Mercury removal control method and device for coal-fired power plants

Abstract

The present invention relates to the field of coal-fired power generation technology, and in particular to a method and device for controlling mercury removal in coal-fired power plants. The present invention controls a coal mill to grind and crush retired circuit boards and raw coal, thereby transporting the resulting mixed coal powder to the furnace of the coal-fired power plant. Feature extraction is performed on aligned optical remote sensing images of plants and synthetic aperture radar remote sensing images of equipment to obtain optical remote sensing features of the plants and synthetic aperture radar remote sensing features of the equipment. These features are then fused to obtain fused features. A prediction result is determined based on the fused features, and the amount and timing of adding retired circuit boards are determined based on the prediction result. Through this configuration, the present invention can more efficiently and simply control mercury removal in coal-fired power plants.

Description

Technical Field

[0001] The present invention relates to the technical field of coal-fired power generation, and in particular to a mercury removal control method and device for a coal-fired power plant. Background Art

[0002] The current mercury removal technology systems used in coal-fired power plants generally lack a feedback mechanism for mercury removal effectiveness. Specifically, whether it's physical coal washing and chemical treatment before combustion, adsorbent addition during combustion, or post-combustion synergistic mercury removal or specialized equipment, it's difficult to accurately monitor mercury removal results in real time and effectively feed these results back into operational processes to dynamically adjust mercury removal process parameters and optimize mercury removal results.

[0003] Based on this, the present invention proposes a mercury removal control method and device for a coal-fired power plant to solve the above technical problems. Summary of the Invention

[0004] The present invention describes a method and device for controlling mercury removal in a coal-fired power plant, which can more efficiently and simply control mercury removal in a coal-fired power plant.

[0005] According to a first aspect, the present invention provides a method for controlling mercury removal in a coal-fired power plant, the method comprising:

[0006] Controlling the coal mill to grind and crush the retired circuit boards and raw coal to deliver the resulting mixed coal powder to the furnace of the coal-fired power plant;

[0007] Acquire multiple plant optical remote sensing images and multiple equipment synthetic aperture radar remote sensing images of a coal-fired power plant within a preset time period;

[0008] Arranging the plurality of plant optical remote sensing images and the plurality of equipment synthetic aperture radar remote sensing images respectively in time series, to obtain arranged plant optical remote sensing images and arranged equipment synthetic aperture radar remote sensing images in sequence;

[0009] performing feature extraction on the arranged plant optical remote sensing image and the arranged equipment synthetic aperture radar remote sensing image, respectively, to obtain plant optical remote sensing features and equipment synthetic aperture radar remote sensing features; wherein the plant optical remote sensing features are used to characterize the effects of mercury and mercury oxides on plant growth, and the equipment synthetic aperture radar remote sensing features are used to characterize the effects of mercury and mercury oxides on coal-fired power plant equipment;

[0010] Fusing the plant optical remote sensing features and the device synthetic aperture radar remote sensing features according to a dynamic adjustment factor to obtain a fused feature; wherein the dynamic adjustment factor is determined based on the vegetation coverage rate in the area of ​​the coal-fired power plant and the weather influence factor within the preset time period;

[0011] Inputting the fused features into a preset mercury emission detection model to obtain a prediction result of mercury emission;

[0012] Based on the predicted result of the mercury emission, the amount and time of adding the retired circuit boards are determined.

[0013] According to a second aspect, the present invention provides a mercury removal control device for a coal-fired power plant, comprising:

[0014] a control unit configured to control a coal mill to grind and crush the retired circuit boards and raw coal, so as to transport the resulting mixed coal powder into a furnace of a coal-fired power plant;

[0015] an acquisition unit configured to acquire a plurality of plant optical remote sensing images and a plurality of equipment synthetic aperture radar remote sensing images of the coal-fired power plant within a preset time period;

[0016] The first processing unit is configured to arrange the plurality of plant optical remote sensing images and the plurality of equipment synthetic aperture radar remote sensing images respectively in a time series, and sequentially obtain arranged plant optical remote sensing images and arranged equipment synthetic aperture radar remote sensing images;

