Water body odor feature recognition method and system

Abstract

The invention provides a water body odor feature recognition method and system, and relates to the technical field of water body odor recognition, and the specific steps are as follows: collecting sample feature data of an odor water body in a historical time period, and combining with data of a current to-be-detected water body, including volatile organic compounds, humic acid, sulfide concentration and other chemical components, to obtain an odor feature of the water body; the method comprises the following steps: extracting an odor characteristic matrix through historical data according to environmental parameters such as environmental temperature and water flow velocity, constructing the odor characteristic matrix of a current to-be-detected water body by using an odor level label training model determined by an expert, obtaining the odor level, and performing correlation analysis on the environmental and chemical parameters to obtain the odor level of the current to-be-detected water body. Generating an environmental influence coefficient and a chemical influence coefficient, calculating a comprehensive scoring coefficient, and evaluating the odor grade change condition of the to-be-detected water body. According to the method, the recognition precision of the water body odor characteristics can be remarkably improved, the odor grade change of the to-be-detected water body can be effectively pre-judged, the potential water quality problem is recognized, and measures are taken in time for intervention.

Description

Technical Field

[0001] The present invention relates to the technical field of water body odor recognition, and specifically to a method and system for recognizing water body odor characteristics. Background Art

[0002] Water body odor characteristic recognition refers to determining the odor condition of the water body and its possible pollution degree by analyzing the chemical components and physical properties in the water body sample. Water body odor usually refers to the odor experienced by people when perceiving the water body, and this odor may originate from components such as volatile organic compounds, humic acids, and sulfides present in the water body. Water body odor characteristic recognition is of great significance. It not only helps to monitor water quality but also provides a scientific basis for water resource management. By identifying the odor characteristics of the water body, relevant departments can take timely measures to control pollution sources and ensure the safety of drinking water and ecological environment protection.

[0003] In the prior art, a method for recognizing algal source odor characteristics with the publication number CN116502130A includes the following steps: obtaining data on algal source odor-causing organic substances and mass spectrometry data; obtaining the MACCS organic molecule fingerprints corresponding to the data on algal source odor-causing organic substances and the MACCS molecule fingerprints corresponding to the mass spectrometry data; training different machine learning models respectively with the MACCS organic molecule fingerprints corresponding to the training set in the algal source odor-causing organic substance odor category data to obtain the optimal odor classification model; training different machine learning models respectively with the MACCS organic molecule fingerprints corresponding to the training set in the odor threshold data to obtain the optimal odor threshold prediction model; inputting the data on the algal source odor-causing organic substances to be measured into the model, and the odor recognition result of the algal source odor-causing organic substances can be output. This method has the characteristics of low cost, simplicity, rapidity, and saving a large amount of manpower, material resources, and financial resources.

[0004] However, there are still the following deficiencies. From the above statements, it can be seen that the prior art focuses on the recognition of algal source odor-causing organic substances and their odor characteristics, without considering environmental parameters (such as temperature, water flow rate, rainfall, etc.) and chemical parameters (such as chemical oxygen demand, biochemical oxygen demand, etc.). And the odor of the water body is not only affected by chemical components but also significantly affected by environmental conditions, which reduces the accuracy of the recognition result. There is also only a focus on the current odor characteristics of the water body, lacking a prediction of the odor change trend, resulting in the inability to give early warnings for pollution incidents in advance.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for recognizing water body odor characteristics to solve the problems raised in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for identifying the odor characteristics of water bodies, the specific steps include:

[0009] S1. Collect the sample characteristic data of the odoriferous water bodies within T historical time periods, and collect the sample characteristic data of the water body to be detected at the current moment. The sample characteristic data includes the concentration of volatile organic compounds, the concentration of humic acid, the concentration of sulfides, the pH value of the water body, and the dissolved oxygen content. And collect the environmental parameters and chemical parameters that affect the condition of the water body to be detected at the current moment. The environmental parameters include environmental temperature, water flow rate, rainfall, light intensity, and wind speed. The chemical parameters include chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity;

[0010] S2. Extract the odor characteristic matrix of the odoriferous water bodies from the sample characteristic data of the odoriferous water bodies within T historical time periods, including the average concentration of volatile organic compounds, the average concentration of humic acid, the average concentration of sulfides, the average pH value of the water body, and the average dissolved oxygen content;

[0011] S3. Use the odor characteristic matrix of the odoriferous water bodies as the input and the odor level as the label to train the model. The odor level is determined by an expert group. Train the odor level model. The odor level is 1 or 2 or 3. 1 represents a slightly odoriferous water body, 2 represents a moderately odoriferous water body, and 3 represents a severely odoriferous water body;

[0012] S4. Assign different water body coefficients to slightly odoriferous water bodies, moderately odoriferous water bodies, and severely odoriferous water bodies respectively;

[0013] S5. Use the sample characteristic data of the water body to be detected at the current moment to construct the odor characteristic matrix of the water body to be detected, and input the odor characteristic matrix of the water body to be detected into the trained odor level model to obtain the odor level of the water body to be detected at the current moment;

[0014] S6. Perform data processing and correlation analysis on the environmental temperature, water flow rate, rainfall, light intensity, and wind speed that affect the water body to be detected at the current moment to generate an environmental impact coefficient for evaluating the deterioration trend of the environmental parameters on the water body to be detected. Perform data processing and correlation analysis on the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity that affect the water body to be detected at the current moment to generate a chemical impact coefficient for evaluating the deterioration trend of the chemical parameters on the water body to be detected. Perform data processing on the environmental impact coefficient and chemical impact coefficient of the water body to be detected at the current moment to generate a comprehensive scoring coefficient of the water body to be detected at the current moment;

[0015] S7. Evaluate the change of the odor level of the water body to be detected according to the comprehensive scoring coefficient, odor level, and water system coefficient of the water body to be detected at the current moment.

[0016] Furthermore, obtain the average concentration of volatile organic compounds, average concentration of humic acid, average concentration of sulfide, average water body pH value, and average dissolved oxygen content. The formulas are as follows:

[0017]

[0018]

[0019] Among them, μ hy 、μ fz 、μ lh 、μ ss 、μ ry are the average concentration of volatile organic compounds, average concentration of humic acid, average concentration of sulfide, average water body pH value, and average dissolved oxygen content respectively. hy i 、fz i 、lh i 、ss i 、ry i are the concentration of volatile organic compounds, concentration of humic acid, concentration of sulfide, water body pH value, and dissolved oxygen content in the i-th historical time period respectively. i is the index of the historical time period, and i ∈ [1, T].

