normalization layer;

[0202] The sixth convolution subunit includes: a sixth convolution layer, a sixth batch normalization layer, and a sixth activation function layer;

[0203] The sixth convolution layer is a convolution layer with a convolution kernel of 3×3, a stride of 1, and a padding of 1; the input of the sixth convolution layer is the output of the fifth convolution subunit;

[0204] The input of the sixth batch normalization layer is the output of the sixth convolutional layer;

[0205] The sixth activation function layer is a silu activation function, and the input of the sixth activation function layer is the output of the sixth batch normalization layer;

[0206] The seventh convolution subunit is a convolution layer with a convolution kernel of 1×1 and a stride of 1. The input of the seventh convolution subunit is the output of the sixth convolution subunit. The seventh convolution subunit outputs a feature map with a specification of 80×60×64.

[0207] The 8th convolution subunit includes: the 8th convolution layer, the 8th batch normalization layer, and the 8th activation function layer;

[0208] The eighth convolutional layer is a convolutional layer with a convolution kernel of 3×3, a stride of 1, and a padding of 1; the input of the eighth convolutional layer is the feature map X1;

[0209] The input of the eighth batch normalization layer is the output of the eighth convolutional layer;

[0210] The eighth activation function layer is a silu activation function, and the input of the eighth activation function layer is the output of the eighth batch normalization layer;

[0211] The ninth convolution subunit includes: a ninth convolution layer, a ninth batch normalization layer, and a ninth activation function layer;

[0212] The ninth convolution layer is a convolution layer with a convolution kernel of 1×1 and a stride of 1; the input of the ninth convolution layer is the output of the eighth convolution subunit;

[0213] The input of the ninth batch normalization layer is the output of the ninth convolutional layer;

[0214] The ninth activation function layer is a silu activation function, and the input of the ninth activation function layer is the output of the ninth batch normalization layer;

[0215] The tenth convolution subunit includes: a tenth convolution layer, a tenth batch normalization layer, and a tenth activation function layer;

[0216] The tenth convolutional layer is a convolutional layer with a convolution kernel of 3×3, a stride of 1, and a padding of 1; the input of the tenth convolutional layer is the output of the ninth convolutional subunit;

[0217] The input of the tenth batch normalization layer is the output of the tenth convolutional layer;

[0218] The tenth activation function layer is a silu activation function, and the input of the tenth activation function layer is the output of the tenth batch normalization layer;

[0219] The 11th convolution subunit includes: an 11th convolution layer, an 11th batch normalization layer, and an 11th activation function layer;

[0220] The eleventh convolutional layer is a convolutional layer with a convolution kernel of 1×1 and a stride of 1; the input of the eleventh convolutional layer is the output of the tenth convolutional subunit;

[0221] The input of the eleventh batch of normalization layers is the output of the eleventh convolutional layer;

[0222] The eleventh activation function layer is a silu activation function, and the input of the eleventh activation function layer is the output of the eleventh normalization layer;

[0223] The 12th convolution subunit is a convolution layer with a convolution kernel of 1×1 and a stride of 1. The input of the 12th convolution subunit is the output of the 11th convolution subunit. The 12th convolution subunit outputs a feature map with a specification of 80×60×80.

[0224] The fifth concatenation subunit is used to concatenate the output of the seventh convolution subunit and the output of the twelfth convolution subunit to obtain a feature map X'0; the feature map X'0 is a feature map with a specification of 80×60×144.

[0225] The structure of the feature map X1 processing unit is the same as that of the feature map X0 processing unit; the input of the feature map X1 processing unit is the feature map X1, and the output is the feature map X'1; the feature map X'1 is a feature map with a size of 40×30×144;

[0226] The structure of the feature map X2 processing unit is the same as that of the feature map X0 processing unit; the input of the feature map X2 processing unit is the feature map X2, and the output is the feature map X'2; the feature map X'2 is a feature map with a specification of 20×15×144.

[0227] The result output module includes: an image mapping unit and a non-maximum suppression unit;

[0228] The image mapping unit is used to scale the feature map X'0, the feature map X'1 and the feature map X'2 to the same size as the original image of the heating system equipment and then map them to the original image of the heating system equipment to obtain the original image of the heating system equipment after mapping the feature map X'0, the feature map X'1 and the feature map X'2;

[0229] The non-maximum suppression unit is used to perform non-maximum suppression processing on the original image of the heating system equipment after mapping the feature map X'0, the feature map X'1 and the feature map X'2 to obtain the heating system equipment image with the target detection frame position and the heating equipment type.