[0017] a second processing unit configured to perform feature extraction on the arranged plant optical remote sensing image and the arranged equipment synthetic aperture radar remote sensing image, respectively, to obtain plant optical remote sensing features and equipment synthetic aperture radar remote sensing features; wherein the plant optical remote sensing features are used to characterize the effects of mercury and mercury oxides on plant growth, and the equipment synthetic aperture radar remote sensing features are used to characterize the effects of mercury and mercury oxides on coal-fired power plant equipment;

[0018] a third processing unit configured to perform feature fusion on the plant optical remote sensing feature and the device synthetic aperture radar remote sensing feature according to a dynamic adjustment factor to obtain a fused feature; wherein the dynamic adjustment factor is determined based on the vegetation coverage rate in the area of ​​the coal-fired power plant and the weather influence factor within the preset time period;

[0019] a fourth processing unit configured to input the fused features into a preset mercury emission detection model to obtain a prediction result of mercury emission;

[0020] The fifth processing unit is configured to readjust the amount and time of adding the retired circuit boards based on the prediction result of the mercury emission amount.

[0021] In a third aspect, an embodiment of this specification further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in any embodiment of this specification is implemented.

[0022] In a fourth aspect, an embodiment of this specification further provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method described in any embodiment of this specification.

[0023] According to the present invention, the method and device for controlling mercury removal in coal-fired power plants first controls a coal mill to grind and crush decommissioned circuit boards and raw coal, as circuit boards contain bromide. The resulting mixed pulverized coal is then transported to the power plant's furnace to control mercury removal. Due to the reaction between bromide and mercury, the present invention can improve the efficiency of mercury removal in coal-fired power plants. Within a preset time interval, the present invention uses high-resolution optical remote sensing equipment to collect multiple optical remote sensing images of plants reflecting the vegetation conditions surrounding the coal-fired power plant. Simultaneously, synthetic aperture radar (SAR) technology is used to collect multiple SAR remote sensing images of coal-fired power plant equipment. After data collection is complete, the acquired optical remote sensing images of plants and equipment are sequenced based on the time dimension. Using a time series analysis algorithm, the plant optical remote sensing images are arranged in chronological order of acquisition time to construct a complete and continuous time series of plant optical remote sensing images. Similarly, the equipment SAR remote sensing images are subjected to the same time series sorting process to generate a time series of equipment SAR remote sensing images, ensuring the temporal correlation of the data and the consistency of the analysis. Next, deep feature mining was performed on the sorted plant optical remote sensing image time series and the equipment SAR remote sensing image time series. This process extracted plant optical remote sensing features and equipment SAR remote sensing features. The plant optical remote sensing features accurately reflect the processes and underlying mechanisms of mercury and mercury oxides' effects on plant growth. Mercury emissions can cause pollution stress in surrounding vegetation, altering its internal physiological and biochemical structure and ultimately leading to significant changes in its spectral characteristics. These changes are effectively encoded in the plant optical remote sensing features. The equipment SAR remote sensing features, on the other hand, are primarily used to characterize the mechanisms and effects of mercury and mercury oxides on coal-fired power plant equipment. Mercury and mercury oxides chemically react with the surface materials of coal-fired power plant equipment, causing corrosion and fouling. These changes in equipment status can be reflected in the equipment SAR remote sensing features through differences in the backscattering properties of SAR remote sensing images. Based on these characteristics, indirect detection of mercury emissions can be achieved through a comprehensive analysis of plant optical remote sensing features and equipment SAR remote sensing features. During actual data collection, plant optical remote sensing images are susceptible to interference from environmental factors such as cloud cover (cloud cover is positively correlated with weather factors) and vegetation coverage. Clouds block some light, weakening the light signal reflected by vegetation received by the sensor and reducing image quality. Therefore, image quality exhibits a significant negative correlation with weather factors. Higher vegetation coverage, on the other hand, provides richer vegetation information, helping to improve the quality of plant optical remote sensing images and the accuracy of feature extraction. Therefore, image quality is positively correlated with vegetation coverage.In contrast, SAR remote sensing images of equipment are negatively correlated with vegetation coverage. This is because vegetation scatters and attenuates SAR signals. Higher vegetation coverage interferes with the reflected signal from the equipment surface, reducing the quality of the SAR remote sensing image and the reliability of equipment feature extraction. To fully integrate the advantages of both remote sensing data types and improve the accuracy of mercury emission prediction, a dynamic adjustment factor is introduced to scientifically and rationally fuse the plant optical remote sensing features with the equipment SAR remote sensing features. The dynamic adjustment factor is determined by comprehensively considering real-time weather factors and vegetation coverage. An adaptive weighting algorithm dynamically adjusts the weights of the two features during the fusion process based on their reliability and effectiveness under different environmental conditions. After feature fusion, a fused feature containing rich information is obtained. This fused feature more comprehensively and accurately reflects the complex relationship between mercury emissions, the surrounding environment, and equipment status, significantly improving the accuracy of mercury emission predictions. Based on the predicted results, the amount and timing of decommissioned circuit boards to be added are determined. Through this configuration, the present invention enables more efficient and simple control of mercury removal in coal-fired power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0025] Figure 1 A schematic flow chart of a mercury removal control method for a coal-fired power plant according to one embodiment is shown;