[0020] Furthermore, construct the odor feature matrices of the odor water body and the water body to be detected. The specific process is as follows:

[0021]

[0022] Among them, X is the odor feature matrix of the odor water body;

[0023] Use the sample feature data of the water body to be detected at the current moment to construct the odor feature matrix of the water body to be detected:

[0024]

[0025] Among them, d is the odor feature matrix of the water body to be detected, μ hy ’、μ fz ’、μ lh ’、μ ss ’、μ ry ’ are the average concentration of volatile organic compounds, average concentration of humic acid, average concentration of sulfide, average water body pH value, and average dissolved oxygen content of the water body to be detected respectively.

[0026] Further, the odor level model is composed of a deep neural network based on a multi-layer perceptron. The deep neural network of the multi-layer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer, and an output layer. The first hidden layer, the second hidden layer, and the third hidden layer each have at least two neurons and all use ReLU as the activation function;

[0027] The process of training the environmental impact model is as follows:

[0028] Based on the odor characteristic matrix of the odoriferous water body in T historical time periods, the corresponding odor level is determined by the expert scoring method. The higher the odor level, the higher the degree of water body deterioration;

[0029] Using the odor characteristic matrix of the odoriferous water body in T historical time periods as the input quantity and the odor level as the output label for training, and using the mean square error as the loss function. When the mean square error is within the range of [0, 0.01], the training of the odor level model is completed.

[0030] Further, different water body coefficients are assigned to slightly odoriferous water bodies, moderately odoriferous water bodies, and severely odoriferous water bodies. The specific process is as follows:

[0031] The odor levels are divided into 1, 2, or 3. 1 represents a slightly odoriferous water body, 2 represents a moderately odoriferous water body, and 3 represents a severely odoriferous water body;

[0032] The water body coefficient of the slightly odoriferous water body is assigned a value of 0.3, the water body coefficient of the moderately odoriferous water body is assigned a value of 0.5, and the water body coefficient of the severely odoriferous water body is assigned a value of 0.7.

[0033] Further, the temperature, water flow rate, rainfall, light intensity, and wind speed of the water body to be detected at the current moment are processed and analyzed for correlation to generate an environmental impact coefficient for evaluating the deterioration trend of the environmental parameters on the water body to be detected. The formula is as follows:

[0034]

[0035] Among them, HJxs is the environmental impact coefficient at the current moment, WD is the environmental temperature, GZ is the light intensity, FS is the wind speed, SL is the water flow rate, JY is the rainfall, ω 1 is the weight coefficient of the environmental temperature, ω 2 is the weight coefficient of the light intensity, ω 3 is the weight coefficient of the wind speed, ω 4 is the weight coefficient of the combination of the water flow rate and rainfall. On the basis of ω 1 + ω 2 + ω 3 + ω 4 = 1, let 0 < ω 3 < ω2 <ω 1 <ω 4 <1。

[0036] Furthermore, the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity of the water body to be detected at the current moment are processed and analyzed for correlation to generate a chemical influence coefficient for evaluating the deterioration trend of the chemical parameters on the water body to be detected. The formula is as follows:

[0037] HXxs = β 1 (HY×SY) ZD +β 2 (NN×LN)

[0038] where HXxs is the chemical influence coefficient at the current moment, HY is the chemical oxygen demand, SY is the biochemical oxygen demand, NN is the nitrogen concentration, LN is the phosphorus concentration, ZD is the turbidity, and β 1 is the weight coefficient of the combination of chemical oxygen demand and biochemical oxygen demand, and β 2 is the weight coefficient of the combination of nitrogen concentration and phosphorus concentration. On the basis of β 1 +β 2 = 1, let 0 < β 2 < β 1 < 1.

[0039] Furthermore, the environmental influence coefficient and chemical influence coefficient of the water body to be detected at the current moment are processed to generate a comprehensive scoring coefficient of the water body to be detected at the current moment. The formula is as follows:

[0040] PFxs = γ 1 HJxs + γ 2 HXxs

[0041] where PFxs is the comprehensive scoring coefficient of the water body to be detected at the current moment, and γ 1 is the weight coefficient of the environmental influence coefficient, and γ 2 is the weight coefficient of the chemical influence coefficient. The specific values of γ 1 and γ 2 are determined by the analytic hierarchy process.

[0042] Furthermore, according to the comprehensive scoring coefficient, odor level, and water system coefficient of the water body to be detected at the current moment, the change of the odor level of the water body to be detected is evaluated. The specific process is as follows:

[0043] When the comprehensive scoring coefficient is less than the water system coefficient, that is, PFxs < SZxs, it means that the water quality score of the water body to be detected is lower than the standard associated with the current odor level, indicating that the water body has improved and the odor characteristics have become better;

[0044] The odor level of the water body to be detected at the current moment is above moderate, and the odor level is reduced by one level.

[0045] The odor level of the water body to be detected at the current moment is mild, and the odor level is maintained at mild.

[0046] When the comprehensive score coefficient is equal to the water body coefficient, that is, PFxs = SZxs, it means that the water quality score of the water body to be detected is equal to the standard associated with the current odor level, indicating that the water body condition remains unchanged and the odor characteristics remain unchanged.

[0047] Keep the odor level of the water body to be detected at the current moment unchanged.

[0048] When the comprehensive score coefficient is greater than the water body coefficient, that is, PFxs > SZxs, it means that the water quality score of the water body to be detected is higher than the standard associated with the current odor level, indicating that the water body deteriorates and the odor characteristics become worse.

[0049] The odor level of the water body to be detected at the current moment is below severe, and the odor level is increased by one level.

[0050] The odor level of the water body to be detected at the current moment is severe, and the odor level is maintained at severe.

[0051] A water body odor characteristic recognition system, which is used to execute any one of the above-mentioned water body odor characteristic recognition methods, including:

[0052] A data acquisition module, which is used to collect sample characteristic data of odor water bodies within T historical time periods, and collect sample characteristic data of the water body to be detected at the current moment. The sample characteristic data includes volatile organic compound concentration, humic acid concentration, sulfide concentration, water body pH value, and dissolved oxygen content, and collect environmental parameters and chemical parameters that affect the condition of the water body to be detected at the current moment. The environmental parameters include environmental temperature, water flow rate, rainfall, light intensity, and wind speed, and the chemical parameters include chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity.

[0053] A matrix construction module, which is used to extract the odor characteristic matrix of the odor water body from the sample characteristic data of the odor water body within T historical time periods, including average volatile organic compound concentration, average humic acid concentration, average sulfide concentration, average water body pH value, and average dissolved oxygen content.

[0054] An odor model construction module, which is used to use the odor characteristic matrix of the odor water body as input and the odor level as a label to train the model. The odor level is determined by an expert group, and the odor level model is trained. The odor level is 1 or 2 or 3, where 1 represents a mild odor water body, 2 represents a moderate odor water body, and 3 represents a severe odor water body.