[0230] The C2f_add module, SCDowm module, C2fCIB_add module, SPPF module, and PSA module mentioned in step 2 are existing modules, specifically:

[0231] As shown in FIG3( a ), the C2f_add module in this step includes: a first CBS unit, a split unit, n Bottleneck units, a channel fusion unit, and a second CBS unit;

[0232] The first CBS unit and the second CBS unit each include a 1*1 convolution layer, a batch normalization layer, and an activation function layer;

[0233] The output of the first CBS unit is the input of the split unit. The split unit divides the output of the first CBS unit into two feature maps along the channel dimension. The input of the first Bottleneck unit is one of the feature maps output by the split unit. The output of the first Bottleneck unit is the input of the second Bottleneck unit, until the output of the n-1th Bottleneck unit is input to the nth Bottleneck unit. The outputs of n Bottleneck units are all input to the channel fusion unit. The input of the channel fusion unit is another feature map output by the split unit and the output of n Bottleneck units. The output of the channel fusion unit is input to the second CBS unit.

[0234] As shown in FIG3( b ), the SCDowm module in this step includes a CBS unit and a CB unit;

[0235] The CB unit includes a 3*3 convolutional layer and a batch normalization layer;

[0236] As shown in FIG3( c ), the C2fCIB_add module includes: a first CBS unit, a Split unit, a first CIB unit, a second CIB unit, a feature connection unit, and a second CBS unit;

[0237] The output of the first CBS unit is the input of the Split unit. The Split unit splits the output of the first CBS unit into two feature maps P0 and P1, P1 is input to the first CIB unit, the output of the first CIB unit is input to the second CIB unit and the feature connection unit, the input of the second CIB unit is input to the feature connection unit; the feature map P0 is input to the feature connection unit; the feature connection unit splices the three feature maps and inputs them to the second CBS unit;

[0238] The first CBS unit and the second CIB unit each include a 1*1 convolution layer, a batch normalization layer, and an activation function layer;

[0239] As shown in FIG3( d ), the CIB unit includes: a first CBS subunit, a second CBS subunit, a third CBS subunit, a fourth CBS subunit, a fifth CBS subunit, and a feature addition subunit;

[0240] The first CBS subunit, the third CBS subunit, and the fifth CBS subunit each include a 3*3 convolution layer, a batch normalization layer, and an activation function layer;

[0241] The second CBS subunit and the fourth CBS subunit each include a 1*1 convolution layer, a batch normalization layer, and an activation function layer;

[0242] The output of the first CBS subunit in the CIB unit is the input of the second CBS subunit, the output of the second CBS subunit is the input of the third CBS subunit, the input of the third CBS subunit is the input of the fourth CBS subunit, the output of the fourth CBS subunit is the input of the fifth CBS subunit, and the output of the fifth CBS subunit and the input of the first CBS subunit are the inputs of the feature addition subunit.

[0243] As shown in FIG3( e ), the SPPF module includes: a first CBS unit, a first maximum pooling layer unit, a second maximum pooling layer unit, a third maximum pooling layer unit, a feature connection unit, and a second CBS unit;

[0244] The first CBS unit and the second CBS unit each include a 1*1 convolution layer, a batch normalization layer, and an activation function layer;

[0245] The output of the first CBS unit is the input of the first maximum pooling layer unit and the feature connection unit, the output of the first maximum pooling layer unit is the input of the second maximum pooling layer unit and the feature connection unit, the output of the second maximum pooling layer unit is the input of the third maximum pooling layer and the feature connection unit; the output of the second maximum pooling layer unit is the input of the feature connection unit; the feature connection unit splices the outputs of the first CBS unit, the first maximum pooling layer unit, the second maximum pooling layer unit, and the third maximum pooling layer unit according to the channel dimension, and inputs the splicing result to the second CBS unit.

[0246] As shown in Figure 3(f), the PSA module includes: a first CBS unit, a Spilt unit, an attention unit, a first feature addition unit, a second CBS unit, a CB unit, a second feature addition unit, a feature connection unit, and a third CBS unit;

[0247] The first CBS unit, the second CBS unit, and the third CBS unit all include a 1*1 convolution layer, a batch normalization layer, and an activation function layer;

[0248] The output of the first CBS unit is the input of the Spilt unit. The Spilt unit divides the output of the first CBS unit into feature maps M0 and M1. M1 is input to the attention unit and the first feature addition unit. The output of the attention unit is input to the first feature addition unit. The first feature addition unit adds the feature map M1 and the output feature map of the attention unit element by element. The output of the first feature addition unit is input to the second CBS unit and the second feature addition unit. The output of the second CBS unit is used as the input of the CB unit. The output of the CB unit is used as the input of the second feature addition unit. The second feature addition unit adds the output of the CB unit and the output of the first feature addition unit element by element, and sends the addition result to the feature connection unit. The feature map M0 is input to the feature connection unit. The feature connection unit splices the output of the second feature addition unit and the feature map M0 according to the channel dimension, and sends the splicing result to the third CBS unit.