[0026] Figure 2 A schematic block diagram of a mercury removal control device for a coal-fired power plant according to an embodiment is shown. DETAILED DESCRIPTION

[0027] The solution provided by the present invention is described below with reference to the accompanying drawings.

[0028] Figure 1 A flow chart of a method for controlling mercury removal in a coal-fired power plant according to an embodiment is shown. It is understood that the method can be executed by any device, equipment, platform, or equipment cluster with computing and processing capabilities. Figure 1 As shown, the method includes:

[0029] Step 100: Controlling a coal mill to grind and crush the retired circuit boards and raw coal, so as to transport the resulting mixed coal powder to a furnace of a coal-fired power plant;

[0030] Step 102: Acquire multiple optical remote sensing images of plants and multiple synthetic aperture radar remote sensing images of equipment in the coal-fired power plant within a preset time period;

[0031] Step 104: Arrange the plurality of plant optical remote sensing images and the plurality of equipment synthetic aperture radar remote sensing images in time series, to obtain arranged plant optical remote sensing images and arranged equipment synthetic aperture radar remote sensing images in sequence;

[0032] Step 106: Feature extraction is performed on the arranged plant optical remote sensing images and the arranged equipment synthetic aperture radar remote sensing images, respectively, to obtain plant optical remote sensing features and equipment synthetic aperture radar remote sensing features; wherein the plant optical remote sensing features are used to characterize the effects of mercury and mercury oxides on plant growth, and the equipment synthetic aperture radar remote sensing features are used to characterize the effects of mercury and mercury oxides on coal-fired power plant equipment;

[0033] Step 108: Fusing the plant optical remote sensing features and the equipment synthetic aperture radar remote sensing features according to a dynamic adjustment factor to obtain a fused feature; wherein the dynamic adjustment factor is determined based on the vegetation coverage rate in the area of ​​the coal-fired power plant and the weather influence factor within a preset time period;

[0034] Step 110: Input the fused features into a preset mercury emission detection model to obtain a prediction result of mercury emission.