[0055] An assignment module for assigning different water system coefficients to slightly odorous water bodies, moderately odorous water bodies, and severely odorous water bodies respectively;

[0056] A water body to be detected recognition module for using the sample feature data of the water body to be detected at the current moment to construct an odor feature matrix of the water body to be detected, inputting the odor feature matrix of the water body to be detected into the trained odor level model, and obtaining the odor level of the water body to be detected at the current moment;

[0057] A data processing and analysis module for processing and performing correlation analysis on the environmental temperature, water flow rate, rainfall, light intensity, and wind speed that affect the water body to be detected at the current moment to generate an environmental impact coefficient for evaluating the deterioration trend of the environmental parameters on the water body to be detected, processing and performing correlation analysis on the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity that affect the water body to be detected at the current moment to generate a chemical impact coefficient for evaluating the deterioration trend of the chemical parameters on the water body to be detected, and processing the environmental impact coefficient and chemical impact coefficient of the water body to be detected at the current moment to generate a comprehensive scoring coefficient of the water body to be detected at the current moment;

[0058] A water body evaluation module for evaluating the change in the odor level of the water body to be detected according to the comprehensive scoring coefficient, odor level, and water system coefficient of the water body to be detected at the current moment.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] By introducing environmental and chemical parameters, the present invention can more comprehensively understand the odor change of water bodies. This method ensures that when identifying odor characteristics, it not only depends on the chemical composition of the water body but also considers the influence of the external environment on water quality, significantly improving the identification accuracy of water body odor characteristics; by constructing an odor feature matrix of the water body to be detected, inputting the odor feature matrix of the water body to be detected into the trained odor level model, obtaining the odor level of the water body to be detected at the current moment, and based on the odor level, comprehensive scoring coefficient, and water system coefficient, it can predict the change in the odor level of the water body to be detected, identify potential water quality problems, and take measures for intervention in a timely manner. Brief Description of the Drawings

[0061] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0062] Figure 2 It is a block diagram of the module composition of the present invention. Detailed Embodiments

[0063] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the following further elaborates on the present invention in conjunction with specific embodiments.

[0064] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the ordinary meanings understood by those of ordinary skill in the field to which the present invention pertains. The "first", "second" and similar terms used in the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "comprising" or "including" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0065] Example 1:

[0066] Please refer to Figure 1 , the present invention provides a technical solution:

[0067] A method for identifying the odor characteristics of water bodies, the specific steps include:

[0068] S1. Collect the sample characteristic data of odoriferous water bodies within T historical time periods, and collect the sample characteristic data of the water body to be detected at the current moment. The sample characteristic data includes the concentration of volatile organic compounds, the concentration of humic acid, the concentration of sulfides, the pH value of the water body, and the dissolved oxygen content. And collect the environmental parameters and chemical parameters that affect the condition of the water body to be detected at the current moment. The environmental parameters include environmental temperature, water flow rate, rainfall, light intensity, and wind speed. The chemical parameters include chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity;

[0069] S2. Extract the odor characteristic matrix of the odoriferous water bodies from the sample characteristic data of the odoriferous water bodies within T historical time periods, including the average concentration of volatile organic compounds, the average concentration of humic acid, the average concentration of sulfides, the average pH value of the water body, and the average dissolved oxygen content;

[0070] S3. Use the odor characteristic matrix of the odoriferous water bodies as the input and the odor level as the label to train the model. The odor level is determined by the expert group. Train the odor level model. The odor level is 1 or 2 or 3. 1 represents a slightly odoriferous water body, 2 represents a moderately odoriferous water body, and 3 represents a severely odoriferous water body;

[0071] S4. Assign different water body coefficients to the slightly odoriferous water body, the moderately odoriferous water body, and the severely odoriferous water body respectively;

[0072] S5. Using the sample feature data of the water body to be detected at the current moment, construct an odor feature matrix of the water body to be detected, and input the odor feature matrix of the water body to be detected into the trained odor level model to obtain the odor level of the water body to be detected at the current moment;

[0073] S6. Perform data processing and correlation analysis on the environmental temperature, water flow rate, rainfall, light intensity, and wind speed that affect the water body to be detected at the current moment to generate an environmental impact coefficient for evaluating the deterioration trend of the water body to be detected affected by environmental parameters. Perform data processing and correlation analysis on the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity that affect the water body to be detected at the current moment to generate a chemical impact coefficient for evaluating the deterioration trend of the water body to be detected affected by chemical parameters. Perform data processing on the environmental impact coefficient and chemical impact coefficient of the water body to be detected at the current moment to generate a comprehensive scoring coefficient of the water body to be detected at the current moment;

[0074] S7. Evaluate the change in the odor level of the water body to be detected based on the comprehensive scoring coefficient, odor level, and water system coefficient of the water body to be detected at the current moment.

[0075] Based on the above embodiments, the collection method and equipment of the sample feature data are as follows:

[0076] Extract a certain amount of water sample from the water body sample. In the laboratory, use a gas chromatograph to analyze the VOCs concentration in the sample;

[0077] Filter the sample of the odoriferous water body to remove suspended matter, add a specific reagent, react the humic acid with the reagent to form a colored compound, and use a spectrophotometer to measure the absorbance of the sample at a specific wavelength to calculate the concentration of humic acid;

[0078] Extract a certain amount of water sample from the water body sample and add a reagent, usually a lead ion solution. The reacted sample is measured for absorbance by a spectrophotometer, and the concentration of sulfide is calculated according to the standard curve;

[0079] Immerse the electrode of the pH meter into the water body sample, wait for the reading to stabilize, and record the pH value of the water body;

[0080] Immerse the dissolved oxygen detector into the water body sample, ensure that the electrode or sensor is completely submerged, and read and record the dissolved oxygen content.

[0081] Based on the above embodiments, the collection method and equipment of the environmental parameters of the water body to be detected are as follows:

[0082] Immerse the thermometer probe into the water body sample, wait for the reading to stabilize, and record the temperature value;

[0083] Place the flowmeter in a water sample, ensure that the device is in line with the water flow direction, read and record the water flow velocity;

[0084] Install the rain gauge in an open area, avoid obstruction, and regularly read and record the rainfall;

[0085] Place the light intensity meter on the surface of the water sample or near the water surface, read and record the light intensity value;

[0086] Place the anemometer in an unobstructed place, ensure its normal operation, read and record the wind speed value.