[0249] The labels of the CBS unit, CB unit, feature concatenation unit, feature addition unit, and Split unit in each module introduced here are all the labels in the modules to which they belong, and are not the same processing unit.

[0250] Step 3: Use the test set and validation set to test and verify the trained equipment identification network. If the accuracy of the trained equipment identification network is greater than the preset accuracy threshold, the currently trained fault detection network is used as the heating equipment identification model; otherwise, return to step 1.

[0251] Thermal imaging technology is a detection method that uses infrared detectors to capture infrared radiation energy (i.e., heat) emitted by objects or human bodies and converts it into electrical signals. These electrical signals generate visual thermal images through image processing systems, showing areas of different temperatures. This technology originates from a basic physical law: any object with a temperature above absolute zero will emit infrared radiation, and the higher the temperature, the stronger the radiation. Thermal imaging equipment measures the surface temperature of an object by capturing these infrared radiations and converting them into images. This means that thermal imaging technology does not rely on external light sources, and can still perform effective detection and imaging even in completely dark or low-light environments. It also has the advantages of no contact, high sensitivity, and the ability to achieve real-time detection. Therefore, the use of thermal imaging technology can well reduce labor costs and save energy. Any object with temperature will emit infrared radiation outward, and its radiation intensity can be described by Planck's Law. Therefore, the present invention uses an infrared detector to scan the heating equipment to be diagnosed to obtain a thermal image of the heating equipment to be diagnosed.

[0252] S2. Scan the heating equipment to be diagnosed using an infrared detector to obtain a thermal image of the heating equipment to be diagnosed, specifically:

[0253] S201, obtaining the total radiation intensity corresponding to each pixel in the image of the heating equipment to be diagnosed, and obtaining the electrical signal corresponding to each pixel of the image of the heating equipment to be diagnosed output by the detector based on the total radiation intensity corresponding to each pixel of the image of the heating equipment to be diagnosed, specifically:

[0254] S201-1. Scan the heating equipment to be diagnosed using an infrared detector to obtain the total radiation intensity E of each pixel in the image of the heating equipment to be diagnosed. I i R (T);

[0255] The infrared detector scans the heating equipment to be diagnosed to obtain the total radiation intensity E of the heating equipment to be diagnosed I i R (T) Specifically:

[0256] First, the radiation intensity of the i-th pixel point corresponding to the image of the heating equipment to be diagnosed at different wavelengths is obtained as follows:

[0257]

[0258] Among them, E i (λ,T) is the total radiation intensity of the i-th pixel in the image of the heating equipment to be diagnosed at wavelength λ, T iis the absolute temperature of the i-th pixel in the image of the heating equipment to be diagnosed, λ is the infrared wavelength, 8μm≤λ≤14μm, and h is the Planck constant, which is 6.626×10 -34 J·s, c is the speed of light, c is 3×10 8 m / s, k is the Boltzmann constant, k is 1.381×10 - 23 J / K;

[0259] The total radiation intensity of the corresponding i-th pixel point in the image of the heating equipment to be diagnosed is obtained by integrating the radiation intensity of the corresponding i-th pixel point in the image of the heating equipment to be diagnosed at different wavelengths, specifically:

[0260]

[0261] Among them, λ max is the minimum wavelength in the infrared band, λ min It is the maximum wavelength in the infrared band;

[0262] S201-2, using the total radiation intensity E of the corresponding i-th pixel point in the image of the heating equipment to be diagnosed IR (T i ) Get the output electrical signal V of the detector i :

[0263] V i = k'·E IR (T i )

[0264] Where k' is the sensitivity coefficient of the infrared detector.

[0265] S202, the output electrical signal V i Amplify, filter and correct the processed electrical signal Using processed electrical signals Obtain thermal images of the heating equipment to be diagnosed, specifically:

[0266] S202-1, the output electrical signal V of the infrared detector i Amplify, filter and correct the processed electrical signal

[0267] Signal amplification: Use a low noise amplifier (LNA) to amplify the electrical signal V i ;

[0268] Signal filtering: Use a low-pass filter to remove the electrical signal V after signal amplification i High frequency noise in

[0269] Signal correction: Use blackbody calibration technology to correct the filtered electrical signal to obtain the processed electrical signal

[0270] A blackbody source of known temperature calibrates the sensor's response to obtain accurate radiation intensity;

[0271] S202-2. Using the processed electrical signal Get the temperature value of the heating equipment to be diagnosed corresponding to the i-th pixel in the image of the heating equipment to be diagnosed:

[0272]

[0273] Where k* is the proportional factor of the electrical signal into temperature, and ε represents the emissivity;

[0274] ∈=1: It means that the heating equipment to be diagnosed is a perfect black body, which can emit radiation with maximum efficiency at all wavelengths.