[0035] In this embodiment, because circuit boards contain bromide, the present invention first controls a coal mill to grind and crush the decommissioned circuit boards and raw coal. The resulting mixed pulverized coal is then transported to the furnace of a coal-fired power plant to control mercury removal in the power plant. Due to the reaction between bromide and mercury, the present invention can improve the efficiency of mercury removal in coal-fired power plants. Within a preset time interval, the present invention uses high-resolution optical remote sensing equipment to collect multiple plant optical remote sensing images reflecting the vegetation conditions around the coal-fired power plant. Simultaneously, synthetic aperture radar (SAR) technology is used to collect multiple SAR remote sensing images of coal-fired power plant equipment. After data collection is completed, the acquired plant optical remote sensing images and equipment SAR remote sensing images are sequenced based on the time dimension. Using a time series analysis algorithm, the plant optical remote sensing images are arranged in chronological order of acquisition time to construct a complete and continuous plant optical remote sensing image time series. Similarly, the equipment SAR remote sensing images are subjected to the same time series sorting process to generate a time series of equipment SAR remote sensing images, ensuring the temporal correlation of the data and the consistency of the analysis. Next, deep feature mining was performed on the sorted plant optical remote sensing image time series and the equipment SAR remote sensing image time series. This process extracted plant optical remote sensing features and equipment SAR remote sensing features. The plant optical remote sensing features accurately reflect the processes and underlying mechanisms of mercury and mercury oxides' effects on plant growth. Mercury emissions can cause pollution stress in surrounding vegetation, altering its internal physiological and biochemical structure and ultimately leading to significant changes in its spectral characteristics. These changes are effectively encoded in the plant optical remote sensing features. The equipment SAR remote sensing features, on the other hand, are primarily used to characterize the mechanisms and effects of mercury and mercury oxides on coal-fired power plant equipment. Mercury and mercury oxides chemically react with the surface materials of coal-fired power plant equipment, causing corrosion and fouling. These changes in equipment status can be reflected in the equipment SAR remote sensing features through differences in the backscattering properties of SAR remote sensing images. Based on these characteristics, indirect detection of mercury emissions can be achieved through a comprehensive analysis of plant optical remote sensing features and equipment SAR remote sensing features. During actual data collection, plant optical remote sensing images are susceptible to interference from environmental factors such as cloud cover (cloud cover is positively correlated with weather factors) and vegetation coverage. Clouds block some light, weakening the light signal reflected by vegetation received by the sensor and reducing image quality. Therefore, image quality exhibits a significant negative correlation with weather factors. Higher vegetation coverage, on the other hand, provides richer vegetation information, helping to improve the quality of plant optical remote sensing images and the accuracy of feature extraction. Therefore, image quality is positively correlated with vegetation coverage.In contrast, SAR remote sensing images of equipment are negatively correlated with vegetation coverage. This is because vegetation scatters and attenuates SAR signals. Higher vegetation coverage interferes with the reflected signal from the equipment surface, reducing the quality of the SAR remote sensing image and the reliability of equipment feature extraction. To fully integrate the advantages of both remote sensing data types and improve the accuracy of mercury emission prediction, a dynamic adjustment factor is introduced to scientifically and rationally fuse the plant optical remote sensing features with the equipment SAR remote sensing features. The dynamic adjustment factor is determined by comprehensively considering real-time weather factors and vegetation coverage. An adaptive weighting algorithm dynamically adjusts the weights of the two features during the fusion process based on their reliability and effectiveness under different environmental conditions. After feature fusion, a fused feature containing rich information is obtained. This fused feature more comprehensively and accurately reflects the complex relationship between mercury emissions, the surrounding environment, and equipment status, significantly improving the accuracy of mercury emission predictions. Based on the predicted results, the amount and timing of decommissioned circuit boards to be added are determined. Through this configuration, the present invention enables more efficient and simple control of mercury removal in coal-fired power plants.

[0036] In one embodiment of the present invention, the dynamic adjustment factor is determined by the following formula:

[0037]

[0038] Wherein, α is the dynamic adjustment factor, x is the vegetation coverage rate, y is the weather influence factor, w1 is the first weight coefficient, w2 is the second weight coefficient, t1 is the starting time point of the preset time period, and t2 is the ending time point of the preset time period.

[0039] In one embodiment of the present invention, after inputting the fused features into a preset mercury emission detection model to obtain a mercury emission prediction result, the method further includes:

[0040] Get the total power generation and thermal power ratio of coal-fired power plants within a preset time period;

[0041] The total power generation, thermal power generation ratio, and mercury removal efficiency of coal-fired power plants are input into the dynamic prediction equation for mercury emissions to obtain the corrected mercury emissions.

[0042] The final results of mercury emissions are determined based on the corrected mercury emissions and the predicted mercury emissions.