[0087] On the basis of the above embodiments, the collection methods and devices for the chemical parameters of the water body to be detected are as follows:

[0088] In a water sample, add an oxidant (such as potassium dichromate) and acid, react under heating conditions, and measure the absorbance of the sample by a spectrophotometer after the reaction to calculate the chemical oxygen demand;

[0089] Put the water sample into a biochemical oxygen demand bottle, seal it, and culture it in a constant temperature incubator for 5 days. Then, measure the dissolved oxygen content again, and calculate the biochemical oxygen demand according to the difference in DO before and after cultivation;

[0090] Add a specific reagent to the water sample to form a measurable colored compound of nitrogen and phosphorus, use a spectrophotometer to measure the absorbance, and calculate the concentrations of nitrogen and phosphorus;

[0091] Place the water sample in a turbidimeter and read the turbidity measurement value.

[0092] Among them, the data acquisition module includes a gas chromatograph, a pH meter, a dissolved oxygen detector, a thermometer, a flowmeter, a rain gauge, a light intensity meter, an anemometer, an optical dissolved oxygen analyzer, and a turbidimeter, which are respectively used to collect the concentrations of VOCs, pH value, dissolved oxygen content, temperature, water flow velocity, rainfall, light intensity, wind speed, biochemical oxygen demand, and turbidity. The spectrophotometer can be used to collect the concentrations of humic acid, chemical oxygen demand, nitrogen concentration, and phosphorus concentration. The above-mentioned collection devices can all adopt the models in existing devices, and no restrictions are imposed here.

[0093] The gas chromatograph, pH meter, dissolved oxygen detector, spectrophotometer, thermometer, flowmeter, rain gauge, light intensity meter, anemometer, optical dissolved oxygen analyzer, and turbidimeter are all in multiple groups (such as 3 groups), and are placed in the water sample to measure the concentrations of VOCs, humic acid concentration, sulfide concentration, water body pH value, dissolved oxygen content, temperature, water flow velocity, rainfall, light intensity, wind speed, chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity. Then, average the same data collected multiple times, and use the finally obtained average value as the corresponding data to avoid accidental errors in single-point sampling.

[0094] On the basis of the above embodiments, the average concentration of volatile organic compounds, the average concentration of humic acid, the average concentration of sulfide, the average water body pH value, and the average dissolved oxygen content are obtained, and the formulas are as follows:

[0095]

[0096] Among them, μ hy 、μ fz 、μ lh 、μ ss 、μ ry are the average concentration of volatile organic compounds, the average concentration of humic acid, the average concentration of sulfide, the average water body pH value, and the average dissolved oxygen content respectively, and hy i 、fz i 、lh i 、ss i 、ry i are the concentration of volatile organic compounds, the concentration of humic acid, the concentration of sulfide, the water body pH value, and the dissolved oxygen content in the i-th historical time period respectively, i is the index of the historical time period, and i ∈ [1, T].

[0097] On the basis of the above embodiments, the odor characteristic matrix of the odor-containing water body is constructed, and the specific process is as follows:

[0098]

[0099] Among them, X is the odor characteristic matrix of the odor-containing water body.

[0100] On the basis of the above embodiments, the odor level model is composed of a deep neural network based on a multi-layer perceptron. The deep neural network of the multi-layer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer, and an output layer. The first hidden layer, the second hidden layer, and the third hidden layer all have at least two neurons, and ReLU is used as the activation function;

[0101] In this embodiment, the input features of the deep learning network of the multi-layer perceptron include: the average concentration of volatile organic compounds, the average concentration of humic acid, the average concentration of sulfide, the average water body pH value, and the average dissolved oxygen content, 5 features;

[0102] The structure of the deep learning network of the multi-layer perceptron is as follows:

[0103] Input layer: Receive the input of 5 features;

[0104] First hidden layer: Has 64 neurons and uses ReLU as the activation function;

[0105] Second hidden layer: It has 32 neurons and also uses the ReLU activation function;

[0106] Third hidden layer: It has 16 neurons and uses the ReLU activation function;

[0107] Output layer: It has a single neuron, the odor level.

[0108] The process of training the environmental impact model is as follows:

[0109] According to the odor characteristic matrix of the odor-containing water body in T historical time periods, the corresponding odor level is determined by the expert scoring method between 1 and 9. The higher the odor level, the higher the degree of water body deterioration;

[0110] Using the odor characteristic matrix of the odor-containing water body in T historical time periods as the input quantity and the odor level as the output label for training, and using the mean square error as the loss function. When the mean square error is within the range of [0, 0.01], the training of the odor level model is completed.

[0111] On the basis of the above embodiment, different water body coefficients are assigned to the slightly odoriferous water body, moderately odoriferous water body, and severely odoriferous water body. The specific process is as follows:

[0112] The odor level is divided into 1 or 2 or 3. 1 represents the slightly odoriferous water body, 2 represents the moderately odoriferous water body, and 3 represents the severely odoriferous water body;

[0113] Assign the water body coefficient of the slightly odoriferous water body as 0.3, assign the water body coefficient of the moderately odoriferous water body as 0.5, and assign the water body coefficient of the severely odoriferous water body as 0.7.

[0114] The water body coefficient is a preset standard value, which is associated with a specific odor level and represents the lowest water quality requirement acceptable for a specific odor level.

[0115] On the basis of the above embodiment, using the sample characteristic data of the water body to be detected at the current moment, an odor characteristic matrix of the water body to be detected is constructed. The specific process is as follows:

[0116]

[0117] Among them, d is the odor characteristic matrix of the water body to be detected, μ hy ’, μ fz ’, μ lh ’, μ ss ’, μ ry ’ are respectively the average concentration of volatile organic compounds, average concentration of humic acid, average concentration of sulfide, average water body pH value, and average dissolved oxygen content of the water body to be detected;

[0118] The odor characteristic matrix X of the water body to be detectedd Input it into the trained odor level model to obtain the odor level of the water body to be detected, and the odor level of the water body to be detected is a slightly odorous water body, a moderately odorous water body, or a severely odorous water body.

[0119] On the basis of the above embodiments, data processing and correlation analysis are performed on the temperature, water flow rate, rainfall, light intensity, and wind speed of the water body to be detected at the current moment to generate an environmental impact coefficient for evaluating the deterioration trend of the water body to be detected by environmental parameters. The formula is as follows:

[0120]

[0121] Among them, HJxs is the environmental impact coefficient at the current moment, WD is the environmental temperature, GZ is the light intensity, FS is the wind speed, SL is the water flow rate, JY is the rainfall, ω 1 is the weight coefficient of the environmental temperature, ω 2 is the weight coefficient of the light intensity, ω 3 is the weight coefficient of the wind speed, ω 4 is the weight coefficient of the combination of the water flow rate and the rainfall.