[0275] ∈<1: Indicates that the emissivity of the heating equipment to be diagnosed is lower than that of a perfect black body. The surface characteristics, smoothness, temperature and wavelength of different materials are the main factors affecting the emissivity.

[0276] S203, color mapping is performed using the temperature values ​​of the heating equipment to be diagnosed corresponding to all pixels in the image of the heating equipment to be diagnosed, to obtain a thermal image of the heating equipment to be diagnosed:

[0277] The temperature value of the heating equipment to be diagnosed corresponding to the i-th pixel in the image of the heating equipment to be diagnosed is color mapped, specifically:

[0278]

[0279] in, yes The mapped color, yes Color mapping function.

[0280] Each pixel of the thermal imager corresponds to a temperature value or radiation intensity value, that is, the thermal imager collects temperature data for each point in the entire image.

[0281] The color mapping in this step usually corresponds to red or yellow for higher temperatures and blue or green for lower temperatures; this can enhance the visualization of temperature differences.

[0282] The specific processing process of the color mapping function is:

[0283] Normalize the temperature to be mapped to the range of [0,1] using the following formula:

[0284]

[0285] in, yes The normalized value, T max is the maximum value of the temperature of the heating equipment to be diagnosed corresponding to the pixel in the image of the heating equipment to be diagnosed, T min is the minimum value of the temperature of the heating equipment to be diagnosed corresponding to the pixels in the image of the heating equipment to be diagnosed;

[0286] For each normalized value Determine its position in the color gradient. Use linear interpolation to make the color transition smoothly. The formula is as follows:

[0287]

[0288] Among them, (R0, G0, B0) and (R1, G1, B1) are two adjacent color nodes, and (R, G, B) is the linear interpolation between them.

[0289] The (R, G, B) obtained by interpolation is combined into the final color, and the temperature value of each pixel is converted into the corresponding color to generate the final thermal image. The thermal image can be displayed on a monitor or stored in a computer for subsequent analysis.

[0290] S3, mapping the position of the heating equipment in the image of the heating equipment to be diagnosed obtained in S1 to the thermal image of the heating equipment to be diagnosed obtained in S3, obtaining the center coordinates of the target detection frame in the thermal image, the ratio of the width of the target detection frame in the thermal image to the overall width of the thermal image, and the ratio of the height of the target detection frame in the thermal image to the overall height of the thermal image, and then obtaining the range of the target detection frame in the thermal image;

[0291] The specific coordinates of the center of the target detection frame in the thermal image are:

[0292]

[0293] Among them, (x heat ,y heat ) is the center coordinate of the target detection box in the thermal image, (x noraml ,y normal ) is the center coordinate of the heating equipment target detection box in the heating equipment image to be diagnosed, W heat is the thermal image width, W heat is the thermal image height, W normal is the image width of the heating equipment to be diagnosed, H normal is the image height of the heating equipment to be diagnosed;

[0294] The ratio of the width of the target detection frame in the thermal image to the overall width of the thermal image is obtained by:

[0295]

[0296] Among them, w heat is the ratio of the width of the target detection box in the thermal image to the overall width of the thermal image, w normal is the ratio of the width of the heating equipment target detection box in the heating equipment image to be diagnosed to the width of the heating equipment image to be diagnosed, W heat is the thermal image width, W normal is the image width of the heating equipment to be diagnosed;

[0297] The ratio of the height of the target detection frame in the thermal image to the overall height of the thermal image is obtained by:

[0298]

[0299] Among them, h heat is the ratio of the height of the target detection box in the thermal image to the overall height of the thermal image, h normal H is the ratio of the height of the heating equipment target detection frame in the heating equipment image to be diagnosed to the height of the heating equipment image to be diagnosed. heat is the thermal image height, H normal It is the height and width of the image of the heating equipment to be diagnosed.