[0043] In this embodiment, mercury emissions prediction using optical and synthetic aperture radar (SAR) remote sensing images is susceptible to interference from cloud cover and vegetation. Optical remote sensing images rely on the reflection of visible and near-infrared light by objects. Clouds can obscure objects, blocking light transmission, resulting in shadows or blurred areas in the image, severely degrading image quality. High vegetation cover can obscure some object information. Furthermore, the complex spectral characteristics of different vegetation types increase the difficulty of object identification and analysis. While SAR remote sensing images are capable of operating all day and all weather, vegetation scatters and attenuates microwave signals. High vegetation cover can interfere with radar signal detection of device surfaces, leading to biased device feature extraction. This can lead to errors in mercury emissions predicted solely based on optical and SAR remote sensing image features. To improve prediction accuracy, a corrected mercury emissions calculation is incorporated into the optimization of the prediction results. An adaptive weight allocation strategy based on information entropy and least squares is employed to fuse the corrected mercury emissions with the mercury emissions predicted based on remote sensing image features. Information entropy is used to measure the uncertainty of the information contained in different data sources (i.e., corrected emissions and remote sensing imagery-predicted emissions). The lower the uncertainty, the higher the weight in the fusion process. The least squares method further optimizes the weight distribution by constructing an objective function to minimize the sum of squared errors between the fusion result and the historical actual mercury emissions data. Based on this, the two are fused according to the determined weights to determine the final mercury emissions result. This multi-source data fusion and scientific weight distribution method can effectively integrate the advantages of different data, significantly improving the accuracy and reliability of the final mercury emissions forecast. The final mercury emissions forecast is then used to further improve the accuracy of the mercury removal efficiency calculation of coal-fired power plants.

[0044] In one embodiment of the present invention, the dynamic prediction equation of mercury emissions is constructed by the following formula:

[0045]

[0046] Where M is the corrected mercury emissions, E is the total electricity generation, and P S is the proportion of thermal power, P C is the reference thermal power ratio of the same type of coal-fired power plants under standard operating conditions, P X is the correction coefficient of thermal power proportion, a j is the benchmark value of mercury emissions per unit thermal power generation, w i is the weight of the i-th impact factor, a i is the correction value of the i-th factor affecting mercury emissions per unit thermal power generation, η s is the mercury removal efficiency of coal-fired power plants, η j Design mercury removal efficiency for coal-fired power plants, η tIt is the comprehensive adjustment coefficient of mercury removal efficiency.

[0047] In this embodiment, considering that the thermal power ratio may deviate from the reference value due to factors such as seasons and fluctuations in electricity demand in the actual operation of coal-fired power plants, the thermal power ratio correction coefficient can be set between 0.8 and 1.2. i is the weight of the i-th influencing factor, which is determined by experimental data or professional research and reflects the relative magnitude of the impact of each influencing factor on mercury emissions. i is the correction value of the i-th factor that affects mercury emissions per unit of thermal power generation. For example, factors such as sulfur content and ash content in coal quality will affect mercury release. Each factor corresponds to a correction value. t = is the comprehensive adjustment factor for mercury removal efficiency, taking into account factors such as equipment aging and unstable operating conditions. Its value range is 0.9-1.1 and is determined by evaluating equipment operating conditions and fitting historical mercury removal efficiency data. The above formula accurately determines corrected mercury emissions.

[0048] In one embodiment of the present invention, after determining the final result of mercury emissions based on the corrected mercury emissions and the predicted mercury emissions, the method further includes:

[0049] Determining a mean square error loss of a preset mercury emission detection model based on the final mercury emission result and the mercury emission prediction result;

[0050] Back propagation is performed according to the mean square error loss to update the parameters of the preset mercury emission detection model, and the mercury emission detection model after the parameter update is completed is used as the detection model for the next time period.

[0051] In this example, after obtaining the mean squared error loss, a backpropagation algorithm is used to propagate the loss value back along the model's computational path, calculating the contribution of each parameter to the loss. Using optimization algorithms such as gradient descent, the parameters of the pre-set mercury emission detection model are updated based on the calculated gradient information, enabling the model to more accurately approximate the true value in subsequent predictions. The updated mercury emission detection model will be used for mercury emission detection in the next time period, continuously improving the accuracy and stability of the model's predictions.

[0052] In one embodiment of the present invention, the mean square error loss is determined by the following formula:

[0053]

[0054] Where L is the mean square error loss, n is the number of samples, G is the predicted result of mercury emissions, and M is the corrected mercury emissions.

[0055] In one embodiment of the present invention, the preset mercury emission detection model is a long short-term memory network model.