[0122] It should be noted that the higher the environmental temperature WD, the greater the environmental impact coefficient HJxs, and the faster the deterioration trend of the water body. The stronger the light intensity GZ, the greater the environmental impact coefficient HJxs, and the faster the deterioration trend of the water body. The faster the wind speed FS, the greater the environmental impact coefficient HJxs, and the faster the deterioration trend of the water body. The faster the water flow rate SL, the smaller the environmental impact coefficient HJxs, and the slower the deterioration trend of the water body. The greater the rainfall JY, the smaller the environmental impact coefficient HJxs, and the slower the deterioration trend of the water body. Therefore, the environmental impact coefficient HJxs is positively correlated with the environmental temperature WD, the light intensity GZ, and the wind speed FS, and the environmental impact coefficient HJxs is negatively correlated with the water flow rate SL and the rainfall JY;

[0123] The increase in rainfall JY will not only increase the water level and flow rate of the water body, but also change the water flow rate SL. During rainfall, the fluidity of the water body increases, resulting in the dispersion of pollutants, and high flow rates can dilute the pollutants brought by rainfall. When the rainfall JY is large, the change in flow rate will also affect the concentration of pollutants. The interaction between the water flow rate SL and the rainfall JY will jointly determine the water body. Therefore, the water flow rate SL and the rainfall JY are set in a multiplicative form;

[0124] Since the influence degrees of different environmental parameters on the water body are different, it is necessary to set the weight coefficients of the environmental parameters. By setting the weight coefficients ω 1 、ω 2 、ω 3 and ω 4 , the influence degrees of environmental parameters on the water body can be more accurately reflected, and the interaction between various factors can be emphasized.

[0125] In summary, the calculation formula of the environmental impact coefficient HJxs in the above form is set.

[0126] The value range of the environmental impact coefficient HJxs is [0, 1];

[0127] When HJxs is close to 0, it indicates that the environmental parameters have little impact on the deterioration trend of the water body;

[0128] When HJxs is close to 1, it indicates that the environmental parameters have a great impact on the deterioration trend of the water body;

[0129] Therefore, the larger the environmental impact coefficient HJxs, the higher the impact of environmental conditions on the deterioration trend of the water body to be detected, and the faster the water body deterioration trend.

[0130] The water flow velocity SL is a key factor affecting the self-purification ability of the water body. A higher water flow velocity SL can effectively dilute the pollutants in the water and promote their diffusion, thereby slowing down the water body deterioration. Fast water flow can enhance the fluidity of the water body, help dilute the pollutants to a lower concentration, and reduce their negative impact on the water body; rainfall will cause changes in the fluidity of the water body, thereby affecting the concentration and distribution of pollutants. Especially when the rainfall is large, the interaction between the dilution effect of the water body and the introduction of pollutants will accelerate the water body change trend. Therefore, the combined action of the water flow velocity SL and the rainfall JY will have a significant impact on the water body. The combination of the water flow velocity SL and the rainfall JY has the greatest correlation with the environmental impact coefficient HJxs, so the highest weight coefficient ω is set 4 。

[0131] The influence of temperature WD on the water body is more comprehensive, involving the improvement of biological metabolism and chemical reaction speed, so its influence degree is greater. Although the light intensity GZ has a direct impact on the growth of algae, its influence is relatively limited to the promotion of photosynthesis, and its overall impact on the water body is not as significant as that of temperature WD. Since the influence of temperature WD is significant and wide-ranging, the temperature weight coefficient ωω 1 is higher, and the light intensity weight coefficient ωω 2 is lower.

[0132] The light intensity GZ promotes the photosynthesis of algae. Under strong light conditions, algae will multiply in large numbers, causing oxygen deficiency in the water body. The wind speed FS mainly affects gas exchange and improves the dissolved oxygen level in the water, but its influence on the growth of algae is small, and under the action of water flow and rainfall, its influence will be diluted. Therefore, the weight of light intensity is higher than that of wind speed, ωω 2 >ωω 3 。

[0133] In summary, that is, in ωω 1 +ωω 2 +ωω3 +ωω 4 Based on ωω = 1, let 0 < ωω 3 < ωω 2 < ωω 1 < ωω 4 < 1.

[0134] As an embodiment, the value range of ωω 1 is 0.35 - 0.4, the value range of ωω 2 is 0.25 - 0.3, the value range of ωω 3 is 0.1 - 0.2, the value range of ωω 4 is 0.45 - 0.5. The specific values are set by technicians according to the actual situation and are not limited here.

[0135] Based on the above embodiments, data processing and correlation analysis are performed on the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity of the water body to be detected at the current moment, and a chemical influence coefficient for evaluating the deterioration trend of the water body to be detected affected by chemical parameters is generated. The formula is as follows:

[0136] HXxs = β 1 (HY × SY) ZD + β 2 (NN × LN)

[0137] where HXxs is the chemical influence coefficient at the current moment, HY is the chemical oxygen demand, SY is the biochemical oxygen demand, NN is the nitrogen concentration, LN is the phosphorus concentration, ZD is the turbidity, and β 1 is the weight coefficient of the combination of chemical oxygen demand and biochemical oxygen demand, and β 2 is the weight coefficient of the combination of nitrogen concentration and phosphorus concentration.

[0138] It should be noted that the higher the chemical oxygen demand HY, the greater the chemical influence coefficient HXxs, and the faster the deterioration trend of the water body. The higher the biochemical oxygen demand SY, the greater the chemical influence coefficient HXxs, and the faster the deterioration trend of the water body. The higher the turbidity ZD, the greater the chemical influence coefficient HXxs, and the faster the deterioration trend of the water body. The higher the nitrogen concentration NN, the greater the chemical influence coefficient HXxs, and the faster the deterioration trend of the water body. The higher the phosphorus concentration LN, the greater the chemical influence coefficient HXxs, and the faster the deterioration trend of the water body. Therefore, the chemical influence coefficient HXxs is positively correlated with the chemical oxygen demand HY, biochemical oxygen demand SY, turbidity ZD, nitrogen concentration NN, and phosphorus concentration LN;

[0139] The chemical oxygen demand HY and the biochemical oxygen demand SY act together. The higher the chemical oxygen demand HY, the more serious the organic pollution in the water body; the higher the biochemical oxygen demand SY, the stronger the biodegradation ability. However, in the case of excessive organic matter content, it may still lead to water body deterioration. Therefore, the product of the two can comprehensively express the severity of water body pollution;

[0140] When the turbidity ZD increases, the impact of the organic matter (HY×SY) in the water body on the water body will intensify. By raising (HY×SY) to the power of the turbidity (HY×SY) ZD , this impact can be emphasized. For example, when the turbidity increases, it means there are more suspended solids in the water. At this time, the photosynthesis of aquatic plants is inhibited, and the decomposition rate of organic matter will slow down, resulting in more serious water body deterioration:

[0141] Nitrogen and phosphorus are essential nutrients for the growth of water body plants. Excessive nitrogen and phosphorus will lead to the outbreak of algae. When nitrogen and phosphorus coexist at high concentrations, different from the individual effects, their combined effect will accelerate the eutrophication process of the water body, resulting in further deterioration of the water body;

[0142] In summary, the calculation formula of the chemical influence coefficient HXxs in the above form is set.