[0300] S4, using the average temperature in the target detection frame in the thermal image to determine whether the heating equipment to be diagnosed is abnormal, if abnormal, an alarm is issued, otherwise, the process ends, specifically:

[0301] S401, obtaining the average temperature in the range of the target detection frame in the thermal image, specifically:

[0302]

[0303] Among them, T aver is the average temperature within the range of the target detection box in the thermal image, T j is the temperature value of the jth pixel in the range of the target detection box in the thermal image, M is the total number of pixels in the range of the target detection box in the thermal image, and j is the pixel label in the range of the target detection box in the thermal image.

[0304] S402: judging whether the heating equipment to be diagnosed is abnormal according to the type of heating equipment obtained in S1 and the average temperature in the range of the target detection frame in the thermal image; if abnormal, an alarm is given; otherwise, the detection is terminated, specifically:

[0305] If the heating equipment type is a boiler, when 70℃≤T aver ≤90℃ means that the boiler has no heating failure. aver <70℃, it means the boiler temperature is too low, and a low boiler temperature alarm is issued; if Taver >90℃, it means the boiler temperature is too high and a boiler temperature alarm is issued;

[0306] If the heating equipment type is a radiator, when 60℃≤T aver ≤80℃ means that the radiator has no heating failure. aver <60℃, it means the radiator temperature is too low, and a low radiator temperature alarm is issued; if T aver >80℃, it means the radiator temperature is too high and a radiator temperature alarm is issued;

[0307] If the heating equipment type is a heat exchanger, when 50℃≤T aver ≤90℃ means that the heat exchanger has no heating failure. aver <50℃, it means the heat exchanger temperature is too low, and a heat exchanger low temperature alarm is issued; if T aver >90℃, it means the heat exchanger temperature is too high and a heat exchanger temperature alarm is issued;

[0308] If the heating equipment type is a circulating pump, when 30℃≤T aver ≤60℃ means that the circulation pump has no heating failure. aver <30℃, it means the circulating pump temperature is too low, and a circulating pump low temperature alarm is issued; if T aver >60℃, it means the circulating pump temperature is too high, and a circulating pump over-temperature alarm is issued;

[0309] For the early warning prompts of the network, different treatment measures need to be taken according to different types of equipment. When the boiler temperature is too high, the fuel input should be reduced, the heating capacity should be reduced, the cooling water flow should be injected in time and the safety valve should be checked to ensure its normal operation and prevent the risk of explosion; when it is too low, the fuel input should be increased to improve the heating capacity. When the radiator temperature is too high, the heat source output should be reduced (such as closing the valve), the thermostat setting should be checked to ensure that the temperature parameters are reasonable, and there are no obstacles around the radiator that hinder the heat dissipation; when it is too low, the heat source output should be increased (such as opening the valve), the air blockage in the system should be checked, the water flow should be smooth, and the thermostat should be checked to ensure its normal operation. When the heat exchanger temperature is too high, adjust the cooling water flow to ensure sufficient cooling, check for blockage, ensure smooth flow, check the cleanliness of the heat exchanger, and clean it if necessary; when it is too low, increase the fluid flow or adjust the heating source output, check the insulation measures, and ensure that heat is not lost. When the temperature of the circulating pump is too high, check the flow setting of the pump to ensure that the flow is moderate, check the cooling system to ensure that the pump has good heat dissipation and its operating frequency is within a reasonable range; when it is too low, increase the flow of the pump or adjust the operating frequency, check for blockage, and ensure smooth flow.

Claims

1. A heating system fault diagnosis method based on artificial intelligence technology and Internet of Things technology, characterized in that The specific process of the method is: S1. Obtain an image of the heating equipment to be diagnosed, input the image of the heating equipment to be diagnosed into a heating equipment recognition model, and obtain the position of the heating equipment target detection frame and the type of the heating equipment in the image of the heating equipment to be diagnosed; S2. Scan the heating equipment to be diagnosed using an infrared detector to obtain a thermal image of the heating equipment to be diagnosed; S3, mapping the position of the target detection frame of the heating equipment in the image of the heating equipment to be diagnosed obtained in S1 to the thermal image of the heating equipment to be diagnosed obtained in S3, obtaining the center coordinates of the target detection frame in the thermal image, the ratio of the width of the target detection frame in the thermal image to the overall width of the thermal image, and the ratio of the height of the target detection frame in the thermal image to the overall height of the thermal image, and then obtaining the range of the target detection frame in the thermal image; S4. Use the average temperature in the target detection frame in the thermal image to determine whether the heating equipment to be diagnosed has any abnormality. If an abnormality occurs, an alarm is issued; otherwise, the detection is terminated directly.