[0056] In this embodiment, the Long Short-Term Memory (LSTM) model is well-suited as a mercury emission detection model. Its ability to process time series data demonstrates that mercury emission data exhibits distinct time series characteristics, with mercury emissions at different time points interrelated. The LSTM model possesses a unique gating mechanism, including input, forget, and output gates, which effectively handles long-term dependencies, memorizes information from the past, and accurately captures and analyzes temporal trends in mercury emissions. For example, by learning from historical emission data, it can predict future fluctuations in mercury emissions. Furthermore, its ability to model complex nonlinear relationships demonstrates that mercury emissions have complex nonlinear relationships with plant optical remote sensing characteristics and equipment SAR remote sensing characteristics. The LSTM model is a powerful nonlinear model. Through its complex internal structure and parameter adjustments, it can effectively fit these complex relationships, uncovering patterns hidden in the data and enabling accurate prediction of mercury emissions. For example, it can effectively model the complex relationships between changes in vegetation spectral characteristics and the degree of pollution caused by mercury emissions, and between equipment corrosion and scaling and mercury emissions. Interference Resistance: In actual data collection, both optical remote sensing images of plants and SAR images from equipment are subject to interference from environmental factors such as cloud cover and vegetation cover, resulting in noisy feature data. During training, the LSTM model uses its inherent learning mechanism to filter and process this noisy data to a certain extent, preventing significant deviations in overall prediction results due to localized noise interference. This ensures the stability and reliability of mercury emission detection. Continuous Learning and Dynamic Updates: Over time, the operating conditions and surrounding environment of coal-fired power plants may change, and the factors affecting mercury emissions will also change accordingly. The LSTM model continuously learns based on newly acquired data, constantly updating model parameters through a backpropagation algorithm, enabling it to adapt to dynamically changing environments and continuously and accurately detect mercury emissions. For example, if a coal-fired power plant replaces new mercury removal equipment or the surrounding vegetation type changes, the model can quickly learn from this new information and adjust its prediction strategy.

[0057] The foregoing description describes specific embodiments of the present invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0058] According to another embodiment, the present invention provides a mercury removal control device for a coal-fired power plant. Figure 2 A schematic block diagram of a mercury removal control device for a coal-fired power plant according to an embodiment is shown. It is understood that the device can be implemented by any device, equipment, platform, or device cluster with computing and processing capabilities. Figure 2 As shown, the device includes: a control unit 200, an acquisition unit 202, a first processing unit 204, a second processing unit 206, a third processing unit 208, a fourth processing unit 210 and a fifth processing unit 212. The main functions of each component unit are as follows:

[0059] a control unit configured to control a coal mill to grind and crush the retired circuit boards and raw coal, so as to transport the resulting mixed coal powder into a furnace of a coal-fired power plant;

[0060] an acquisition unit configured to acquire a plurality of plant optical remote sensing images and a plurality of equipment synthetic aperture radar remote sensing images of the coal-fired power plant within a preset time period;

[0061] The first processing unit is configured to arrange the plurality of plant optical remote sensing images and the plurality of equipment synthetic aperture radar remote sensing images respectively in a time series, and sequentially obtain arranged plant optical remote sensing images and arranged equipment synthetic aperture radar remote sensing images;

[0062] a second processing unit configured to perform feature extraction on the arranged plant optical remote sensing image and the arranged equipment synthetic aperture radar remote sensing image, respectively, to obtain plant optical remote sensing features and equipment synthetic aperture radar remote sensing features; wherein the plant optical remote sensing features are used to characterize the effects of mercury and mercury oxides on plant growth, and the equipment synthetic aperture radar remote sensing features are used to characterize the effects of mercury and mercury oxides on coal-fired power plant equipment;

[0063] a third processing unit configured to perform feature fusion on the plant optical remote sensing feature and the device synthetic aperture radar remote sensing feature according to a dynamic adjustment factor to obtain a fused feature; wherein the dynamic adjustment factor is determined based on the vegetation coverage rate in the area of ​​the coal-fired power plant and the weather influence factor within the preset time period;

[0064] a fourth processing unit configured to input the fused features into a preset mercury emission detection model to obtain a prediction result of mercury emission;

[0065] The fifth processing unit is configured to readjust the amount and time of adding the retired circuit boards based on the prediction result of the mercury emission amount.

[0066] In one embodiment of the present invention, the dynamic adjustment factor is determined by the following formula:

[0067]

[0068] In the formula, α is the dynamic adjustment factor, x is the vegetation coverage rate, y is the weather influence factor, w1 is the first weight coefficient, w2 is the second weight coefficient, t1 is the starting time point of the preset time period, and t2 is the ending time point of the preset time period.