[0143] The value range of the chemical influence coefficient HXxs is [0, 1];

[0144] When HJxs is close to 0, it means that the chemical parameters have little influence on the deterioration trend of the water body;

[0145] When HJxs is close to 1, it means that the chemical parameters have a great influence on the deterioration trend of the water body;

[0146] Therefore, the larger the chemical influence coefficient HXxs, the higher the influence of chemical conditions on the deterioration trend of the water body to be detected, and the faster the water body deterioration trend.

[0147] Since the increase of the chemical oxygen demand HY and the biochemical oxygen demand SY is directly related to the pollution degree of the water body, giving a larger weight to β1 can emphasize the impact of organic matter on the water body. For example, in the case of serious pollution, even if the biochemical oxygen demand SY is high, the self-purification ability of the water body will not be sufficient to offset the impact of organic matter;

[0148] Although the increase of nitrogen concentration NN and phosphorus concentration LN will cause water body problems, their influence usually occurs on the basis of the existence of organic matter. In the case of low organic matter pollution, the influence of nitrogen and phosphorus is relatively small. Giving β 2 a smaller weight can reflect the relatively minor relationship of the influence of nitrogen and phosphorus on the water body.

[0149] In summary, that is, in β 1 +β2 On the basis of = 1, let 0 < β 2 < β 1 < 1.

[0150] As an implementation manner, the value range of β 1 is 0.5 - 0.7, and the value range of β 2 is 0.3 - 0.4. The specific values are set by technicians according to the actual situation and are not limited here.

[0151] On the basis of the above embodiments, the environmental impact coefficient and chemical impact coefficient of the water body to be detected at the current moment are processed to generate the comprehensive scoring coefficient of the water body to be detected at the current moment. The formula is as follows:

[0152] PFxs = γ 1 HJxs + γ 2 HXxs

[0153] Among them, PFxs is the comprehensive scoring coefficient of the water body to be detected at the current moment. The comprehensive scoring coefficient is used to combine the environmental impact coefficient and chemical impact coefficient to comprehensively evaluate the water body to be detected. The larger the comprehensive scoring coefficient PFxs, the faster the deterioration trend of the water body to be detected;

[0154] It should be noted that the greater the environmental impact coefficient, the faster the water body deterioration trend, and the greater the chemical impact coefficient, the faster the water body deterioration trend. Therefore, the comprehensive scoring coefficient PFxs is positively correlated with both the environmental impact coefficient HJxs and the chemical impact coefficient HXxs. Therefore, the calculation formula of the comprehensive scoring coefficient PFxs in the form of weighted summation is set;

[0155] In the formula, γ 1 is the weight coefficient of the environmental impact coefficient, γ 2 is the weight coefficient of the chemical impact coefficient, and γ 1 , γ 2 The specific values are determined by the analytic hierarchy process. The specific logic is as follows:

[0156] Mark the two indicators of the environmental impact coefficient HJxs and the chemical impact coefficient HXxs, and determine the numerical values of the relative importance between them through the nine - scale method to construct a judgment matrix. Among them, mark the index of the environmental impact coefficient as 1 and the index of the chemical impact coefficient as 2. The constructed judgment matrix [q uv 3×3 is:

[0157]

[0158] ​Among them, both u and v represent the indices of coefficients, where u ∈ [1, 2] and v ∈ [1, 2], indicating the importance of the coefficient with index u relative to the coefficient with index v for the comprehensive scoring coefficient, q uv The specific value of q uv is determined by relevant experts using the 1 - 9 scoring method. q uv = 9 means that the coefficient with index u is extremely important for the comprehensive scoring coefficient compared to the coefficient with index v, q

[0159] Divide each element value in the judgment matrix by the sum of its column to obtain a normalized judgment matrix. Calculate the mean value of each row element value in the normalized judgment matrix, and take the mean value of the first - row element values as the proportionality coefficient of the environmental impact coefficient HJxs, and the mean value of the second - row element values as the proportionality coefficient of the chemical impact coefficient HXxs. With the constraint that the sum of the scaled - down values is equal to 1, scale the two proportionality coefficients proportionally, and take the scaled - down values as the weights of the corresponding coefficients.

[0160] Based on the above - mentioned embodiment, according to the comprehensive scoring coefficient, odor level, and water system coefficient of the water body to be detected at the current moment, evaluate the change situation of the odor level of the water body to be detected. The specific process is as follows:

[0161] When the comprehensive scoring coefficient is less than the water system coefficient, that is, PFxs < SZxs, it means that the water quality score of the water body to be detected is lower than the standard associated with the current odor level, indicating that the water body has improved and the odor characteristics have become better;

[0162] If the odor level of the water body to be detected at the current moment is above moderate, lower the odor level by 1 level;

[0163] If the odor level of the water body to be detected at the current moment is mild, keep the odor level at mild;

[0164] When the comprehensive scoring coefficient is equal to the water system coefficient, that is, PFxs = SZxs, it means that the water quality score of the water body to be detected is equal to the standard associated with the current odor level, indicating that the water body condition remains unchanged and the odor characteristics remain unchanged;

[0165] Keep the odor level of the water body to be detected at the current moment unchanged;

[0166] When the comprehensive scoring coefficient is greater than the water system coefficient, that is, PFxs > SZxs, it means that the water quality score of the water body to be detected is higher than the standard associated with the current odor level, indicating that the water body has deteriorated and the odor characteristics have become worse;

[0167] If the odor level of the water body to be detected at the current moment is below severe, raise the odor level by 1 level;

[0168] The odor level of the water body to be detected at the current moment is severe, and the odor level is maintained at severe.