2. According to claim 1, a heating system fault diagnosis method based on artificial intelligence technology and Internet of Things technology is characterized by: The heating equipment identification model in S1 is obtained by: Step 1: Obtain the original images of the heating system equipment and preprocess them. Use the preprocessed original images of the heating system equipment to form a data set, and then divide the data set into a training set, a validation set, and a test set: Step 11: Obtaining original images of heating system equipment; Step 12: preprocessing the original image of the heating system equipment to obtain the preprocessed original image of the heating system equipment; Step 13: obtain the position of the target detection frame in the preprocessed original image of the heating system equipment, use the type of the heating system equipment as the label of the preprocessed original image of the heating system equipment, form a data set with the preprocessed original image of the heating system equipment, the preprocessed original image of the heating system equipment with the position of the target detection frame, and the label, and divide the data set into a training set, a validation set, and a test set; The heating system equipment types include: boiler image, radiator image, heat exchanger image, and circulating pump image; Step 2: Use the training set to train the equipment recognition network, use the test set and validation set to verify and test the trained equipment recognition network, and finally obtain the heating equipment recognition model.

3. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 2 is characterized in that: The preprocessing of the original image of the heating system equipment in the steps 1 and 2 is specifically as follows: First, data enhancement is performed on the original images of the heating system equipment; Then, bilinear interpolation is used to scale the original image of the heating system equipment after data enhancement to obtain the preprocessed original image of the heating system equipment.

4. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 3 is characterized in that: The device identification network includes: a first convolution module, a second convolution module, a first C2f_add module, a second C2f_add module, a third convolution module, a third C2f_add module, a fourth C2f_add module, a fifth C2f_add module, a sixth C2f_add module, a first SCDown module, a seventh C2f_add module, an eighth C2f_add module, a ninth C2f_add module, a tenth C2f_add module, a second SCDown module, a third SCDown module, a fourth SCDown module, a fifth SCDown module, a sixth SCDown module, The seventh SCDown module, the eighth SCDown module, the first C2fCIB_add module, the SPPF module, the PSA module, the first Upsample module, the first splicing module, the first C2f module, the second Upsample module, the second splicing module, the second C2f module, the third C2f module, the fourth convolution module, the third splicing module, the second C2fCIB_add module, the third C2fCIB_add module, the ninth SCDown module, the fourth splicing module, the third C2fCIB_add module, the fourth C2fCIB_add module, the header module, and the result output module; The first convolution module includes: a first convolution layer, a first batch of normalization layers, and a first activation function layer; the input of the first convolution module is the preprocessed original image of the heating system equipment; The first convolution layer is a convolution layer with a convolution kernel size of 3×3; the input of the first convolution layer is the preprocessed original image of the heating system equipment; The input of the first batch of normalization layers is the output of the first convolutional layer; The first activation function layer is a Silu activation function, and the input of the first activation function layer is the output of the first batch of normalization layers; The second convolution module includes: a second convolution layer, a second batch normalization layer, and a second activation function layer; the input of the second convolution module is the output of the first convolution module; The second convolution layer is a convolution layer with a convolution kernel size of 3×3; The input of the second batch normalization layer is the output of the second convolutional layer; The second activation function layer is a Silu activation function, and the input of the second activation function layer is the output of the second batch normalization layer; The input of the first C2f_add module is the output of the second convolution module; The input of the second C2f_add module is the output of the first C2f_add module; The third convolution module includes: a third convolution layer, a third batch normalization layer, and a third activation function layer; the input of the third convolution module is the output of the second C2f_add module; The third convolution layer is a convolution layer with a convolution kernel size of 3×3; the input of the third convolution layer is the output of the second C2f_add module; The input of the third batch normalization layer is the output of the third convolutional layer; The third activation function layer is a Silu activation function, and the input of the third activation function layer is the output of the third batch normalization layer; The input of the third C2f_add module is the output of the 3 convolutional modules; The input of the fourth C2f_add module is the output of the third C2f_add module; The input of the fifth C2f_add module is the output of the fourth C2f_add module; The input of the sixth C2f_add module is the output of the fifth C2f_add module; The input of the first SCDown module is the output of the sixth C2f_add module; The input of the seventh C2f_add module is the output of the first SCDown module; The input of the eighth C2f_add module is the output of the seventh C2f_add module; The input of the ninth C2f_add module is the output of the eighth C2f_add module; The input of the tenth C2f_add module is the output of the ninth C2f_add module; The input of the second SCDown module is the output of the tenth C2f_add module; The input of the third SCDown module is the output of the second SCDown module; The input of the fourth SCDown module is the output of the third SCDown module; The input of the fifth SCDown module is the output of the fourth SCDown module; The input of the sixth SCDown module is the output of the fifth SCDown module; The input of the seventh SCDown module is the output of the sixth SCDown module; The input of the eighth SCDown module is the output of the seventh SCDown module; The input of the first C2fCIB_add module is the output of the eighth SCDown module; The input of the SPPF module is the output of the first C2fCIB_add module; The input of the PSA module is the output of the SPPF module; The input of the first Upsample module is the output of the PSA module; The first splicing module is used to splice the output of the first Upsample module and the output of the tenth C2f_add module to obtain a first splicing result; The input of the first C2f module is the first splicing result; The input of the second Upsample module is the output of the first C2f module; The second splicing module is used to splice the output of the second Upsample module and the output of the sixth C2f_add module to obtain a second splicing result; The input of the second C2f module is the second splicing result; The input of the third C2f module is the output of the second C2f module; the third C2f module outputs a feature map X0; The fourth convolution module includes: a fourth convolution layer, a fourth batch normalization layer, and a fourth activation function layer; the input of the fourth convolution module is the output of the third C2f module; The fourth convolutional layer is a convolutional layer with a convolution kernel size of 3×3; the input of the fourth convolutional layer is the output of the third C2f module; The fourth batch of normalization layer inputs are outputs of the fourth convolutional layer; The fourth activation function layer is a Silu activation function, and the input of the fourth activation function layer is the output of the fourth batch normalization layer; The third splicing module is used to splice the output of the first C2f module and the output of the fourth convolution module to obtain a third splicing result; The input of the second C2fCIB_add module is the output of the third splicing module; The input of the third C2fCIB_add module is the output of the second C2fCIB_add module; the third C2fCIB_add module outputs a feature map X1; The input of the ninth SCDown module is the output of the third C2fCIB_add module; The fourth splicing module is used to splice the output of the PSA module and the output of the ninth SCDown module to obtain a fourth splicing result; The input of the third C2fCIB_add module is the fourth splicing result; The fourth C2fCIB_add module input is the output of the third C2fCIB_add module; the fourth C2fCIB_add module outputs a feature map X2; The input of the head module is feature map X0, feature map X1 and feature map X2 to obtain feature map X'0, feature map X'1 and feature map X'2; The result output module is used to obtain the heating system equipment image and heating equipment type with the target detection box position through the processed feature map X0, the processed feature map X1 and the processed feature map X2.

5. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 4 is characterized in that: The head module includes: a feature map X0 processing unit, a feature map X1 processing unit, and a feature map X2 processing unit; The feature map X0 processing unit includes: a fifth convolution subunit, a sixth convolution subunit, a seventh convolution subunit, an eighth convolution subunit, a ninth convolution subunit, a tenth convolution subunit, an eleventh convolution subunit, a twelfth convolution subunit, and a fifth concatenation subunit; the input of the feature map X0 processing unit is the feature map X0; The fifth convolution subunit includes: a fifth convolution layer, a fifth batch normalization layer, and a fifth activation function layer; The fifth convolution layer is a convolution layer with a convolution kernel of 3×3; the input of the fifth convolution layer is the feature map X0; The input of the fifth batch normalization layer is the output of the fifth convolutional layer; The fifth activation function layer is a silu activation function, and the input of the fifth activation function layer is the output of the fifth batch normalization layer; The sixth convolution subunit is the same as the fifth convolution subunit; the input of the sixth convolution subunit is the output of the fifth convolution subunit; The seventh convolution subunit is a convolution layer with a convolution kernel of 1×1; the input of the seventh convolution subunit is the output of the sixth convolution subunit The 8th convolution subunit is the same as the 5th convolution subunit; the input of the 8th convolution subunit is the feature map X1; The 9th convolution subunit includes: a 9th convolution layer, a 9th batch normalization layer, and a 9th activation function layer; the input of the 9th convolution subunit is the output of the 8th convolution subunit; The ninth convolution layer is a convolution layer with a convolution kernel of 1×1; the input of the ninth convolution layer is the output of the eighth convolution subunit; The input of the ninth batch normalization layer is the output of the ninth convolutional layer; The ninth activation function layer is a silu activation function, and the input of the ninth activation function layer is the output of the ninth batch normalization layer; The 10th convolution subunit is the same as the 5th convolution subunit; the input of the 10th convolution subunit is the output of the 9th convolution subunit; The 11th convolution subunit is the same as the 9th convolution subunit; the input of the 11th convolution subunit is the output of the 10th convolution subunit; The 12th convolution subunit is a convolution layer with a convolution kernel of 1×1; the input of the 12th convolution subunit is the output of the 11th convolution subunit; The fifth concatenation subunit is used to concatenate the output of the seventh convolution subunit and the output of the twelfth convolution subunit to obtain a feature map X'0; The structure of the feature map X1 processing unit is the same as that of the feature map X0 processing unit; the input of the feature map X1 processing unit is the feature map X1, and the output is the feature map X'1; The structure of the feature map X2 processing unit is the same as that of the feature map X0 processing unit; the input of the feature map X2 processing unit is the feature map X2, and the output is the feature map X'2.

6. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 5 is characterized in that: The result output module includes: an image mapping unit and a non-maximum suppression unit; The image mapping unit is used to scale the feature map X'0, the feature map X'1 and the feature map X'2 to the same size as the original image of the heating system equipment and then map them to the original image of the heating system equipment to obtain the original image of the heating system equipment after mapping the feature map X'0, the feature map X'1 and the feature map X'2; The non-maximum suppression unit is used to perform non-maximum suppression processing on the original image of the heating system equipment after mapping the feature map X'0, the feature map X'1 and the feature map X'2 to obtain the heating system equipment image with the target detection frame position and the heating equipment type.

7. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 6 is characterized in that: The center coordinates of the target detection frame in the thermal image in S3 are obtained by: Among them, (x heat ,y heat ) is the center coordinate of the target detection box in the thermal image, (x noraml ,y normal ) is the center coordinate of the heating equipment target detection box in the heating equipment image to be diagnosed, W heat is the thermal image width, W heat is the thermal image height, W normal is the image width of the heating equipment to be diagnosed, H normal It is the image height of the heating equipment to be diagnosed.

8. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 7 is characterized in that: The ratio of the width of the target detection frame in the thermal image to the overall width of the thermal image is obtained by: Among them, w heat is the ratio of the width of the target detection box in the thermal image to the overall width of the thermal image, w normal It is the ratio of the width of the heating equipment target detection frame in the heating equipment image to be diagnosed to the width of the heating equipment image to be diagnosed.

9. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 8 is characterized in that: The ratio of the height of the target detection frame in the thermal image to the overall height of the thermal image is specifically: Among them, h heat is the ratio of the height of the target detection box in the thermal image to the overall height of the thermal image, h normal It is the ratio of the height of the heating equipment target detection frame in the heating equipment image to be diagnosed to the height of the heating equipment image to be diagnosed.

10. The method for diagnosing heating system faults based on artificial intelligence technology and Internet of Things technology according to claim 9 is characterized in that: In S4, the average temperature in the target detection frame in the thermal image is used to determine whether the heating equipment to be diagnosed is abnormal. If abnormal, an alarm is issued. Otherwise, the detection is terminated directly. Specifically, S401, obtaining the average temperature in the range of the target detection frame in the thermal image, specifically: Among them, T aver is the average temperature within the range of the target detection box in the thermal image, T j is the temperature value of the jth pixel in the range of the target detection frame in the thermal image, M is the total number of pixels in the range of the target detection frame in the thermal image, and j is the pixel number in the range of the target detection frame in the thermal image; S402: judging whether the heating equipment to be diagnosed is abnormal according to the type of heating equipment and the average temperature in the range of the target detection frame in the thermal image; if abnormal, giving an alarm; otherwise, ending the detection, specifically: If the heating equipment type is a boiler, when 70℃≤T aver ≤90℃ means that the boiler has no heating failure and the test is ended; if T aver <70℃, it means the boiler temperature is too low, and a low boiler temperature alarm is issued; if T aver >90℃, it means the boiler temperature is too high and a boiler temperature alarm is issued; If the heating equipment type is a radiator, when 60℃≤T aver ≤80℃ indicates that the radiator has no heating failure and the test ends; if T aver <60℃, it means the radiator temperature is too low, and a low radiator temperature alarm is issued; if T aver >80℃, it means the radiator temperature is too high and a radiator temperature alarm is issued; If the heating equipment type is a heat exchanger, when 50℃≤T aver ≤90℃ indicates that the heat exchanger has no heating failure and the test is terminated; if T aver <50℃, it means the heat exchanger temperature is too low, and a heat exchanger low temperature alarm is issued; if T aver >90℃, it means the heat exchanger temperature is too high and a heat exchanger temperature alarm is issued; If the heating equipment type is a circulating pump, when 30℃≤T aver ≤60℃ indicates that the circulation pump has no heating failure and the test ends; if T aver <30℃, it means the circulating pump temperature is too low, and a circulating pump low temperature alarm is issued; if T aver >60℃, it means the circulation pump temperature is too high and a circulation pump high temperature alarm is issued.

Images