[0069] In one embodiment of the present invention, after determining the mercury removal efficiency of the coal-fired power plant according to the predicted result of the mercury emission, the method further includes:

[0070] Obtain the total power generation and thermal power proportion of the coal-fired power plant within the preset time period;

[0071] Inputting the total power generation, the thermal power ratio, and the mercury removal efficiency of the coal-fired power plant into a dynamic prediction equation for mercury emissions to obtain corrected mercury emissions;

[0072] A final result of mercury emissions is determined based on the corrected mercury emissions and the predicted result of mercury emissions.

[0073] In one embodiment of the present invention, the dynamic prediction equation of mercury emissions is constructed by the following formula:

[0074]

[0075] Where M is the corrected mercury emission, E is the total power generation, and P S is the proportion of thermal power, P C is the reference thermal power ratio of the same type of coal-fired power plants under standard operating conditions, P X is the correction coefficient of thermal power proportion, a j is the benchmark value of mercury emissions per unit thermal power generation, w i is the weight of the i-th impact factor, a i is the correction value of the i-th factor affecting mercury emissions per unit thermal power generation, η s is the mercury removal efficiency of the coal-fired power plant, η j Design mercury removal efficiency for coal-fired power plants, η t It is the comprehensive adjustment coefficient of mercury removal efficiency.

[0076] In one embodiment of the present invention, after determining the final result of mercury emissions based on the corrected mercury emissions and the predicted mercury emissions, the method further includes:

[0077] Determining a mean square error loss of a preset mercury emission detection model based on the final result of the mercury emission and the predicted result of the mercury emission;

[0078] Back propagation is performed according to the mean square error loss to update the parameters of the preset mercury emission detection model, and the mercury emission detection model after the parameter update is completed is used as the detection model for the next time period.

[0079] In one embodiment of the present invention, the mean square error loss is determined by the following formula:

[0080]

[0081] Wherein, L is the mean square error loss, n is the number of samples, G is the predicted result of the mercury emission, and M is the corrected mercury emission.

[0082] In one embodiment of the present invention, the preset mercury emission detection model is a long short-term memory network model.

[0083] According to another embodiment, there is also provided a computer readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute a combination of Figure 1 The method described.

[0084] According to another embodiment, an electronic device is provided, comprising a memory and a processor, wherein the memory stores an executable code, and when the processor executes the executable code, the system realizes the combination of Figure 1 The method described.

[0085] The various embodiments of the present invention are described in a progressive manner. Similar portions between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the device embodiments are described briefly because they are generally similar to the method embodiments. For relevant portions, refer to the description of the method embodiments.

[0086] Those skilled in the art will appreciate that, in one or more of the above examples, the functions described herein may be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium.

[0087] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solution of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for controlling mercury removal in a coal-fired power plant, characterized in that: The method comprises: Controlling the coal mill to grind and crush the retired circuit boards and raw coal to deliver the resulting mixed coal powder to the furnace of the coal-fired power plant; Acquire multiple plant optical remote sensing images and multiple equipment synthetic aperture radar remote sensing images of a coal-fired power plant within a preset time period; Arranging the plurality of plant optical remote sensing images and the plurality of equipment synthetic aperture radar remote sensing images respectively in time series, to obtain arranged plant optical remote sensing images and arranged equipment synthetic aperture radar remote sensing images in sequence; performing feature extraction on the arranged plant optical remote sensing image and the arranged equipment synthetic aperture radar remote sensing image, respectively, to obtain plant optical remote sensing features and equipment synthetic aperture radar remote sensing features; wherein the plant optical remote sensing features are used to characterize the effects of mercury and mercury oxides on plant growth, and the equipment synthetic aperture radar remote sensing features are used to characterize the effects of mercury and mercury oxides on coal-fired power plant equipment; Fusing the plant optical remote sensing features and the device synthetic aperture radar remote sensing features according to a dynamic adjustment factor to obtain a fused feature; wherein the dynamic adjustment factor is determined based on the vegetation coverage rate in the area of ​​the coal-fired power plant and the weather influence factor within the preset time period; Inputting the fused features into a preset mercury emission detection model to obtain a prediction result of mercury emission; Based on the predicted result of the mercury emission, the amount and time of adding the retired circuit boards are determined.