[0169] Please refer to Figure 2 , the present invention also provides a technical solution:

[0170] A water body odor characteristic recognition system, which is used to execute any one of the above-mentioned water body odor characteristic recognition methods, including:

[0171] A data acquisition module, which is used to collect sample characteristic data of odoriferous water bodies in T historical time periods, and collect sample characteristic data of the water body to be detected at the current moment. The sample characteristic data includes the concentration of volatile organic compounds, the concentration of humic acid, the concentration of sulfides, the pH value of the water body, and the dissolved oxygen content, and collect environmental parameters and chemical parameters that affect the condition of the water body to be detected at the current moment. The environmental parameters include environmental temperature, water flow rate, rainfall, light intensity, and wind speed. The chemical parameters include chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity;

[0172] A matrix construction module, which is used to extract the odor characteristic matrix of the odoriferous water body from the sample characteristic data of the odoriferous water bodies in T historical time periods, including the average concentration of volatile organic compounds, the average concentration of humic acid, the average concentration of sulfides, the average pH value of the water body, and the average dissolved oxygen content;

[0173] An odor model construction module, which is used to use the odor characteristic matrix of the odoriferous water body as the input and the odor level as the label to train the model. The odor level is determined by an expert group, and the odor level model is trained. The odor level is 1 or 2 or 3, where 1 represents a slightly odoriferous water body, 2 represents a moderately odoriferous water body, and 3 represents a severely odoriferous water body;

[0174] An assignment module, which is used to assign different water body coefficients to slightly odoriferous water bodies, moderately odoriferous water bodies, and severely odoriferous water bodies respectively;

[0175] A water body to be detected recognition module, which is used to use the sample characteristic data of the water body to be detected at the current moment to construct an odor characteristic matrix of the water body to be detected, and input the odor characteristic matrix of the water body to be detected into the trained odor level model to obtain the odor level of the water body to be detected at the current moment;

[0176] A data processing and analysis module is used to perform data processing and correlation analysis on the environmental temperature, water flow rate, rainfall, light intensity, and wind speed that affect the water body to be detected at the current moment, generate an environmental impact coefficient for evaluating the deterioration trend of the water body to be detected affected by environmental parameters, perform data processing and correlation analysis on the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity that affect the water body to be detected at the current moment, generate a chemical impact coefficient for evaluating the deterioration trend of the water body to be detected affected by chemical parameters, and perform data processing on the environmental impact coefficient and chemical impact coefficient of the water body to be detected at the current moment to generate a comprehensive scoring coefficient of the water body to be detected at the current moment;

[0177] A water body evaluation module is used to evaluate the change situation of the odor level of the water body to be detected according to the comprehensive scoring coefficient, odor level, and water system coefficient of the water body to be detected at the current moment.

[0178] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0179] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by, electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0180] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0181] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. A method for identifying odor characteristics of water bodies, characterized in that: The specific steps include: S1. Collect sample characteristic data of odorous water bodies within T historical time periods, and collect sample characteristic data of the water body to be tested at the current moment, the sample characteristic data include volatile organic compound concentration, humic acid concentration, sulfide concentration, water pH value and dissolved oxygen content, and collect environmental parameters and chemical parameters that affect the condition of the water body to be tested at the current moment, the environmental parameters include ambient temperature, water flow rate, rainfall, light intensity and wind speed, and the chemical parameters include chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration and turbidity; S2. Extract the odor characteristic matrix of the odor water body from the sample characteristic data of the odor water body in T historical time periods, including the average volatile organic compound concentration, the average humic acid concentration, the average sulfide concentration, the average water pH value, and the average dissolved oxygen content; S3. The odor feature matrix of the odorous water body is used as input, and the odor level is used as a label training model. The odor level is determined by an expert group, and the odor level model is trained. The odor level is 1, 2, or 3, where 1 represents a water body with a light odor, 2 represents a water body with a moderate odor, and 3 represents a water body with a heavy odor; S4. Assign different water system indexes to water bodies with mild odor, water bodies with moderate odor, and water bodies with severe odor; S5. Using the sample characteristic data of the water body to be detected at the current moment, constructing the odor characteristic matrix of the water body to be detected, inputting the odor characteristic matrix of the water body to be detected into the odor grade model after training, and obtaining the odor grade of the water body to be detected at the current moment; S6. Perform data processing and correlation analysis on the ambient temperature, water flow rate, rainfall, light intensity and wind speed that affect the water body to be detected at the current moment, and generate an environmental impact coefficient for evaluating the deterioration trend of the environmental parameters affecting the water body to be detected; perform data processing and correlation analysis on the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration and turbidity that affect the water body to be detected at the current moment, and generate a chemical impact coefficient for evaluating the deterioration trend of the chemical parameters affecting the water body to be detected; perform data processing on the environmental impact coefficient and chemical impact coefficient of the water body to be detected at the current moment, and generate a comprehensive scoring coefficient for the water body to be detected at the current moment; S7. Evaluate the change in the odor level of the water body to be tested based on the comprehensive scoring coefficient, odor level and water system coefficient of the water body to be tested at the current moment.

2. The method for identifying the odor characteristics of water bodies according to claim 1, characterized in that: The average volatile organic compound concentration, average humic acid concentration, average sulfide concentration, average water pH value, and average dissolved oxygen content are obtained based on the following public announcements: Among them, μ hy , μ fz , μ lh , μ ss , μ ry They are the average volatile organic compound concentration, average humic acid concentration, average sulfide concentration, average water pH value, average dissolved oxygen content, hy i 、fz i , lh i 、ss i 、ry i are the concentration of volatile organic compounds, humic acid concentration, sulfide concentration, water pH value, and dissolved oxygen content in the i-th historical time period, respectively. i is the index of the historical time period, and i∈[1,T].

3. The method for identifying the odor characteristics of water bodies according to claim 2, characterized in that: Construct the odor feature matrix of the odorous water body and the water body to be detected. The specific process is as follows: Among them, X is the odor characteristic matrix of the odor water body; Using the sample feature data of the water body to be detected at the current moment, the odor feature matrix of the water body to be detected is constructed: Among them, d is the odor characteristic matrix of the water body to be detected, μ hy’ , μ fz’ , μ lh’ , μ ss’ , μ ry’ They are respectively the average volatile organic compound concentration, average humic acid concentration, average sulfide concentration, average water pH value, and average dissolved oxygen content of the water body to be tested.

4. The method for identifying the odor characteristics of water bodies according to claim 3, characterized in that: The olfactory grade model is constructed by a deep neural network based on a multilayer perceptron, wherein the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, wherein the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; The process of training the environmental impact model is as follows: According to the odor characteristic matrix of odorous water bodies in T historical time periods, the corresponding odor level is determined based on the expert scoring method. The higher the odor level, the higher the degree of water body deterioration. The odor feature matrix of odorous water bodies in T historical time periods is used as input, and the odor level is used as the output label for training. The mean square error is used as the loss function. When the mean square error is in the range of [0, 0.01], the training of the odor level model is completed.

5. The method for identifying the odor characteristics of water bodies according to claim 4, characterized in that: Different water system coefficients are assigned to water bodies with slight odor, water bodies with moderate odor, and water bodies with severe odor. The specific process is as follows: The odor level is 1, 2 or 3, 1 represents water with a slight odor, 2 represents water with a moderate odor, and 3 represents water with a heavy odor. The water system index of water bodies with slight odor is assigned to 0.3, the water system index of water bodies with moderate odor is assigned to 0.5, and the water system index of water bodies with severe odor is assigned to 0.

7.

6. The method for identifying the odor characteristics of water bodies according to claim 5, characterized in that: The temperature, water flow rate, rainfall, light intensity and wind speed of the water body to be tested at the current moment are processed and correlated to generate an environmental impact coefficient for evaluating the deterioration trend of environmental parameters on the water body to be tested. The formula is as follows: Among them, HJxs is the environmental impact coefficient at the current moment, WD is the ambient temperature, GZ is the light intensity, FS is the wind speed, SL is the water flow velocity, JY is the rainfall, ω1 is the weight coefficient of the ambient temperature, ω2 is the weight coefficient of the light intensity, ω3 is the weight coefficient of the wind speed, ω4 is the weight coefficient of the combination of water flow velocity and rainfall, and on the basis of ω1+ω2+ω3+ω4=1, let 0<ω3<ω2<ω1<ω4<1.

7. The method for identifying the odor characteristics of water bodies according to claim 6, characterized in that: The chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration and turbidity that affect the water body to be tested at the current moment are processed and correlated to generate a chemical influence coefficient for evaluating the deterioration trend of the chemical parameters affecting the water body to be tested. The formula is as follows: HXxs=β1(HY×SY) ZD +β2(NN×LN) Among them, HXxs is the chemical influence coefficient at the current moment, HY is the chemical oxygen demand, SY is the biochemical oxygen demand, NN is the nitrogen concentration, LN is the phosphorus concentration, ZD is the turbidity, β1 is the weight coefficient of the combination of chemical oxygen demand and biochemical oxygen demand, β2 is the weight coefficient of the combination of nitrogen concentration and phosphorus concentration, and on the basis of β1+β2=1, let 0<β2<β1<1.

8. The method for identifying the odor characteristics of water bodies according to claim 7, characterized in that: The environmental impact coefficient and chemical impact coefficient of the water body to be tested at the current moment are processed to generate a comprehensive scoring coefficient of the water body to be tested at the current moment, based on the following formula: PFxs=γ1HJxs+γ2HXxs Among them, PFxs is the comprehensive scoring coefficient of the water body to be tested at the current moment, γ1 is the weight coefficient of the environmental impact coefficient, γ2 is the weight coefficient of the chemical impact coefficient, and the specific values ​​of γ1 and γ2 are determined by the hierarchical analysis method.

9. The method for identifying the odor characteristics of water bodies according to claim 8, characterized in that: According to the comprehensive scoring coefficient, odor level and water system coefficient of the water body to be tested at the current moment, the change of the odor level of the water body to be tested is evaluated. The specific process is as follows: When the comprehensive scoring coefficient is less than the water system coefficient, i.e., PFxs < SZxs, it means that the water quality score of the water body to be detected is lower than the standard associated with the current odor level, indicating that the water body has improved and the odor characteristics have become better; When the odor level of the water body to be detected at the current moment is above moderate, reduce the odor level by 1 level; When the odor level of the water body to be detected at the current moment is mild, keep the odor level at mild; When the comprehensive scoring coefficient is equal to the water system coefficient, i.e., PFxs = SZxs, it means that the water quality score of the water body to be detected is equal to the standard associated with the current odor level, indicating that the water body condition remains unchanged and the odor characteristics remain unchanged; Keep the odor level of the water body to be detected at the current moment unchanged; When the comprehensive scoring coefficient is greater than the water system coefficient, i.e., PFxs > SZxs, it means that the water quality score of the water body to be detected is higher than the standard associated with the current odor level, indicating that the water body has deteriorated and the odor characteristics have become worse; When the odor level of the water body to be detected at the current moment is below severe, increase the odor level by 1 level; When the odor level of the water body to be detected at the current moment is severe, keep the odor level at severe.

10. A water body odor feature recognition system, the system being used to execute a water body odor feature recognition method according to any one of claims 1 to 9, characterized in that: Including: A data acquisition module, which is used to collect the sample characteristic data of the odor water body in T historical time periods, and collect the sample characteristic data of the water body to be detected at the current moment. The sample characteristic data includes the concentration of volatile organic compounds, the concentration of humic acid, the concentration of sulfide, the water body pH value, and the dissolved oxygen content, and collect the environmental parameters and chemical parameters that affect the water body condition to be detected at the current moment. The environmental parameters include environmental temperature, water flow rate, rainfall, light intensity, and wind speed, and the chemical parameters include chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration, and turbidity; A matrix construction module, which is used to extract the odor characteristic matrix of the odor water body from the sample characteristic data of the odor water body in T historical time periods, including the average concentration of volatile organic compounds, the average concentration of humic acid, the average concentration of sulfide, the average water body pH value, and the average dissolved oxygen content; An odor model construction module, which is used to use the odor characteristic matrix of the odor water body as the input and the odor level as the label to train the model. The odor level is determined by the expert group, and the odor level model is trained. The odor level is 1 or 2 or 3, where 1 represents a mildly odorous water body, 2 represents a moderately odorous water body, and 3 represents a severely odorous water body; An assignment module, which is used to assign different water system coefficients to mildly odorous water bodies, moderately odorous water bodies, and severely odorous water bodies respectively; A water body to be detected identification module, which is used to use the sample characteristic data of the water body to be detected at the current moment to construct the odor characteristic matrix of the water body to be detected, and input the odor characteristic matrix of the water body to be detected into the trained odor level model to obtain the odor level of the water body to be detected at the current moment; A data processing and analysis module, for performing data processing and correlation analysis on the ambient temperature, water flow rate, rainfall, light intensity and wind speed that affect the water body to be detected at the current moment, and generating an environmental impact coefficient for evaluating the deterioration trend of the environmental parameters affecting the water body to be detected, performing data processing and correlation analysis on the chemical oxygen demand, biochemical oxygen demand, nitrogen concentration, phosphorus concentration and turbidity that affect the water body to be detected at the current moment, and generating a chemical impact coefficient for evaluating the deterioration trend of the chemical parameters affecting the water body to be detected, performing data processing on the environmental impact coefficient and the chemical impact coefficient of the water body to be detected at the current moment, and generating a comprehensive scoring coefficient for the water body to be detected at the current moment; The water body assessment module is used to assess the change in the odor level of the water body to be detected based on the comprehensive scoring coefficient, odor level and water system coefficient of the water body to be detected at the current moment.

Images