2. The method according to claim 1, characterized in that The dynamic adjustment factor is determined by the following formula: In the formula, α is the dynamic adjustment factor, x is the vegetation coverage rate, y is the weather influence factor, w1 is the first weight coefficient, w2 is the second weight coefficient, t1 is the starting time point of the preset time period, and t2 is the ending time point of the preset time period.

3. The method according to claim 2, characterized in that After determining the mercury removal efficiency of the coal-fired power plant according to the mercury emission prediction result, the method further includes: Obtain the total power generation and thermal power proportion of the coal-fired power plant within the preset time period; Inputting the total power generation, the thermal power ratio, and the mercury removal efficiency of the coal-fired power plant into a dynamic prediction equation for mercury emissions to obtain corrected mercury emissions; A final result of mercury emissions is determined based on the corrected mercury emissions and the predicted result of mercury emissions.

4. The method according to claim 3, characterized in that The dynamic prediction equation of mercury emissions is constructed by the following formula: Where M is the corrected mercury emission, E is the total power generation, and P S is the proportion of thermal power, P C is the reference thermal power ratio of the same type of coal-fired power plants under standard operating conditions, P X is the correction coefficient of thermal power proportion, a j is the benchmark value of mercury emissions per unit thermal power generation, w i is the weight of the i-th impact factor, a i is the correction value of the i-th factor affecting mercury emissions per unit thermal power generation, η s is the mercury removal efficiency of the coal-fired power plant, η j Design mercury removal efficiency for coal-fired power plants, η t It is the comprehensive adjustment coefficient of mercury removal efficiency.

5. The method according to claim 4, characterized in that After determining the final result of mercury emissions based on the corrected mercury emissions and the predicted mercury emissions, the method further includes: Determining a mean square error loss of a preset mercury emission detection model based on the final result of the mercury emission and the predicted result of the mercury emission; Back propagation is performed according to the mean square error loss to update the parameters of the preset mercury emission detection model, and the mercury emission detection model after the parameter update is completed is used as the detection model for the next time period.

6. The method according to claim 5, characterized in that The mean square error loss is determined by the following formula: Wherein, L is the mean square error loss, n is the number of samples, G is the predicted result of the mercury emission, and M is the corrected mercury emission.

7. The method according to claim 6, characterized in that The preset mercury emission detection model is a long short-term memory network model.

8. A mercury removal control device for a coal-fired power plant, characterized in that: include: a control unit configured to control a coal mill to grind and crush the retired circuit boards and raw coal, so as to transport the resulting mixed coal powder into a furnace of a coal-fired power plant; an acquisition unit configured to acquire a plurality of plant optical remote sensing images and a plurality of equipment synthetic aperture radar remote sensing images of the coal-fired power plant within a preset time period; The first processing unit is configured to arrange the plurality of plant optical remote sensing images and the plurality of equipment synthetic aperture radar remote sensing images respectively in a time series, and sequentially obtain arranged plant optical remote sensing images and arranged equipment synthetic aperture radar remote sensing images; a second processing unit configured to perform feature extraction on the arranged plant optical remote sensing image and the arranged equipment synthetic aperture radar remote sensing image, respectively, to obtain plant optical remote sensing features and equipment synthetic aperture radar remote sensing features; wherein the plant optical remote sensing features are used to characterize the effects of mercury and mercury oxides on plant growth, and the equipment synthetic aperture radar remote sensing features are used to characterize the effects of mercury and mercury oxides on coal-fired power plant equipment; a third processing unit configured to perform feature fusion on the plant optical remote sensing feature and the device synthetic aperture radar remote sensing feature according to a dynamic adjustment factor to obtain a fused feature; wherein the dynamic adjustment factor is determined based on the vegetation coverage rate in the area of ​​the coal-fired power plant and the weather influence factor within the preset time period; a fourth processing unit configured to input the fused features into a preset mercury emission detection model to obtain a prediction result of mercury emission; The fifth processing unit is configured to readjust the amount and time of adding the retired circuit boards based on the prediction result of the mercury emission amount.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, and when the computer program is executed in a computer, the computer is caused to execute the method according to any one of claims 1 to 7.

Images